Diff pinion seal replacement

Good morning,

I’m helping a friend who bought a 340.

Oil is dripping out the diff nose.

I’m just wondering if there are any tricky steps.

Does the prop shaft have to come out, or is there room to move it aside?

I plan to mark the nut, and count turns to remove. I guess I’s need to put a bar through an axle to prevent the pinion shaft from turning.

Pry out seal, install new, count turns on nut back on…done, I hope.

Am I missing anything?



Does your year of car have a crush sleeve or shims for pinion preload adjustment? Crush sleeve diffs take special care not to alter the preload when changing the pinion seal. Others will have more detail and info.

My mk1 /2 workshop manual shows no reference to the crush sleeve which I think was used on later models.
So, easy job. Mark the nut , unscrew it, pull off the flange, pull out the seal, push in a new one and retighten the nut to the same place.

1 Like

You can still count turns. Too much = full disassembly time.

Why this hassle with counting turns ? Isn’t this nut tightened with torque required ?
Shims won’t get any slimmer to increase preload … I bet.
And the nut must be tight as there’s no split pin to keep it in place (not a castle nut).

Yes indeed, and in the manual the crush sleeve version is supposed to be done by torque not turns! The end result should be the same, right?
And the nut will be slightly oval for keeping it in place, tight enough not to move in any situation. Probably. Otherwise the torque will do…

Rob asked if the driveshaft is in the way. Is it?

No prop is not in the way , it can move on splines in and out !

They can be changed in situ with relative ease

some tips;

Ideally take a torque measurement, preferably a rotational one, but if you do not have that sort of torque dial gauge, at least the “breakaway” torque with an ordinary Torque wrench

It should read the same after as before. Not strictly needed on this job

They have shims. Just mark nut/flange with prick punch and nail polish, you will know when its tight, so you do not need to count turns

To break the torque on the pinion nut can be difficult unless you have heavy duty impact gear. If not, place a trolley jack under a breaker bar, and use that to undo the nut

Next, you may need a heavy duty puller to get the flange off…or it may just slide off

As to the seal, its a tough bugger, and I always attack them the same way, drill one or two tiny holes in the steel outer case, tape your drill so you dont go too deep.

Insert self tapper screw in tiny holes. Best bet is a slide hammer, but you can lever

Slide hammer will pop it right out

I made a simple jig to hold the flange in place while you undo the nut, jamming something through is inadvisable, the force is to high and it will bend.

My jig is a length of roughly 2" x 1/4" steel with an arc carved out with a grinder and two holes drilled to fit the flange bolts

You will see what I mean when you do the job

Correct. Had to use a 3/4" impact tool on my XJ6. Smaller 1/2" wouldn’t look at it. P

On my XJ the half inch impact did well but getting the flange off took half a day. This was a manufacturing issue for a short time back then apparently, the pinion normally just slides off. If the nut is too tight brake one wheel and turn the other for a 3½:1 torque mulitipication with the wrench jammed against the ground.

1 Like

I am pretty sure mine must have had original factory Loctite (it was an OEM leather seal)

Suspecting that, I would use heat if they are stuck.
Mine needed a very heavy duty puller and 3/4" drive gear to get it off

doesnt need all that much heat to soften it

Impressive. I just acquired an Ingersol-Rand 3/4" air impact for $50

It has a larger air fitting and I am not sure my compressor will even spin it hard enough

I need a larger compressor…which I have, but it will only start if I spin it by hand, so I suspect the Start Capacitor is failed

sorry OP, I should start a new thread on that :smiley:

Its a good idea to use sealant on the diff seal outer and splines

99% sure I used a Dana 44 pinion seal, half price same thing, but dont trust me, @Mike_S may be able to confirm they fit ?

The splines were a tight interference fit, the puller failed me so I took the flange off a transmission and with nuts and shims in between it worked like a great puller, took forever to gain a few mm each time, unbolt, add shim, etc.
This won’t be rob‘s problem though I‘m sure.

I have a linishing belt withe the same problem. Have to give it a spin! Remanent flux is not enough to get it moving.

Interested in this outcome. As per leslioq, I counted threads exposed, popped the nut off with an impact gun while I had it on a two post lift, regular small flywheel puller got the flange off easy. Swapped the seal out and zipped the nut back up. Shim sets pinion depth, crush collar sets bearing preload, I figure if its close it should be ok. Drove away with ring and pinion whine. That was years ago, just now getting to pulling the rear end for a full refurb and ratio change. Don’t forget to clean and seal the pinion flange splines with anaerobic sealer or it’ll pour gear oil everywhere.

Thank you all for taking time to coach me on this job.

I see there are some “gotchas” when doing this job.

I’ll make sure I have the sealant, puller, punch etc ready before starting.

This forum is great, as always.

I’ll post results when the job’s done.



Hello David,
Are you referring to the torque required to rotate the Pinion Shaft with no Crown-wheel assembly in place, or the torque for the Companion Flange nut, for differentials using the Crush Sleeve?

In the S3 E Type Manual and XJ6 manual, it states quite emphatically that the Companion Flange nut is to be tightened to 120 to 140 lb/ft torque and if inadvertently tightened beyond 140 lb/ft torque, the current Crush Spacer must be replaced with a new on and the operation of tightening the Companion Flange nut repeated.

The big issue with the above instruction is that it takes well over 200 lb/ft torque to get the Crush Spacer to start to collapse and in my experience, doesn’t drop below 140 lb/ft torque for the remainder of the journey until the Taper Roller bearing cones make intimate contact with their respective cups.

What it should have said, is that the torque figure is disregarded completely, and when intimate contact between bearing Cups and Cones is made, ie. zero End Float of the Pinion Shaft, a Torque Wrench, or other means of measuring the torque required to rotate the shaft is used. The nut is further tightened until a torque of circa 26 lb/in torque is required to rotate the Pinion Shaft

When replacing the seal of differentials that used the Crush Spacer, the position of the Companion Flange nut should be marked before removing so that it can be pulled up to the same mark. Without this mark, a torque wrench, or other means of measuring torque, can be used to measure the amount of torque requited to rotate the Pinion Shaft through the Backlash between the Pinion Shaft and the Crown-wheel.

With a differential that uses Shims to set the preload of the Pinion Shaft, as will be the case with Robert’s friend’s car, there is no need to mark the position of the Companion Flange nut, just torque it up to circa 130 lb/ft torque.



Thank you Bill, good that you ask, I was referring to the torque reqired to turn the pinion through the backlash. I should have said ‚even the crush sleeve version‘, with the washers it of course doesn’t matter, torque it up and done.