MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG Midget and Sprite Technical - Diff pinion seal - decisions, decisions.

I have a, new to me, 3.9 diff ready to go into the Frogeye axle (off the car). I've no idea if the diff was leaking oil before I bought it. I've read the archives extensively about pinion seal renewal and can't decide whether to attempt it or leave it. Undoing the nut only seems to be half the battle. But how then to deal with the required pre-load? Do I just change the seal and tighten the nut to the pre-load torque or do it properly (it seems) and use a new crush washer? Using the existing crush washer feels wrong as you can't surely get that original tightness back. I don't want to make matters worse.
Bill Bretherton

I would leave it until you know if it leaks.

Whenever I have renewed a diff seal I have carefully marked the nut and shaft end and then re-tightened the nut to the same marks.
Chris at Octarine Services

You either carefully mark nut and shaft and reuse the crush spacer tightening back up to the marks or use a new crush spacer and tighten to the correct bearing preload.


When i did mine I used a new crush washer and concocted a bag of nuts and bolts of a known weight, string and a ruler to measure the pre-load. Once you get close it is very easy to over do it. A fraction of a turn on the nut changes the pre-load by a big amount so you have to go carefully and check often. Go too far and you need a new crush spacer.

Googling the correct pre-load torque figure, and if that is with the half shafts connected, taken out, will yield lots of different answers. In the end I took the average, and what 'felt right'.

Chris Madge

Just discovered I don't have a large enough suitable socket (I assume it's UNF) and maybe a 6 point one would be better for the needed torque so I might try it as is.
Bill Bretherton

I just mark the pinion and nut, as the Chrises have said.
Dave O'Neill 2

Surely, once the crush washer has deformed it stays like that so I'm wondering if sufficient torque will be imposed next time you tighten the nut.
Bill Bretherton

When the crush washer is permanently deformed it changes in length but still will exhibit elastic behaviour the same as before deformation but will return to a shorter length when unloaded than the original. Here is a pretty good graph of the concept https://www.nuclear-power.net/wp-content/uploads/2019/11/Stress-strain-curve-permanent-deformation.png .
David Billington

Thanks David. So there is still some elasticity in thd crush washer. Theoretically then you tighten the nut to the original point, as Chris, Chris and Dave have said, although is it not best to go carefully and keep checking the pre-load anyway? I also read from the archive that a given diff pinion pre-load may have "relaxed" in usage so it may, indeed, be best to tighten a bit more than originally i.e. to achieve the desired pre-load. Does that make sense?
Bill Bretherton

Bill,

I would say to check the pre-load anyway as wear in the bearings may have reduced it. When I helped my neighbour replace the CWP in his Morgan final drive we used a simple arm and Salter spring scale to measure the pre-load. In that case we had a comprehensive axle rebuild manual as the same Salisbury axle was used in Winget dump trucks and readily available online and gave pinion torques for just the pinion, pinion and crownwheel/diff fitted, and with the halfshafts as well.
David Billington

Bill,
I had a leaking pinion oil seal on the 3.9 diff I fitted to my frogeye some years ago. I marked the nut and shaft and re-tightened carefully to the marks.
That was several years ago and no problems.

Shouldn't have said that, I've probably jinxed it now, bu**er!!

Dennis
D Stanfield

I'm a bit naughty doing seals
Mark the shaft and nut as the troops have said, undo it remove the flange replace the seal reassemble and I use a drip of loctite on the thread , do it back up to the marks and then a slight, tiddle further, probably 20thou past when the marks line up, just to get it nice and tight.
Sometimes I have found if you're using the original nut and collapsable sleeve and do it just up to the original mark, the nut can come loose through lack of preload
The golden rule with collapsable spacers new, when assembling a diff is not to go backwards at all ,just creep it up to spec. so that's the reason I give them just that tiny little extra when using an original sleeve,
willy
William Revit

Bill if you need to borrow any tools give me a shout. I'm in Thornbury South Gloucestershire so not far away.

I have a used crush washer which might may have more 'give' left in it that are welcome to have.
Chris Madge

Willy, that all makes a lot of sense, thanks.
Chris, thanks, that's very kind of you.
Dennis, I'm sure it'll be ok as you've done what many seem to do. I suppose you could check the pre-load on the car although there'd be drag from the half shafts and wheel bearings presumably (with brake drums off so maybe do it when checking/ changing brake shoes).
Bill Bretherton

I don't think I'd use the sleeve out of another diff, the distance between the bearings could be different
I'd be sticking to the set you've got and just give it that tiny little extra
And for my money I'd go ahead and replace the seal
You'd be peeed off if you go to the trouble of fitting it up and it leaks
William Revit

My recent experience with rebuilding 2 diffs is that the drive flange shaft is usually worn out so a new seal won't work. The diff in the car has now started leaking, so I have ordered a new seal and will put a Speedi Sleeve on the flange, as I did with the other 2 units. The one to use is 99150, but prices vary widely. Best I can find is item 301388131740 on eBay. At £9.78 it's worth having it to hand before getting the flange off.

Les
L B Rose

Well the oil seal has been changed (thanks to Chris M. for assistance) but I made an error when measuring pre-load and it is a bit higher than recommended after having gone up to 140 ft lbs torque on the nut. I'm getting about 9 or 10 inch lbs pre-load using a spring balance (checked for accuracy) whilst rotating the pinion. Is this too much over the 4 to 6 figure recommended. It's definitely tighter than it was but revolves smoothly. I'm guessing the bearings have been more compressed slightly. Is this bad and should I go in there again and replace thecrush washer (it's currently original)?
Bill Bretherton

Is this with the CW and diff cage in or not.
David Billington

In David.
Bill Bretherton

Bill,

Do you have the various torques for turning just the pinion, pinion + CW and diff cage etc. I tried to find this information for the BMC A series final drive and couldn't find it, I'm sure it exists somewhere but couldn't find it online. I have seen it quoted for other final drive assemblies but they differ somewhat from what you mention.
David Billington

I don't have pre-load figures other than what has been quoted in an archive thread on this forum, which is 4 to 6 inch lbs minimum for an assembled diff (not new). But I don't know what the permissible MAXIMUM is. The risk, of course, is overheating the pinion shaft bearings.
Bill Bretherton

Bill
It'll be fine, the spec is 11-13 in/lb for new bearings---you need a minimum of 5in/lb for used bearings, absolute minimum, so that gives the range of 5---13. If you've got 9-10 you're in the middle there and allowing a bit for having the c/w carrier in there and a new seal as well would have to bring you back a couple to 7 or 8
If it turns smoothly as you have stated, I'm sure it will be fine
I've never used a tension wrench on pinion nuts, just go by feel
So how far out were the marks on the pinion/nut after you pulled it to 140---
William Revit

Willy, thanks for that. Embarrassingly I neglected to mark it (!). I was going to rely on the pre-load measurement, which I was doing incorrectly. I hadn't apparently reached a minimum pre-load so kept going to 140 ft lbs. And that is TIGHT.

The problem is, I've read the archives extensively especially Norm Kerr's contribution and it does seem that the crush washer length is the important thing, as that sets the bearing position. Surely that can only be achieved with a new crush washer? I must have pushed the bearings closer and it did feel looser before I started. It does seem to have slightly reduced the backlash too. But I speak from zero experience with diffs other than removing/ fitting them as is.
Bill Bretherton

I just changed the oilseal on my diff and noticed that the workshop manual says nothing about the crush sleeve when doing this. Just 'reassemble'. Here's the page. I found that the torque on the nut was very low to get near the original preload. I marked the nut and flange and found I had to tighten beyond the marks to get to that preload. Even so the torque is only about 30 lb/ft. I have used Loctite and am hoping for the best.

Les

L B Rose

Am I right in thinking that there were two types of differential - one with a collapsible spacer and the other with a solid spacer and shims?
Dave O'Neill 2

Dave as I was going to say the earlier ones were shimmed then later crush sleeves were used.
David Billington

Les, assuming original 4.22 diff it may be that your Frogeye has the fixed sleeve plus shims. Although in that case, you'd think the original torque would be needed to achieve the pre-load.
Bill Bretherton

My manual only describes the crush sleeve type. It's for Sprites Mark 2-4.
L B Rose

Bill, If you've pulled it up to 140 and have that 9-10 in/lbs drag --your work is done

Crush washers were introduced to save all the hassle of having to select shims for the correct preload--
With the solid spacer type it can be a pain, assemble, measure, dismantle, reshim, reassemble, measure, dismantle and on and on and this is usually done without the seal in there to save having to remove it all the time--then when finished it's apart again to fit the seal and reassemble--

With the crush sleeve it's easier, specially on the production line, the strength of the sleeve determines the tension needed for the pinion nut and these do vary from aftermarket suppliers, which is the reason I don't use a tension wrench for this job ,just creep it up till the preload feels right

The MGB manual quotes exactly as you have done, reassemble and 140 ft/lb

If you've got it up to 140 the collapsible sleeve will be well and truely doing it's job correctly--If it's length was out the drag on the pinion would be up the creek
It's all good, forget about it-------------
willy

With the decrease in backlash, maybe the nut hadn't been tweeked up previously, some people run them without much pinion drag for some reason--How much backlash do you have now----------
William Revit

Willy, thanks, that's reassuring. You've convinced me! I don't have a dial indicator so cannot measure the backlash, it was more "notional".

For others reading this, with the pre-load measurement, you do seem to have to get it rotating with the spring balance attached i.e. to overcome the "stricktion". I think that attaching a weight from a static position may not give the correct reading.
Bill Bretherton

Bill yes i found that when I did mine, the initial sticktion was a factor. A gentle turn of the pinion was needed.


I have a dial indicator if you want to measure the backlash.


Chris Madge

Thanks Chris.
Bill Bretherton

Les
Just read your bit---30 ft/lb is a bit of a worry, You probably need to have another look at that-Maybe the seal is really tight, sometimes, some seals are fairly 'stiff' and can get tight if you happen to have a Speedy sleeve on board--? If you're doing it in the car, are your brakes nice and free (adjusters backed off--

On marking the nut--You have to mark the nut and the shaft, not the nut and flange unless you're 100% sure the flange goes back on exactly the same spline (which you need to mark to the shaft to get right)otherwise your mark will be in the wrong position relevant to the shaft

I'd be tempted to pull the flange off, sneak the seal out, refit the flange and creep the nut up till the drag is just on the loose side of spec. like around that 5-6 in/lb drag,making sure the brakes are free if you have the axles in, then, mark the 'shaft and nut', pull it apart, fit the seal,flange,nut and tighten the nut back to the marks and go just a little past the mark, just enough so you can see it's just past the mark like 20thou max. on your paint marks
William Revit

Thanks Willy, indeed I marked the flange not the spline. And yes the seal is new and there is a sleeve on the flange.

Looks like another session under the car coming up.....

Les
L B Rose

You were right Willy, the oil seal was really messing up the preload reading. Damned difficult working under the car but I think I have it about right now. No room to swing the torque wrench on the nut but it's as tight as I can get it with an ordinary socket spanner. My home made pound/inch tool shows the preload is OK.

Les
L B Rose

Good stuff Les, I was a bit worried about that 30ft/lb
You'd get a laugh out of me, I read your vehicle profile, and couldn't get my head around what a heated tacho was, I reckon I read it 3-4 times before it hit me I was reading-- heated tacho-- instead of heater - tacho
Cheers
willy
William Revit

Ha ha Willy! Yes the heater and tacho were optional ex works. I have heard about a car with no tacho, but never with no heather in Northern Europe!

Les
L B Rose

is that Romany lucky heather Les?
;-)
David Smith

Very belated ha ha David.
L B Rose

This thread was discussed between 27/01/2020 and 19/02/2020

MG Midget and Sprite Technical index

This thread is from the archives. Join the live MG Midget and Sprite Technical BBS now