Welcome to our resource for MG Car Information.
MG Midget and Sprite Technical - 'Glitter' in diff oil
|I got round to changing the diff oil last weekend and am somewhat concerned to find a collection of small metal flakes in the oil. Something tells me that this is probably not a good thing.|
I'm contemplating removing the diff this weekend for investigation purposes. Anyone else had this and any clues as to what I should be looking for?
The things that do wear quite rapidly in the diff are the two "olive" spacers on the planet gear shaft. I think they are made of phosphor-bronze which may explain the "glitter" flakes. When worn one symptom is excessive lash at the wheel rim - i.e.more than about 1 1/2". They are cheap and easy to replace without needing to disturb the pre-load on the pinion.
Of course it could be something else, but I cannot think what!
That might make sense. I do have a significant amount of lash at the wheel rims.
18 months ago I had to have the diff oil seal replaced. This was done by a local garage since I couldn't get sufficient leverage on the nut with the car raised up on the ramps. I'm hoping they did their best to restore the pre-load by putting the nut back on in the position it came off in, but I guess they could have toasted the collapsible spacer :(
I'll check the thrust washers this weekend as a first thing, before I go looking any further.
What colour is the glitter?. Likely choices are silver or bronze, may help track down the source.
I'm 99% sure it was silver in colour.
|Hmm, that would suggest its not the olives! But they are easy to inspect once you have removed the diff.|
That doesn't sound good. It would be interesting to know if the particles are attracted to a magnet, I suspect they are as I can't think of anything silver coloured in the final drive assembly that's likely to break up that isn't magnetic. BTW where abouts in Wiltshire are you, big county.
|my vote is that the shop over-torqued the pinion nut and lunched your bearings.|
The crush spacers supplied today are too strong, mine reached the specified 13in-lbs bearing preload at under 120ft-lbs (not the 140 ft-lbs the book says to go to). So, if someone installs a new crush spacer and goes up to 140, without checking the bearing load, and they got one of those strong spacers like I got (twice, since I overshot the first time) it'll overload them and quickly wear them out.
Alternatively, they marked the pinion before removal, and then conscientiously re-installed it and turned the nut to the same mark but DIDN'T notice if the pinion flange, when reinstalled, was put on the same tooth (it is splined to the pinion shaft) so the nut mark would make sense (an easy mistake to make).
In either case, the repair is the same: strip the diff to R & R the pinion bearings (and, as long as you are in there, the olive and thrust spacers in the cage, as already mentioned).
These can all be done without special tools, if great care is taken to get everything back right where it came from, and if that new crush spacer is monitored very closely during the pinion nut torque.
Of course, all of this is dependent on whether your ring and pinion were damaged by the metal flakes, but they are pretty stout, if this has been in over a few miles, they might still be ok.
Thanks for the advice. I will remove the diff this weekend and report back my findings.
David, I'm in Salisbury.
Just another option to check - my diff was whiney at certain speeds. Cause of that was that the thrust washers behind the sun gears were disintegrating. There were bits in the oil, but they were carbon-black rather than glittery silver. The sun gears were floating due to the washers having gone, and this was causing the whining. The washers behind the sun gears can be easily replaced while you have the planet gears out if you're going to do them.
But, from the description, I think my money is with Norm and that the bearing pre-load has been over-done. What a pain!
I've run across three type of those thrust washers, steel, bronze, and tufnol. In my limited experience I've only seen the tufnol ones disappear, usually leaving a telltale remnant in the radius at the base of the gear. Unfortunately I think the new replacements are tufnol. I've still not gotten around to making some bronze replacements.
|David - My bits must have been Tufnol - certainly the replacements that I fitted are Tufnol...had no problems with them since I did it.|
|'Should' it be the preload being overdone, anyone found a suitable tool for setting the preload?|
I have found something that looks like it should do while I was trawling some MTB websites:
Mind you, from the look of the workshop manual changing any of the bearings looks very involved and requires several special tools for bearing removal and re-fitting. Not to mention setting the pinion position....
I'm so hoping it's npt the pinion bearing....
its not so much the preload as the postion of the pinion thats the problem, i've a diff in bits at the moment & have tried to find dimension A of the BMC factory tool without any success.
However the new bearings measure exactly the same as the old so i will probably miss the pinion spacer washer adjustment part & hope for the best, i have a load of new shims for the other bearings, though the inscription on the pinion & crown makes no sense to me at all so that'll be another work it out myself task.
|you can do the whole thing with basic machine shop equipment:|
- dial indicator
- press (with various sized arbors for pressing)
- bearing puller (for pulling the "inner" pinion bearing off of the pinion shaft)
- seal puller (screwdriver will do in a pinch)
- ft-lb torque wrench
- blueing ink (for checking tooth engagement)
- clean and orderly work space to keep the parts in so they all go back in exactly the same place (like, the ring gear "clock" position relative to the cage, or the cage bearings on the right / left side).
Oh, and the in-lb torque wrench. That you need special, because typical in-lb torque wrenches only go down to 50 or so. We need 6 ~ 15, so you need a special one designed for that low range. I bought mine from Precision Instruments. You can also use a small oz. spring scale and a bit of ingenuity, but I was surprised at how very useful it was to be able to rotate my pinion 360deg and measure "all of the way around".
Parts I recommend:
pinion bearings, cup washers, thrust washers, pinion seal, crush spacer, crown wheel lock tabs (3), diff pin, and the spring pin that holds that cross pin in the cage.
I took mine apart first, then measured the cup and thrust washer's thickness so I'd order the right ones (there are 4 sizes of thrust, and 2 sizes of cup). That was when I decided my diff bearings, ring & pinion, and the sun/planet gears were all fine, and the diff pin was worn a lot, time to replace. The pinion bearings were not only severely worn, the color of their races was dull, dull, dull (I had overtorqued the pinion nut when I replaced the seal some years ago).
No one could supply me with the little spring pin, and the size was not stocked by any of the local hardware stores, so I ordered it from Grainger ($15 for a bag of 500). I kept 10, and gave the other 490 to my parts guy, Robert Kirk, who is a great source for these kind of parts from. All of the parts fit like a glove, and the whole thing measured right back in spec once I fitted them all.
The ring gear will have written on its perimeter what the backlash should be for that particular one, and that's what you use the dial indicator for, to confirm backlash and runout. But if you just R & R bearings it should be right when you put it back together (I checked anyway, and it was).
All of those fancy shop tools in the Bentley are just there to speed up the job, and for when you are starting from scratch with a new ring and pinion. Essentially, any shop who can do a Mustang rear end can set up one of ours for us, if we want to farm it out.
Or, you can just toss your whole pumpkin in the hopper and buy a used one, as there are supposed to be a lot floating around for low bucks (midgets rust faster than their differentials wear, apparently). My ring and pinion looked like new, so I opted to give it a go (haven't driven on it yet, but it went together super well and I am optimistic). Even if it howls like a banshee, it will have been an interesting project that I was glad to give a try too.
Norm "what's the diff?" Kerr
|He may sell very good torque wrenches, but i'm too tight to give him my money, heck i even used a preloved tie wrap. The big socket weighs a pound.
|Before & after thruster washers, i never did find the other one.
|The diff is much better now, could have done with another thou on the pinion spacer & new differential gear thrust washers which are NLA, hopefully i didn't make a dogs diner of it.|
Handy those wine boxes :-)
Those look like the planet gear thrust washers, the new ones, and the remains of a tufnol sun gear thrust washer. If so the picture is a bit misleading.
|There was no planet thrust washers just heaps of swarf that bit was the only thing left semi intact, i can't remember exatly whrre it came from but its in no shape to be good for anything.|
I've pulled a few bits like that from the radius, undercut actually, at the base of thrust face of the sun gear. Due to the undercut when the rest of the thrust washer disappears and the sun gear runs metal to metal with the diff cage that's all you end up with.
I've seen a few quite worn planet thrust washers but no where near the extent that you ended up with a thin sliver remaining.
|Well i've it all to do again when i pull the one in the car, as thats the spare 4.22 diff i have just rebuilt.|
I really should stop buying midget bits, thats 4 diffs i've got now, amongst other bits.
I like the idea of using the bar a socket hanging on it. Just a thought though, don't you have to take in account the weight of the bar? The original tool from BL had a counterweight to balance it. The weight of the bar must be adding to the torque reading, so when your 'weight' is 7 inches from the centre point of the socket then you have 7 in/lbs plus whatever the bar adds .... or am I missing something completely here (like maybe they're not inches marked on the bar!)?
your right, though as the socket being a tad under 1lb, it's close enough for me for the range of torques specified for the preload.
I don't think my torque wrench is that accurate at 140 lbs ft which is near it's max setting.
|Excellent. My torque wrench max's out at 150, so it's not that accurate at 140 either. Using your system seems like a good plan!|
|The term "crush spacer" is new to me. I thought it was only a rigid distance piece that was not supposed to limit the preload on the bearings. Can the old one be used again?|
I notice from the Moss web catalogues that this part has different numbers in early and late axles. My ´61 Midget has of course the early axle. Is anything known about the quality of these replacements?
|Hmm, removed the diff this morning. Have 3 sorts of 'glitter' in the oil. Silver magnetic, silver non-magnetic and copper coloured.|
Used my parts cleaner (B&Q bucket and some Gunk) to clean it. Some of the glitter appears to be from the diff casing round one of the bearings!
However there seems to be a lot of lash between the crown and pinion. So much so that you can move the crown wheel 1-2mm without any movement in the pinion gear. This makes a nice 'clonk' noise too as the crown gear makes contact with the teeth on the pinion gear.
The planet thrust washers are disappearing too, so I guess that this is copper coloured bits in the oil.
I have not been able to measure the preload yet, but it does seem very tight and not very smooth just turning by hand.
I have a basic bearing puller and a torque wrench, but that about it from Norms list. I'm thinking that this one might be beyond me :(
|And a pic of the rest of the diff:
|Are any of the glittery bits magnetic?|
i used a universal 115mm grinder 'puller' to remove the pinion bearing, if your carefull it's easy enough.
|Tore, there is a way to use a rigid spacer with shims, in place of the crush spacer. It is a more sure way to set up a diff, but takes longer to dial in, so they used a crush spacer in mass production (and most modern RWD diffs are made the same way, with a one-time use crush spacer). If you have a solid spacer, it is re-usable, in answer to your question.|
That huge freeplay between your ring and pinion is likely due to failed pinion bearings. The key question is this: the bearings are thrust washers are all replaceable, you can do the whole diff for about $100 worth of parts. If the R & P teeth are worn, or, for that matter, the diff gears too, then you are probably best off starting with a different one rather than trying to repair this one. But, if the gear teeth all look very good (no dull spots, no chips, no scouring), it might be worth the effort.
Phone around for shops who do hot rods, or cars with RWD that get ratios changed for performance (in the US, this is easy, any shop that specializes in custom engine/driveline work on Mustangs, Camaros, etc.), they should be able to do the whole job, even setting up a new R & P (setting up a new R & P is substantially more involved than just replacing the bearings).
If you think your gears look good, but just need the bearings and washers replaced, then your good, local machinist ought to be able to do the remove / replace for you (and can check all of the lash, clearance, run-out and bearing pre-load for you, if you give them a copy of those pages out of the Bentley that describes those values - frankly, most of the key info is universal to all differentials, and the stuff that isn't is (believe it or not) hand written on the edge of your ring gear.
hope this helps,
|Some of the glittery bits are definitely magnetic.|
Brad, my only concern is that if I do take it apart, I'm commited to putting it back together! If I don't take it apart I can use it as a part exchange against a reconditioned part. That's my current dilemma!
Just want to say thanks for all the great info. I've been thinking about this, and I'm going to give it a go. I've had a good look at the C&P and the teeth look like they're in condition. I'm hoping that if I replace the bearings only, and put everything back where it came from, all will be well. If the worst comes to the worst, I can source a second hand replacement. It'll be interesting finding out anyways!
|My original 1961 factory workshop manual says "Where it is only necessary to fit a replacement oil seal the axle may be reassembled in the reverse order of dismantling, assuming that the original shim thicknesses are retained." It also describes the renewing process very straightforward, and says that the nut should be tightened to 140 lb.ft.|
I suppose this means that early axles are shimmed, and later ones used a crush spacer, as their part numbers are different.
|Good man Tony,|
it makes posting on this BBS worthwhile if occasionally someone steps out of their comfort zone & tries something they normally wouldn't.
keep us updated & you know where to find help if you need it.
I really appreciate the help! Thanks again, and I'll keep you posted.
|in the UK, here are the two shops where you can get the parts you'll need:|
Good luck, and don't hesitate to ask if you've any questions!
PS: here is the tool you will need to make to hold the pinion while you loosen / tighten the pinion nut. Hint: it is the same tool you use to hold a mini hub when doing up that car's front wheel bearing nut to 150 ft - lbs. 1 piece of 1.5 x 1.5" angle iron, and a shorter piece of straight steel stock, a few holes and a few nuts/bolts and you're in business. A very useful way to get leverage on a variety of things.
|Nice reaction bar, i just put the flange in the vice, that held it firmly enough.|
Just wanted to say a big thanks to all those who contributed to this thread! I installed my rebuilt diff on the car last night, went for a blast down some country lanes and all was good :)
The rebuild used a 'universal leverage thingy' as designed by Norm and I measured the pre-load using a tool of Brads design! Sussex Classics supplied the parts and I borrowed a bearing and seal installation tool from a VW van enthusiast at work. Interestingly I also found that the correct preload was achieved at just under 120 ft/lbs so thanks for the warning Norm.
There is still a slightly excessive amount of lash in the diff, but it is way less than before. That and the pinion now turns smoothly. I'll change the oil again in a few hundred miles time but I hopeful that there will be no more shiney bits.
Cheers once again. Without BBS's like this one I wouldn't have the confidence to undertake some of the less routine tasks on the car. It always helps to speak to someone who has 'been there before'.
|Well done Tony,|
|Since Tony has succeeded, i don't feel bad hijacking his thread.|
can someone explain to me what preload refers to in this case, and how it is measured?
I'm not working on a diff, nor planning to any time soon, but I would like to understand!
The pinion on the diff has 2 taper roller bearings. In this case we are measuring the rolling resistance of the bearings, the 'bearing stiffness'. Between the two bearings there is a crush spacer (on my model of diff, see further up the thread for discussion of the non-collapsible spacer type). As the pinion nut is tightened, the two bearings are pulled closer together (as the crush spacer collapses) and therefore further into their races. As the pinion nut is done up tighter the bearings are pulled further against the races and the friction increases. The preload is then said to be higher.
In order to measure the preload I made up a tool which consisted of a 3ft bar with a socket in the middle of it. Once the socket was on the pinion nut in a horizontal plane there were therefore no forces acting on the pinion which would cause it to rotate (since my bar was balanced). I then used a 1lb weight on my balanced bar. With the 1lb weight 12 inches from the socket, and the bar in a horizontal plane, the force applied to the pinion is therefore 12 inch lbs (torque = force x distance). Note that the force applied to the pinion should be measured when the pinion is already rotating. The force required to start the rotation (the breaking force) will be slightly higher.
Hopefully that makes some kind of sense. I'll post a picture of my balanced bar when I get a chance. That's my understanding in any case!
|Ok, i get the preload now, i think. It's a measure of how much tension/pressure is being put against the bearings by the spacer and pinion nut, right? what about the diffs tha have a solid (not crush) spacer, is the preload more or less fixed by torque on the pinion nut?|
the balanced bar and inch/lb measurement also makes sense.
you lost me with this sentence: "the force applied to the pinion should be measured when the pinion is already rotating."
The diffs with a solid spacer have the same preload setting as those with the crush spacer. Instead of using a one time crush spacer between the bearings however there is a solid spacer and a collection of shims. This effectively means that the crush spacer collapses to the same dimensions as the solid spacer plus shims. The crush spacer is obviously quicker and easier to install since using shims and a solid spacer involves a certain amount of trial and error regarding the number and thickness of shims required.
Regarding the measurement of the preload in relation to the rotation of the pinion... it will take slightly more force to start the rotation of the pinion compared to the force that is required to keep it rotating. This is to overcome the stiction of the bearings. By ensuring that the pinion is already moving at the point of measurement then you do not have to take in to account the stiction.
This thread was discussed between 02/08/2010 and 10/09/2010
This thread is from the archive. The Live MG Midget and Sprite Technical BBS is active now.