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 MGB Technical - Banjo axle problem

This may be a rather long post but I may help to explain how I got to where I am.

I have a 1974 MGB which I use in standard class sprints. We are allowed to use parts from any year of MGB. Retrofitting the lighter banjo axle is a common way of making the cars lighter. I acquired a damaged banjo axle from a friend of mine who had used in his trials car.

Prolonged wheel spinning in the trails car had lead to the small bevel gears seizing on their shaft. This had caused the shaft to brake its locating pin and protrude from the rotating carrier until it caught on the end of the pinion. This resulted in some burring to the end of the pinion teeth.

I rebuilt the diff using good quality used small bevel gears, a new shaft, locking pin and diff carrier bearings. Having never rebuilt a diff’ before I relied heavily on the factory workshop manual. It uses special gauges to a) set the pinion position and b) set the crown wheel position. These gauges or their sizes are not available. As I had not changed anything that would affect the pinion position I accepted it as correct (although there may have been some wear during its life). From reading the manual I decided that the use of the gauge in setting the crown wheel position was to compare the effective width of the bearings, taking account sideways play, with their nominal width. I obtained this using parallels and a dial gauge and micrometer. This was followed up by checking the mesh using engineers blue as described on the MGAguru.com site. Although using blue I seemed to get more mess than clarity.

I checked the backlash using the method in the workshop manual where fig H14 shows the dial gauge bearing on the face of a crown wheel tooth at its outer end. Interesting, MGAguru talks about measuring backlash at the input flange which strikes me as more difficult to do.

The axle is now on the car and I took it out for a drive. The good news is that with power on the axle is silent. The bad news is that on the over run it lets out a howl with a slightly graunchy tone. So clearly I have got something wrong.

To try and get some idea as to the problem I raised the rear wheels into the air, locked up the brake on one side, jammed the prop shaft so it could not turn and measured the rotation available on the remaining wheel. There was only 3 or 4 degrees of movement at the wheel.

Any ideas?
David Witham

I assume that you re-used the old ring and pinion gears. The minor damage might have worn off the case hardening on the gear faces and will cause increased wear and noise. It sounds to me like you don't have the gear backlash set correctly. This will give the results that you have described. The wear pattern, when setting up the gears, should place the gear mark in the middle of the corresponding teeth. RAY
rjm RAY

Did you set the carrier bearing preload?
Backlash to marked value?
I think it reasonable to set backlash slightly over marked for well used gears - judgment call.
Did you examine pinion bearings?
Check/set pinion preload?

It can be difficult to set by bluing on worn but still good gears. You really need to understand what you are about. Not really fun!

Gears are deep hardened, not case hardened.

FRM
FR Millmore

I guess I need to take the diff out of the axle and have another go with the blue. Any tips on using engineers blue? Should the blue go on very thinly?
David Witham

Yes. Get a lot of cotton swabs and good oil solvent. Clean all oil off all teeth. Put blue thinly on about 4-5 teeth on the ring gear, both sides. Then put a drag on the ring gear (maybe a wood wedge down the case against the gear, and rotate the pinion about two turns. That will wipe blue off the first teeth, that were originally blued, and maybe deposit some on teeth not blued, in groups obviously spaced by the # of teeth on the pinion. Varying the drag can help sometimes. I used to mark the groups with chalk on the periphery, so I could find and compare them. Examine all carefully, make drawings of the contact, because you WILL forget. (reproduce pages of "blank" teeth) Ideally, the second group of teeth, which get the transferred blue, will show a bare contact pattern outlined in blue. Try to get a look at the pinion teeth too. After a dozen tries you should start to get the hang of it!

Reverse the rotation to check the backside in a similar manner.

This can be done with other thin paint like stuff; white lead was formerly used - good luck on getting that! Possibly some artist's oil paints, whatever is used now. "Engineer's blue" is just the artist's "Prussian blue". White, yellow, pink etc. Point being better visibility - steel is quite dark and thin blue disappears visually.

Used gears will generally have a less well defined contact pattern than the books show, as the very slight crown on the teeth has worn in to shape. That makes it both harder to "read", and more important to get it right. You must avoid heavy contact at any extreme (crest, root, inner or outer ends)of the teeth, on either side.

None of this is much use until you can answer my questions above.

FRM
FR Millmore

FRM, the answers are all yes. I had to make my own preload gauge.

As well as having a further go with the blue following your helpful instructions, I expect I should go through and check all the measurments again.
David Witham

David-
Best of luck, it is a very Zen thing to get into.

There is a wonderful story about one of the low dollar American hot rodders going after the Land Speed Record in the 60s. They found that the car was fast enough to run out of rpm at top speed, so they needed a higher ratio, which did not exist for their final drive. They found a pinion with the right number of teeth to give the ratio with their existing ring gear. Of course this gave a completely mismatched and incorrect set. So they set it up best they could, and spent the night driving the gearset with an electric motor, whilst spooning valve grinding compound over the teeth. Cleaned up and assembled, they set the record the next day. Didn't last long as that was when a bunch of guys were proving that you didn't need billions of dollars to go fast; Mickey Thompson came out on top.

FRM
FR Millmore

low dollar would sum up my aproach.
David Witham

Well I took the diff out and have been checking it on the bench. The backlash seems to be just under 0.002" but the factory scribe on the back of the crown wheel is 005. So I guess I got that wrong.

To change the backlash you move shimms from behind one diff bearing to behind the other. That means removing the nice nwe bearings which cost almost £50 a pair.

In the picture you will see my 2 pullers. I used the claw type to take off the original bearings but it wrecks them so I don't want to use it now. The other puller is of the correct general design but this cheap ebay example is to thick to fit in. So I will have to ask around my friends to see if anyone has something better.

David Witham

You can grind the cheap puller thinner.
Or use thin things like dinner knives to drive behind the bearings to shift them a bit. (not the good Sterling ones!)
Once you get the bearings off once, grind a pair of clearance notches into the locating shoulders on each side, so you can use small chisels as wedges next removal. Some cars have these as standard.

None of this is good for shims, but will save the bearings.

FRM
FR Millmore

This thread was discussed between 11/04/2012 and 19/04/2012

MG MGB Technical index

This thread is from the archives. Join the live MG MGB Technical BBS now