TR2/3/3A Rear differential

[CJD]

Took me a lot of years to notice a correlation between using an impact gun and bearing failure.

That's funny Randall. I used to have so many arguments with mechanics over that same topic. If I had a nickel for every time I was told "I can turn the impact driver waaaaayyy down"!

 

[sp53]

Thanks for the heads up on the impact gun and bearings Randall. I can see how that might push and pop the bearing. It never entered my mind before. It has been a while since I put a bearing in, and I have only done 3 I think, maybe 4. Plus I am not an expert. I reread and remember reading in the Hayes on how to do it and saw the pounds inch. I remember being jammed under the car and thinking how I am going to get this because just turning the flange takes some effort. I got thinking this reminds me of putting in a wheel bearing on my old 65 ford truck, so I set the drag like I would on a wheel bearing because that is what the whole set up looks like to me anyway. I did hit it with the gun because the key did not line up, but only had to move it fuzz.

I did the last one with the body off the tub and did it the same way. Thinking this reminds me of a wheel bearing. This time I was able to get better leverage, as opposed to being under the car, and only used the torque wrench.

I started a new project and I am going to replace the seal and look at the bearing. What do you suggest to do if the cotter Key does not line up with the hole and it is shimmed and torqued out at 95 pounds?

 

[TR3driver]

What do you suggest to do if the cotter Key does not line up with the hole and it is shimmed and torqued out at 95 pounds?

Remove the nut and sand or file just a bit off the back of it. But I have to say that it has never been a problem for me, the nut seems to always line up somewhere between 80 and 100. Don't forget there are two holes through the shaft, so two places where it might line up.

As I see it, the big difference is that with a wheel bearing, it doesn't matter if the hub & wheel can move around by a few .001". On a TR3, they actually are supposed to be set up with some clearance to allow for differential expansion. (Eg, under hard braking, the front hubs get hotter than the spindle does, and so expand more, taking away from the bearing clearance.) But with a differential, it is critical that the gears be held in the proper relation to each other, even under some fairly heavy loads. (Lessee, 100 ftlb at the flywheel, times about 1:3 for 1st gear, divided by the radius of the pinion gear is about 3000 pounds. And that's not counting what happens when you dump the clutch.) So the bearings in the differential get pre-loaded to ensure that they can't develop any clearance when the housing, pinion, carrier, etc. distort slightly under load. And, since it heats and cools slowly, differential expansion is a lot less of a concern.

 

[RC64]

With the road wheel off and both rear wheels off the ground you push & pull on the hub while having a dial indicator in place against it to measure the float. The specs are in the manual. Float is adjusted by the addition or removal of shims behind the hub

Geo, et al...

The posts by you and others in this thread regarding "end float" piqued my interest. Since I have my left wheel off to finish the installation of my fuel line from the tank I thought I'd check the float. Since this work was done by someone else I wasn't familiar with end float requirements. As it turned out, there is no float at all in either rear axle. That prompted me to do some more digging. Like you pointed out, this clip from Haynes illustrates your point. Similar information was found in the Red Manual.

9. Axle Shaft End Float Measurement and Adjustment (TR2,3,3A) - Later Type
It is important that the axle shaft end float is correctly adjusted on the later type rear axles as the differential unit cross pin has a thrust block on it whereas on the earlier axle it was not fitted. The correct end float of each axle shaft should be between 0.004 and 0.006 inch and is adjusted by using shims placed between the brake backplate and the axle sleeve flange on each end of the axle casing.

The next paragraph in Haynes references an "Early Type". However, I didn't see an references to float tolerances for this one mentioned.

So here are my questions.

What is the distinction between early and later types? How can I determine which type I have? My 3A is TS 51502. I also saw something about a change over on axles around TS1300 or so but I wasn't clear what it was that was changing. It may have been referring to a change in the cross pin.

The main question now is, what would be the consequences of operating with no float? It doesn't sound like it would be a good idea.
This is new territory for me so any suggestions would be appreciated.

Rick...

 

[TR3driver]

You should definitely have the later type axle; the early type was only on cars with front drum brakes. Probably the easiest difference to spot (with the drums off) is that the early axle has 4 bolt heads showing where the bearing retainer holds the backplate to the housing. The later axle has no bolt heads there, only six holes for the bolts that come in from the other side.

There is no end float spec for the early axle, as the halfshafts are located only by the wheel bearings. I forget offhand whether they are roller or ball bearings, but they are a type that will take some side force to either side. The later axles use tapered roller bearings that will take a lot of side force but only in one direction. So in a hard corner, the outside hub moves in until the side load is carried through the halfshaft, the thrust block, the other halfshaft and on to the wheel bearing on the inside hub. Sounds hokey, but works quite well because the tapered bearings are so much better at carrying side loads in that one direction.

.004" isn't much, you probably won't be able to feel it, so the dial indicator is a must. I also have the drum removed, as any drag in the brakes may make it hard to see the clearance, plus the backing plate makes a convenient mount for the dial indicator.

Running with no clearance would be hard on the bearings, and the ends of the halfshafts where they ride on the thrust block. Probably not too serious if the clearance is exactly 0.000" and you don't do a lot of hard driving; but I would still want to fix it. Of course, if it won't move at all, you don't know how much preload is on the bearings. They could potentially be loaded until they are about to fail just sitting there.

Ideally, I would want to remove the hubs and inspect the rollers and races for damage; but those hubs are hard to get off without ruining them even if you have the special tool (and impossible without it). So if the bearings feel OK, and you can't see any obvious signs that they are failing, you're probably OK in putting it back together with the right shim packs. I've never actually seen one of those bearings go bad, they seem to be nearly indestructible.

 

[RC64]

Randall...

Once again you've given me a lot to think about. It gives me the incentive to dig into the books and learn how this machine really works. I feel like a 16 year old with his first car.

I definitely do have the later type axle with the round flange with the six holes. Between the Service Manual and Haynes I think I now have a basic understanding of what fits where. As for how it works though, you cleared it up quite nicely with the following...

"The later axles use tapered roller bearings that will take a lot of side force but only in one direction. So in a hard corner, the outside hub moves in until the side load is carried through the halfshaft, the thrust block, the other halfshaft and on to the wheel bearing on the inside hub."

As for the condition of the bearings I'm confident that they are good. When this project was started several years ago the fellow that was doing the work then had the rear end completely apart, cleaned up, and rebuilt with new bearings.

At the moment though I do have a concern about float as you suggested. It makes sense that if there is no float at all I could be looking at excessive wear on the thrust block (a week ago I didn't even know what a thrust block was :friendly_wink. I have the drum off and with a good solid grip on the hub I can't get even a hint of movement. I'd definitely like to see if there are any shims installed, and add some if necessary. That brings me to a couple of questions.

1) Are the shims visible between the flange and the back plate?

2) Let's say I want to attempt to add shims. That means I'm going to have to add the shims between the axle casing flange and the back plate? From what I've read, it looks like it is possible to remove the entire half shaft without removing the hub. As you pointed out, it can be hard to remove the hub, so I'm not ready to take on that challenge at this point. The brakes haven't been serviced yet so pulling the brakes, lines, and cables isn't a problem. Then it looks like removing the bolts securing the hub housing to the flange would free up the entire axle for removal. Any "gottchas" in that sequence that I need to be aware of?

3) Any recommended source for shims? I see that Moss lists a couple sizes staring at .004 I think it was.

Any other thoughts?

Rick...

 

[carpecursusII]

Rick, as a fellow 16yr old with his first car I can tell you your not alone.
1) The shims are barely visible but can be seen if you know what to look for.

2) The job is very simple, tools requires are a screw driver and a 1/2" wrench. Simply pry away the lock tabs from the back side and remove the 6 bolts. I did not have to touch any brake components other than removing the drum. At this point the entire half shaft will slide out of the axle, set it aside and be prepared for some grease to fall out onto the brake mounting plate, I put a rag down so nothing got on my brake shoes. Slide the brake plate forward enough to insert your shims then reassemble. Check float with a dial gauge (I had good luck with the HF gauge and magnetic stand), and repeat process until you get the desired result. Total time 2 hours. A side note to that, once you have done one side, check the other before disassembly, as they are not mutually exclusive measurements.

3) I cannot comment on the sources, I had more shims than I needed but it appears mine were replacements most likely from VB or TRF, many other parts from this basket case were purchased from there by the PO and he did this "rebuild"

 

[TR3driver]

1) Are the shims visible between the flange and the back plate?

I've never looked, but I don't see why not.

2) Let's say I want to attempt to add shims. That means I'm going to have to add the shims between the axle casing flange and the back plate? From what I've read, it looks like it is possible to remove the entire half shaft without removing the hub.

Yup. Turn the lock tabs (if present) and remove the 6 bolts. The hub & halfshaft should slide right out as an assembly. Try to support it as you do, though, if you think you might reuse the oil seal. The shaft is kind of heavy and not polished.

The brakes haven't been serviced yet so pulling the brakes, lines, and cables isn't a problem.

With a bit of care, you can even leave the brakes connected (if you want). The back plate can hang on the housing, then you can support it slightly away from the housing while installing the shims.

Then it looks like removing the bolts securing the hub housing to the flange would free up the entire axle for removal. Any "gottchas" in that sequence that I need to be aware of?

Just what I mentioned, try not to drag it through the seal. In fact, if you're not certain the seals were replaced, I'd probably put new ones in. They're cheap, and easy while it's apart; but a giant PITA if they leak later and let the differential oil contaminate the brakes.

3) Any recommended source for shims? I see that Moss lists a couple sizes staring at .004 I think it was.

I bought a few from TRF, but I'm sure Moss is just as good. Note that the shims are sized so you can make differential adjustments. For example, if you need .002" more, you can take out a .016" and put in (3) .006".

Any other thoughts?

As always, it's best not to try to reuse the locktabs. So far, Loctite has worked for me, but you can get new tabs from the usual suspects. Don't forget to turn them after you've got it all back together the way you want.

Check the surface where the inner seal rides, and polish it up if you can see a mark.

I can't think of anything else offhand. It's tedious, but not difficult at all.

 

[TR3driver]

A side note to that, once you have done one side, check the other before disassembly, as they are not mutually exclusive measurements.

Definitely worth checking. But, if you get a different answer, then you're going to have to pull it all apart and rebalance the shim packs. The only way for the measurements to be different is if the thrust block hits the center pin (which it shouldn't). The whole process is covered pretty well in the appendix at the back of the (later edition) workshop manual (or Bentley reprint).

 

[RC64]

Randall et al,

I was able to take a close look behind the back plate today and it is possible to see the shims. It appeared that there was one shim on the left and none on the right. I also went back and reviewed some of the old photos of the work in progress several years ago. It's clear that the differential and the axle parts were completely overhauled at that time. The fellow that was doing the work at that point had a great deal of experience with TR-3s. I'm confident that he would have been well versed in the axle float requirements so I'm inclined to let it be for the moment, get the car on the road, and see how it looks after I get some break in mileage on it. Besides, as you all have pointed out, it's not hard to check and adjust if need be.

Thanks for the help. Once again I feel peacefully educated.

Rick...