General TR How is the axle tube attached to the diff housing?

One of the axle tubes on my rear axle became slightly loose, allowing maybe 25 mils of movement at the outer end. I noticed that one of the pins in the differential housing that apparently hold the tube in place became loose.

There are four of these on each end. The picture below shows one of them. It's that staked thingy between the opening and the tube.



I always assumed that these were 3/8-inch screws that were tightened down, cut off, and staked, but the loose one rotated freely in the hole, without moving in or out, so obviously it was not a screw. But if it's just a pin that is inserted, perhaps with a little force, and just staked, I can't see how it would be strong enough to support the forces on the tube, which are substantial.

Does anyone know precisely how this joint is put together? Is there more than it appears?

I see this kind of construction on other differentials, so it must be strong enough, but I really can't see why.

I had one missing on my 66 4A. It is threaded and I replaced it with a bolt, cut of the head, then staked it as original. If your bolt spins in the hole (mine didn't), you might want to rethread it to the nearest size. There is also a hole on the top of the diff for the breather/vent.

I might have added some JBWeld to the treads as a sealant, but don't remember if I did or not. Good idea though.

As far as rotational force on the tube, I expect that occurs mainly inside the tube but the tube does support the diff housing.

The tube carries the torque reaction to the springs. So similar torque to that carried by the shaft, but in the opposite direction so the diff housing doesn't turn (much). It also helps support the weight of the car, to a smaller extent. Lever arm between the wheels and springs tries to force the diff end of the tube towards the ground.

I have never had a loose tube on a Triumph. The Dana’s, GM and Ford tubes are all plug welded to the center housing. I’ve both narrowed and repaired these...once again, by welding. The tubes do see a lot of stress in normal use.

Thanks for the info. I was a little puzzled that I could make the pin/bolt turn easily, without moving in or out, but I couldn't remove it completely. Maybe the remains of threads or something like that. In the end, I just clamped it tightly with a C clamp, with a bolt between the clamp and the pin, and welded it to hold it. I think that should be as good as the original.

When I rebuilt the differential on my 63 I found all but one of the screws loose but the tube was not loose in the housing. I just tightened the screws and welded the end of the screws to the housing. I did however find one of the diff carrier bearing journals damaged from bearing rotation and found a suitable used one from a TR6 that worked just fine and set up perfectly.

Interesting coincidence - I'm tackling the same thing on the TR4 I have. In my case, the pin is completely missing - looking at the tube, it seems like it has moved outward a bit. I've sourced a replacement differential and will swap out.

Excellent pic, Randy! I never thought about the tube retention, as in never had to. It appears the tubes are pinned and then the pin is staked and/or spot welded. I would assume the hole in the pic goes through, so we are looking at the axle shaft?

I believe they are also originally a press fit into the pumpkin.

I think they may be press fit, at least on the earlier differentials - I didn't see any obvious threads in the hole. Yes that is the axle shaft showing in the hole, it explains a bit why I had a pretty big puddle of oil under the car after sitting for a while!

Thanks. That pic is quite illuminating.

This thread has me wondering if many of the rather common axle failures were actually caused by misaligned tubes. If the tube is only held by interference and a couple plugs, then it could easily misalign a degree or two...which then places a cyclic load on the axle splines. In ME class that's how we would test the high cycle fatigue life of different metals. We put a small nick in a round rod and then placed it under a bending load while we spun it. Every turn of the rod was a fatigue cycle, so by spinning you could accumulate many millions of cycles rather fast.

Now I'll have to make it a point to check my axle tube alignment!?!

John,
I am not an engineer but isn't the axle housing totally separate from the axle shafts. Therefore a rotational misalignment in the tube housing should make little difference in the axle shafts. I can see where once the housing gets loose that there would be a number of other issues arise affecting support that the housing provides for the axle shafts. Just curious.

Rotational misalignment you are correct. If the tube is loose, it could just as easily have a bending misalignment that would impose the cyclic loading on the axle.

CJD said:

Rotational misalignment you are correct. If the tube is loose, it could just as easily have a bending misalignment that would impose the cyclic loading on the axle.

Click to expand...

Thanks that I understand.

So, maybe it's not a good idea to jack the rear end from the center of the differential ? Or did everybody else know that and I did not ?
ps: I have not been jacking from the differential.
Doug

1964 TR-3B (is it OK to use that year?)

Another good reason not to jack under the pumpkin is that it's very easy for the jack jaw to contact the lip of the rear cover and bend it; causing a leak. ISTR that's mentioned in one of the books; but don't recall which one offhand.

As far as car year; I don't think there is any consensus on whether to use year of manufacture or year first sold. My feeling is that, since Triumph didn't really adhere to any kind of "model year" for these cars; using year of manufacture probably leads to less confusion.

But that's just my preference, it need not be yours. Variety is the spites of life (or something like that )