how do you think this happened?

[Brucey]

Cone precessing and crushing the hub....in my experience, it only ever happens when the hub is used as shown, with the drive on the left....

we have a winner! However the wheel in question had been used with a screw-on multiple freewheel in the usual way...

I think what happened was that the RH bearing cup/cone had become worn, so a new axle (probably a generic weldtite one) had been fitted. However judging from the wear marks on the RH cone, the balls didn't really get a chance to run on the bearing cup in the right place, certainly not for long, possibly at all. The RH cone was not tight against the locknut and had wound itself into the hub, crushing it as Colin says.

I think that in this case precession can cause the cone to wind itself inwards, and at some point the failing bearing started to seize up, winding the cone in wholesale as the wheel turned. I've seen bearings fail before but this is one of the few times I have seen the hub fail in compression in this way. The hub centre isn't especially slim, being 17mm dia; an old Maillard one would be 1mm smaller than this with a similar size hole up the centre....

The LH cup in the hub is also cracked and broken but is still more or less in one piece. The RH cup came out in pieces...

cheers

[ardeidae]

.

[Brucey]

I can see that happening if the wheel is stress relieved in an uncontrolled fashion. Some of those older shimano hubs don't have a perfectly centred bore and can move...

cheers

[531colin]

I have seen many new-ish pressed steel hubs fail in a similar way to Brucey's........always the rear hub, always the right cone that winds in and breaks the hub, although in the case of these modern (ie wide-barrel) pressed steel hubs all that happens is the bearing cup is pushed into the barrel.....the customer presents a ludicrously wobbly wheel, and expects you to adjust the bearings.
In the sixties, we all rode pressed steel (British hub co.) hubs, and i never saw one fail.....?

[Valbrona]

Why don't they just thread one side of the axle in the opposite direction to the other side?

[Mike Sales]

A thread is a long, shallow ramp, a good way to give enough mechanical advantage to crush a hub. Some axles used to have a shoulder to ensure that even a loose cone would not move to cause this damage.

[JonMcD]

In the sixties, we all rode pressed steel (British hub co.) hubs, and i never saw one fail.....?

Didn't a lot of hubs in those days have a fixed right cone so that this mode of failure was impossible? I suppose there was a reason having one fixed cone fell out of fashion, but at first glance it seems like a good idea.

[531colin]

Why don't they just thread one side of the axle in the opposite direction to the other side?

Well, if the cone locknut is properly tightened, there isn't a problem.
But I do remember hubs (long ago) where the right side cone tightened against a sort of collar at the end of the threaded bit of axle, and the adjustment was via the left cone.......and then again, roadster front cones never had locknuts at all....the axle nut performed that function!

[NATURAL ANKLING]

Hi,
I used to turn up at peoples houses to adjust the chain (motorbikes) and you would take the split pin out of the rear wheel then undo the nut with your fingers

Some times you just have to do the nut up..........hard.

[Brucey]

A thread is a long, shallow ramp, a good way to give enough mechanical advantage to crush a hub. Some axles used to have a shoulder to ensure that even a loose cone would not move to cause this damage.

and some still do; I recently stripped down a budget SBF sturmey archer brake hub. This is designed to have cup and cone bearings, with the brake on the left side. The bearing adjustment is also on the left side. I wondered if it might be easier to adjust the bearings on the right side instead, but saw that there was a locknut but there were no flats on the cone to allow adjustment; I thought about grinding some on but when I pulled the axle out, I realised that there were some crimped bumps on the axle which are presumably meant to work like a full shoulder and prevent the RH cone from precessing inwards and causing trouble.

In a similar vein if you get an old roadster with no hub locknuts, the non-adjusting cone fits up against a shoulder and this one goes on the right side; the adjusting cone goes on the left.

I've seen quite a few hubs destroyed through precession (including shimano freehubs where the RH cone and locknut were not tight enough) but this one is probably the worst I've seen in terms of collateral damage...

BTW I am pretty sure the hub nut would have been tight, but that they axle had two locknuts and a spacer as well as a cone on the RHS. The track nut load would have been reacted by the locknuts, allowing the cone free rein to turn...

cheers