Rear gear cluster "floats" when freewheeling

[plancashire]

I had a bike in the volunteer workshop today donated by a friend. It's a fairly cheap youth bike with 24" wheels, 3 chainrings front and 6 or 7 rear (not checked those in detail yet). Mostly I can bend the bits back into position that were abused during vigorous play in the woods (according to his mother) ...

I noticed when spinning the rear wheel in the freewheeling direction that the gear cluster has a slight side to side motion, as if it is moving on a slightly different axis. You could describe it as "floating". Otherwise it runs fine. I don't think it is enough to affect the gear changing and the bike is not very worn generally. I'm just curious. What causes this?

[531colin]

The axis of the freewheel isn't exactly concentric with the wheel axle. Either the hub itself or the freewheel is basically wonky!
It is as you say, on a different axis.
(I'm a bit surprised you haven't noticed this in previous bikes, but I guess it was more common when every bike had screw-on freewheels, before freehubs.)

[853]

I saw this 40+ years ago on my Regina freewheel, and it looked disturbing but wasn't a problem.

I was told the term for this was precesion and that it was due to the weight not being balanced properly in the rotating parts.That's what I was told, but it may not be correct, and it was a very long time ago ...

[gregoryoftours]

It's very common with freewheel rear wheels, especially nowadays when they're likely to be low end. The hub shell is made to low tolerances, often even only stamped steel, the threaded part of the hub that the freewheel screws to is not perfectly central and parallel to the hub axis. In operation it doesn't really matter, being 5-7 speed the sprockets will be widely spaced so it won't affect the operation of the gears.

[MartinC]

It's ubiquitous with freewheels because the precision needed needed to machine the hub/freewheel thread exactly concentric with the bearing race in the freewheel is far too expensive in practice. So the sprockets rotate around a slightly different axis to the hub making the wobble appear.

IIRC Bernard Hinault couldn't tolerate this and his mechanics had to discard (I guess they ended up on other team bikes) them until the found one that that has fortuitously end up concentric. For us mere mortals it makes no difference at all.

[plancashire]

Thanks for all the explanations and reassurances. It is as I suspected.

[Brucey]

The axis of the freewheel isn't exactly concentric with the wheel axle. Either the hub itself or the freewheel is basically wonky!....

a tiny speck of crud on the mountings will do it too. One of my first freehubs (on an MTB) was actually pretty bad. I had to dismantle the thing and subject the body washer (then a new idea) to some fairly serious deburring.

FWIW the OP's freewheel may just have one too many shims in it, thus making the bearings inside super-slack. However I have been fooled by tooth shaping too. One of the tricks they use to improve shifting is to chamfer all the tooth tops. Often they are all chamfered in the same way for half a turn, and then the other half of each sprocket has tooth tops that face the other way. Neighbouring sprockets are treated similarly, so it can give the impression that the sprockets are not true, when in reality, it is just the tooth tops facing different directions.