Jack- Chas makes good points about Normandy specifically and cones in general. I'll add a couple more. Rotating axles is a good idea and one that is done most every time the wheel is removed and reinstalled. Unless the Rr axle is bent and only one rotational position will give a straight tracking bike (or centered wheel between the stays). Then one tries to keep the axle rotation the same (being too lazy to replace the axle). Which leads the discussion to another Rr axle issue. The off set positions of the cones along the length of the axle. The left side cone and bearings are placed further out towards the drop out then the right side. This allows space on the right side to house the freewheel (the spoke flanges do the same). Just as with the wheel's dish and associated spoke tension's being uneven (left side VS right side) the load the axle bearings see are uneven. The right side bearing (and cone by association) will have higher loads then the left's. So a few things happen. The bearing elements wear at a faster rate and the axle will have a greater likelihood to bend/break at the inside edge of the cone. I don't think I have replaced an axle (from a FW hub) that was bent or broken on it's left side, ever. I have replaced many left side cones though. While the right side might see greater loading, the left is more exposed to the out side elements with less distance or shielding (that the FW otherwise gives the right side) to the out side.
The cassette hub design can address this, placing the right side bearing further out the axle towards the drop out. But the cassette design has it's own compromises. Just look at the tiny ball bearings the right side uses in the Mailard Heilecamatic (sp!).
I've kept this as on topic as best I could. There is so much more bearing design discussion that most would say was off topic.
> Just an observation and theory (not original I'm sure): After several
> recent Normandy hub rebuilds, in all cases, I found much greater damage to
> freewheel-side cone than anywhere else (otherside cone or hub
> races)...probably
> due to unsupported lever-effect.
>
> More to my point (which I'm getting to), I also found, in most cases, the
> freewheel-side cone to be brinnelled, spalled, or otherwise pitted at a
> single location along the cone race. I theorize that because the cone
> does not
> rotate (like hub races) a single (downward-side) point on the cone takes
> the
> maximum load...all of the time! I guess some would argue that a
> "properly"
> adjusted and lubricated cup/cone minimizes this effect. I'd buy that...but
> minimize does not equate to eliminate.
>
> So how to further minimize and/or prevent this damage? How 'bout rotating
> axle (w/cones) periodically? Easy enough to do but how often? I'll leave
> that to your discretion.
>
> Except for this disclaimer, the foregoing proposal ignores that
> present-day
> riders may be 50 or more lbs heavier than what designers envisioned, along
> with other contributing factors.
>
> Jack Romans
> Sacramento, California
> _______________________________________________
Andrew R Stewart
Rochester, NY