Originally Posted by
dabac
You're missing a key feature: Wheels are pre-stressed structures. (in vertical) their dominant reaction to load is by losing some of their pre-tension. The only place on a wheel where you can readily measure a change between with axle loading and w/o axle loading is on the spokes directly underneath the hub. So although it's questionable in terms of literature and language the engineering perspective is clear, the hub stands on the lost pre-tension in the bottom spokes.
That's a big assumption (unless you have some proof to back it up). I can believe that a rim goes out of round when a load is applied to the hub. However, the only way to apply a load to the bottom of a rim by using a hub attached to the rim with spokes is to pull down on the rim (you can't push with a spoke). The only spokes that can provide that pulling force are the spokes at the top.
And in order for the rim to go greatly out of round at the bottom, the spokes next to those bottom spokes would need to stretch (increase in tension) to allow for the out of roundness (the material of the rim has to go somewhere). All of this is complicated by how many spokes are used, the cross section of those spokes, and the cross section of the rim. I need to go and read Brandt analysis and see exactly what he modeled and measured though.