Originally Posted by
Point
So, why is press fit “optimal”? There has been no correlation shown between any measure of stiffness and efficiency. Numerous threads on this site have been dedicated to that one. Weight maybe, but we’re talking about pretty small amounts when the bike and rider are taken in account as a whole. A 100 gram savings for a 160 lb rider/bike combination is a 0.137% advantage, and then only really on climbs. A BSA bottom bracket is limited to the diameter of the axle as you stated. A 24 mm axle is stiffer torsionally than say an old 17 mm square taper (higher “J” value), but any amount of “wind-up” (very small) in the axle will be returned with essentially no loss of energy during the pedal stroke. As for external bearings, they don’t add anything to stiffness since that is a function of the frame construction, not the location of the bearings in the BB area.
For a metal bike, sure. Jan Heine also makes the same "planing" argument where he claims that the springiness of a flexy metal bike will allow you to go faster. But we can agree that the optimal race bike is probably made of carbon right? And not metal? Which means that you can throw your ideas about energy return out the window. Carbon doesn't return energy the same way metal does. If you made a carbon frame as flexy as an old steel bike, it would feel like wooden garbage.
lighter is always better holding all things equal. I shouldn't have to explain more.
A wider bearing stance has some benefits, but yes. Moving the cups themselves outboard doesn't help with stiffness. The only purpose of that was to make room for a 24mm spindle. That's my point. BSA is a hack because it doesn't allow for the frame to fully optimize the construction wrt the bearing stance. An optimal race frame would use something like T47 (Trek) where the frame itself has a wider shell and the bearing stance is also wider.
Also I will disagree that torsional stiffness is the only benefit of a larger spindle. You also have a regular ol bending moment. Sure, with an infinitely stiff frame, this wouldn't matter. But a stiffer spindle will do a better job of transferring the load across the shell and will thus keep the bearings in alignment. At least, that's what I think.