Originally Posted by
Brian Ratliff
Yes, there are small accelerations and decelerations during a pedal stroke. They do not lead to net kinetic energy loss for one wheel weight over another, but they will make you more tired in proportion to the wheel's rotational inertia. The small accelerations/decelerations are pure hysteresis loss from the body's perspective. Every pedal stroke, you expend energy to accelerate the wheel, but your body doesn't recover during the small deceleration through the dead spot in the stroke. It's a one way street.
I think we're in agreement on wheel dynamics. I'm not sure about the second part here: by "returned," I'm not saying the energy is returned to the leg - that's a given - but to the system as a whole. The power _at the hub_ is independent of rotational weight (w.r.t. non-rotational), but you bring up a point about how the leg develops that power.
So what it comes down to is if the leg more efficient at an ever-so-slightly more "circular" pedal stroke with a heavier wheel than without. Certainly, TT'ers think so. Elliptical chainrings, biopace, etc., etc. all try to affect the leg dynamics to improve efficiency, but has anyone shown that they do? It would be virtually impossible to measure (?), and even harder to separate out the psychological effects.