An analysis of the nuvinci on a test bench would be great. Sadly most are too large. No chance at my faculty atm. One would have to build s special test bench for e-bike drives. A normal testbed integration platform is still too expensive. Maybe I could find a way to get this to work at my university but that would have to wait a year or two. It's not that simple to get cash for such a project. I will ask my professor in a good moment what he thinks of this idea. I now there is an old testbench for nitro model engines, but I guess the scientific value isn't very widespread. The tricky part are still the torque transducers.
A very simple way to build a testbed for e-bikes would be a hill with straight ways with different acclivity. nothing one could start a true multi dimensional DOE optimization on, but it would be feasible enough for creating an efficiency map of the nuvincy.
Sry for being so OT, maybe miles and me should start a new thread about the sense...
stepping down and then again stepping up does not make much sense I agree. But a second drive at the rear wheel doesn't look simple. Simple looks cool.
an enclosed drive looks "best" for most customers. That does not exclude each other so that's a good approach in my opinion: A "direct" enclosed rear wheel drive. Your approach already comes very close, but the dia of your belt ring is a bit too large to get a belt protection around it properly. Thats why I plan a smaller one -an adaptation of slaweks angle drive. My vision is a drive build into the swing arm, lightweight,cheap like dirt and about 1.5-2kw peak output power.
These Nuvinci drives could be made for less then a hundred bucks if mass produced (what they in fact already are).