So when is the NuVinci Hub Motor coming out? They already have the first half of your design package in production.
That's interesting, I hadn't seen that design. It's pretty amazing how much innovation is going on in the world of bicycles, let alone E-bicycles.
With respect to the NuVinci technology being used in a hub motor, I see a few issues. As you noted, it's going to difficult to get E-Moto torque levels through what is essentially a friction drive.
Second, as we can see in these pictures, the spherical balls eat up a lot more internal volume than a planetary gear cage. Here the planet gears look like nylon. If they were steel, they could be 1/3 as thick.
https://ebikes.ca/pub/media/wysiwyg/ezeestator.jpgThe NuVinci balls need to be substantial to handle the surface loads.
https://i1.wp.com/www.electricbike.com/wp-content/uploads/2016/06/BeltIGH19.png?w=696&ssl=1Lastly, again because it's a friction drive, the efficiency is quite a bit less than a gears. This test showed overall friction loss of 3% for a single speed bicycle chain drive (97% efficient). With a internal geared (Rohloff) hub that went up to 5.5% average for all gears. The NuVinci continuously variable lost an average of 16.5% over it's range of ratios.
https://www.cyclingabout.com/speed-difference-testing-gearbox-systems/Efficiency wasn't so important when internal combustion engines and cheap gas was providing the power, but becomes a bigger factor with limited storage capacity for electrical energy, and definitely when a human is providing the power.
My impression is that the electric motors have such a wide torque band, that in hub form they don't require a lot of ratios to do the job, and don't require continuously variable transmissions at all. The problem a non geared hub motor has, is that on a steep hill it's being asked to delivery peak torque at lowest rpm and something gets burned up, or at minimum, efficiency plummets. Maybe all somebody needs to do is create an internal planetary gearing system that has a high ratio planetary, that can also be bypassed for direct drive at high speed, and the controller shifts based on speed and/or power demand? Seems so obvious, I'll bet it's already out there.
Back to the idea of hub weight creating issues with their unsprung weight, this graphic shows a range of bicycle wheel hub motors. We can see a weight range from 2.3 to 8.5 kg, quite a variation.
https://ebikes.ca/pub/media/wysiwyg/2020MotorSpread.jpgI don't know about E-moto hub motors, but wonder if graphing power vs. weight for the bicycle hub motor will result in a fairly consistent factor that could be extrapolated.
It occurred to me that adding a rear hub motor, battery pack and controller, to a pedal moped, would result in a "hybrid motorcycle" that possibly retains eligibility as a non-licensed "low power E-bike".