Wattage -Saw the video - not sure some answers were correct, or perhaps incomplete.
You mention higher Voltage isn't needed, or imply that it's pointless. Then minutes later you mention that speed is limited by gear ratio and Wattage. I assume the answer was disingenuous, or maybe just too oversimplified.
For others - gearing helps multiply torque, but each gear's use is limited by the power plant's rpm limit or other mechanical limits. With the controllers being used in ebikes (not boosting Voltage), battery Voltage can limit how fast the controller can spin the motor, requiring a gear-change to lower the rpm back down - this loses torque. If you had a 72V battery instead of a 48V battery, you could theoretically spin the motor 50% faster, keeping more of that higher (lower numerical) gear multiplication in play before requiring a shift. Re-phrased, a higher Voltage battery pack can allow a higher speed in all gears, and with a mid-drive as long as you can either keep up with the cadence, or go throttle-only to spin the crank higher than humanly possible, the higher Voltage pack could give you a couple mph max top speed in all situations, depending on the spacing of the gears. This is far more noticeable if you were to go downhill, as you become rpm-limited rather than drag/power-limited. This is the same reason racers build higher-rpm and higher-geared cars - get more torque to the ground for more rapid acceleration, staying in the lower gears for longer, and more easily fight drag at higher speeds.
Not that we need people going 40mph on an ebike The speed advantage is also noticeable uphill, being able to stay a couple gears down lower while the motor spins away. This can be a 20% speed difference while bogged down in the 10-15mph range (for more widely-spaced derailleurs - I doubt the gain is as large on the 14-speed IGH).
Of course people experience this themselves with a normal bike - you can only pedal so fast, before you need to up-shift. If you hit a hill, you need to downshift to get more torque multiplication at the wheel. Try limiting pedaling to 60rpm and go for a ride, then do the same but limit pedaling to 90rpm; the 90rpm test will have a higher top speed, and likely be a gear lower.
1. What I didn't mention is that Ultra can realistically do only up to 3500W (max power). That is a real world upper bound on motor wattage and is still within the design spec of a 52V pack (at 70/80A). If you drive more than 3500 W, the motor heats up and the controller will shut it down.
2. Further, my understanding is that BMSs are not built to serve under a certain amp draw. e.g. if you get a BMS for a 72V battery, the BMS wont support a discharge rate of 5A or 10A. There is a low draw cutoff. May be someone with more knowledge can comment on why / how that works but that is the case.
Where I am going with this is that for the 72V battery pack, the min discharge rate floor will be higher. So most of the times, it actually wont be able to support a street legal 750W motor config (which can go from 400W to 1200W).
What that leads us to is actually a tier of battery voltages (48 or 52) that are kind of universal for the 500-3500W range. This is for theforseeable future, the ideal pack size. If you have a motor under 500W, you are going to have better performance with a 36V pack since that min discharge rate is lower than the floor of the 48V pack. (which is why bosch uses that). Note early versions of bafang used a 24V pack for 100-150W motors. Then they switched to 36 .. and so on .
So in the wattage performance tier we are in, for all practical purposes, higher voltage is not needed because it doesnt serve the use cases we need to serve. Further, a higher voltage pack will actually restrict our ability to run 750W configs. A 72V version becomes a viable alternative only if we run the motor exclusively at 2500W and above.
Throttle - no/little ramp up causing drivetrain shock. The controller can be designed to easily control ramp rate though. Just don't give that full power immediately, even if commanded, and ideally have shift-detection to prevent banging gears. Maybe if a person is already going 25mph, give them a more aggressive ramp rate. More feedback from sensors would be nice, but you can get by with just speed. I'd hope the electronic Rohloff shifter in the future would provide better feedback here also, and maybe ease more power in as the speed increases. This is how some fuel and timing tables work in one of my cars.
I still don't understand restricting throttle at higher speeds. Perhaps you're thinking someone may be in a lower-numerical gear at 20mph, and throttling away (legal limit for this on the street). At any speed I can't imagine the throttle being any different than a person ghost-pedaling, other than you have a cadence sensor to help identify what gear they may be in? If so, then what would be the minimum cadence that the controller will allow more than 750W? Is it limited by combination of cadence + speed?
1. Most use cases of throttle are to use one from a stand still. Or, riders use a throttle while going uphill (they will be pedaling in the taller gear to gain speed, and then push a throttle without consciously changing the gear). These are typically the cases where most of the unintended damage happens.
2. Yes a throttle can be designed for a gentler ramp up. Riders can also feather the throttle. However most riders just twist the throttle and go. If the bike doesn't move (taller / higher gear), they will twist (push) more and break the top gears. Sending more torque instantaneously. The larger the gear (physically) the lower the breaking torque limit.
3. If you are on a flat surface, you can get to the same speed with a 750W throttle, compared to a 2300W throttle. It will just take a little longer.
4. Once you are in a taller / higher gear at say 15MPH or higher.. you can still throttle on the taller gear with less chance of gear failure.