Mastodawn

Riding the #lectricOne around a bit and I'm just astounded that they put a throttle on it and then mapped it to do the stupidest behavior imaginable. The fatbike hub motor has a lot of torque in the 20in wheel (2.3in not-really-fat tires) so pedal-assist level 5 hits hard if you're not pointed uphill, even level 1 is a bit severe. But if you're in level 0, the throttle does nothing, in level 1 it's limited, so you have to turn it to 5 to get full throttle command. Why have a throttle? #eBikes

Throttle should have full range from assist level 0 and never change its max power. I could see some utility of adjusting that on the fly but really just no.

People complain that pedal sensors are too on/off, but the problem is everyone uses a button pad to adjust them when that should really be like a grip shifter or something you can more easily do with mittens and be able to go all the way from 0-9 or whatever in one motion.

The lurching behavior can be controlled somewhat with ramp-up, but why are no e-bikes using an inclinometer, since weight and incline are what you really want to adjust the power to match?

Jef Poskanzer Jan 11

@enobacon Inclinometer is the germ of a great idea but you'd have to compensate for acceleration, which would look to the sensor the same as a change in incline. Seems like the engineering would be tricky but doable.

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐

@jef @enobacon The engineering is easy. The mfg knows the OEM bicycle weight, and the calculation to derive the actual gross vehicle weight from motor output power and actual acceleration is high school math, F=ma. 3D accelerometer chips are built-in to every smartphone, now, and are cheap.

@gcvsa @jef inertial frames of reference? I'm not sure we did that in high school.

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐Jan 12

@enobacon @jef It's a simple vector calculation, if you know the inclination relative to gravity normal, the power output of the motor, and the unloaded mass of the vehicle.

Well, I did go to a specialized high school with a math and science curriculum that is much more advanced that the typical American high school, but even if we say its undergraduate level math, it's easy for a microprocessor motor controller to accomplish. It just requires someone to see profit in actually implementing it.

Jef Poskanzer Jan 12

@gcvsa @enobacon No, you have to figure out the current mass. That will probably require a short observation period on each ride to get data on level ground.

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐Jan 12

@jef @enobacon It's easy to calculate the current mass, if you know the original mass, the inclination relatvie to gravity normal, and the actual power output of the motor.

F=ma. Force = mass times acceleration.

You simply compare the actual acceleration with the expected acceleration of the unloaded vehicle to get the actual mass. Automatic wind resistance compensation, too.

The results will be accurate enough for these purposes, as we aren't calculating an orbital insertion trajectory.

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐Jan 12

@jef @enobacon The 3D accelerometer chips inside cellphones do this fast enough to be able to be used as videogame controllers without perceptible lag.

Jef Poskanzer Jan 12

@gcvsa @enobacon Well, you keep missing my point, so I guess we're done here.

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐Jan 12

@jef @enobacon I don't believe I've missed your point, at all. You are free not to respond, that's your right, but I think I've done a pretty good job of explaining the actual physics and procedures involved.

@gcvsa @jef the actual incline is not able to be sensed without knowing mass and total torque at that instant. Maybe mass could be calibrated under constant speed on sufficient incline, or manual input would be close enough. Similarly, apparent wind speed and wheel speed could account for actual wind speed.

@gcvsa @jef high speedometer resolution could perhaps also account for acceleration vs slope

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐Jan 12

@enobacon @jef The actual incline could easily be determined by 3d acceleromter integrated circuits. The OEM mass is a known quantity, though I acknowledge that a certain amount of error would be introduced by accessory equipment, though this could be user calibrated within a certain range, or by dealer outside of that range. Actual torque and current would also be known to a pretty close margin. And since the goal would be to adjust acceleration, actual wind speed wouldn't matter.

Gemma ⭐️🔰🇺🇸 🇵🇭 🎐Jan 12

@enobacon @jef My 2020 iPhone SE2 has an inclinometer built into it.

@gcvsa @jef Maybe some sensor on a suspension fork would help you know if you're climbing a hill or doing a wheelie.