Forget fiddling with derailleurs and gear levers mid-commute. Instead, Bafang wants to make urban e‑biking smoother than ever with its…
It even runs without any electrical power from the e-bike, since the shifting is performed inertially instead of using an input loop based on a speed sensor.
Oh man, I have some mixed feelings about this. The first few iterations of automatic transmissions for automobiles (circa 1970s I think?) also used mechanical means for shifting, somewhat similar to centrifugal governors. While they did remove the need to manually shift, my understanding is that they were also not exactly predictable. And since automatic transmissions wouldn’t be very useful in performance cars, they would have found use in mid-tier cars with mid-tier performance. So a mis-shift would leave the driver poorly configured if, say, they’re trying to accelerate to merge onto a highway.
For 250 W ebikes limited to 25 kph (15 MPH), this might be less of an issue, but I hope that Bafang can skip over that part of automotive history and that it Just Works™ for typical human pedaling cadences.
In that regard, the calculus should be simpler because the “dynamic range” of pedaling cadences is only some 50-200 RPM, or 4x between slowest and faster. Whereas for automobile engine speeds, the range can be 700-7000 RPM, or 10x between slowest and fastest. And even if Bafang sets the gear ratios consistently too high or too low, the sprocket ratios of the chain/belt can be used to tune that linearly. So bike manufacturers should have a number of parameters they can tweak, for either leisure- or commuter-oriented product lines.
I think all they mean by that is they don’t depend on external sensor data for this. The motor has internal sense of the speed, either through dedicated speed pickup, or through the rotor hall sensors. At least that’s my interpretation.
More recently, look at the shitshow CVTs were for a decade.
CVTs on automobiles were definitely released before all the issues were fully resolved, no question there. Though I think for fixed machinery and personal mobility, CVTs have substantial benefits, if they are properly engineered for the task. The major drawbacks being efficiency and packaging.
Supposing maybe Bafang develops on this present idea and makes an integrated CVT hub motor, an ideal implementation would have the torque sensor adjusting the ratios to keep the human at a constant effort.
Would it result in the pedals having nearly zero relationship to the actual effort of moving the bike, being essentially spin class but in public? Almost certainly. CVTs automobiles aren’t exactly considered fun to drive. But bikes for transportation is a key segment, and some people just want a bike for points A to B.
It’s too bad it doesn’t run the motor through the gears, that would be a huge improvement as even big chunky hub motors struggle to climb steep grades.
doesn’t the motor through
Doesn’t what?
Fixed lol
I could’ve sworn my now-discontinued 250w ebike (D4S) had a geared hub motor, but no idea now. Though no torque sensor and I always assumed it was probably only 2 gears.
Not that it’s very useful at 250w, you definitely still need a derailleur. And for me, the redundancy is a bigger point, because at this low of power it’s always just an assist.
If it does… my bike only does gearing for the motor itself, whereas the article says this gearing does apply to rider (mechanical) input.
EDIT: OK so this doesn’t apply gearing to the motor itself so is a different thing.
I wish the system also worked through the motor itself (i.e. could downshift the motor for more torque on hill climbs), but this is still great progress for the industry
As much as I admire Shimano and its amazing ability to make itself indispensable for what feels like a century and counting, hub gears just work better and this thing looks like the holy grail.
If it runs off inertia, doesn’t that mean that when the friction of the components changes due to lube getting dirty the shifting behaviors will change?
