It makes sense as long as the sensor motor is wound in a way that it’s BEMF voltage doesn’t exceed this reference voltage in the expected speed range to be measured.
Calibrating the ADC offset in this way is what SimpleFOC does also.
I wasn’t suggesting that. I was simply saying these sensors aren’t subject to the problems of the motor sensor, and work well at slow speeds. Their internal processing has bandwidth limitations which come into play at higher speeds, but at low speeds you can get their full resolution. The opposite of the motor sensor, which loses resolution as speed decreases.
Yes, but a center supply do also make sense when you use OP-amps or voltage reducers to alter the signal amplitude. And as I stated - you will always need at least two different amplifications before the AD-converter to adopt to the large speed range.
I think positions sensors will always have the problem, that you need to differentiate the signal to get a speed signal, and it cause increased noise in the speed signal. But with sufficient low noise and high bandwidth it may be the best solution - I don’t know yet. And yes, we agree on the disadvantages of the BLDC generator - I just got the hunch, that you can get a better speed signal this way.
I look forward to hearing the results if you decide to try it. I think it’s been mentioned at least once before on this forum…
In terms of differentiating position I guess maybe you could also devise a velocity sensor which directly outputs a signal proportional to velocity and isn’t subject to the differentiation error on the MCU?