Just praising your work

[positron96]

Hello. I just wanted to chime in and thank you for the open-sourcing for the project. This is one of the best documented projects I've seen, and even better that it's on the topic I am interested in. Motor control logic is in particular very informative. I am making a BLE-based receiver for 1/87 scale trucks and need to solve very similar problems.

By the way, regarding motor control, have you considered PWM dithering for slow speed movement? Or is it already good enough with BEMF sensing and PID regulator?
Also, what frequency PWM are you using? I am noticing that higher-frequency PWM (>20KHz to make it silent) needs higher starting PWM to overcome stiction. And in general, I need higher PWM to start than to move due to the same reason. You mention that during movement, starting PWM is always added to PID output, does it prevent slower movement?

[gab-k]

Hi,

By the way, regarding motor control, have you considered PWM dithering for slow speed movement? Or is it already good enough with BEMF sensing and PID regulator?

the decoder is using 25kHz by default. I think dithering does not really have a benefit for my case as the PWM resolution is not the limitation but rather the ADC resolution. That doesnt mean it can't be improved. As far as i remember, currently not even a tenth of the full resolution of the PWM timer is used.

You mention that during movement, starting PWM is always added to PID output, does it prevent slower movement?

I wouldn't say it prevents slower movement as the PID output can be a negative value. So theoretically the starting PWM which is a little bit higher overcomes the friction and then the PID controller, sets the output a little bit lower when the motor is moving too fast at the starting PWM.

[positron96]

Thank you for response!

Considering ADC sampling, could you also clarify this part from the documentation:

Considering default settings (100us delay, 100 samples, ~2µs sampling time), the complete measurement run, including averaging, takes about 0.3ms to 0.35ms, which effectively reduces the maximum possible duty cycle to about 93% to 94%.

How often do you do ADC sampling? From these numbers, it seems, every 5ms (200Hz)?

[gab-k]

every 5ms (200Hz)?

Sounds about right, yes.