As you've noticed, I'm experimenting in a couple of directions! I've got one motor that I want to drive faster, hence the increased voltage, and another that I want more control over, hence the lower PWM. As you say, lowering the PWM frequency also solves the problem I was seeing with driving these motors at higher voltage.
FWIW, I did try moving one of the motors to another SBrick (at default PWM) and I didn't see a repeat of the problem there.
Hi Paul,
Indeed, higher voltage means higher inrush current, which is what trips the OCP.
The knock-off motors probably have smaller filter capacitors (or just have none :) ), so the variability is due to the motors in that case.
There is a fair bit of variation in the value of the filter capacitors across motors (even if they are all manufactured by LEGO), and also in the current value at which the OCP kicks in across different motor driver ICs built into the SBricks. So in practice, the only thing you can do is to experiment.
Dropping the voltage can be a viable solution, an other one would be lowering the PWM frequency, as it is being discussed here: https://social.sbrick.com/forums/topic/8062/varying-pwm-frequency ;;)
I've got a couple of SBricks that I've been running at 9V and now at around 10.5V, and I've found that the two PF M motors regularly stall at the higher voltage, but work fine at 9V. I've read some of the past threads on driving M motors so I assume the higher voltage results in a higher current which is tripping the over current protection.
The same model has a third, fake Chinese M motor on a different SBrick which seems to work absolutely fine.
Unfortunately it's not very easy to swap stuff around, so I don't know if it's the different motors or the different SBricks that makes the difference. Is there much variation in this behaviour between individual items?
I've read about the capacitor in the M motor being the problem, but I'm not sure where this is the pictures I've seen only have a thermistor in them.
Any other suggestions, or should I just stick another diode in to drop the voltage a bit?