I designed different speed controllers for PMSM and BLDC motors yet, but I never tried to control the velocity of a brushed DC motor. So I gave it a try and didnt expect any difficulties, but it is harder than I tought.

So what I have done is pretty similar to a PMSM motor, I have a speed controller (1000kHz) and an underlying current controller (30kHz), which is actually a PI controller, that transforms the current into a duty cycle for the PWM signals.

Now I feel the motor cogging at very low speed pretty hard and I tought if I can compensate it with a feed forward term like for PMSM, but I could not achieve that, or at least only a bit. Is it possible, that the cogging at low speed of a brushed DC motor cannot be compensated, since I cannot commutate the stator myself like for a PMSM and so this bumps are switching of the poles?

  • \$\begingroup\$ First thing you should figure out is if the output of your controller is steady, or if it's jumping all over the place with the loop chasing the goal. Or run the motor open loop from a bench supply at the same speeds. Brushed motors designed for low speed operation do sometimes have unique mechanical designs to resist cogging, as opposed to those designed for power. \$\endgroup\$ – Chris Stratton Dec 5 '18 at 16:05
  • \$\begingroup\$ Well, the current control can hold a reference pretty good, the motors has actually pretty large cogging when turned by hand, like a BLDC motor mostly has, but it is a brushed motor. \$\endgroup\$ – HansPeterLoft Dec 5 '18 at 16:51
  • \$\begingroup\$ With any sort of synchronous machine (PMSM or BLDC), you more or less know the armature position, and can use that to overcome cogging. With a brushed DC motor, unless you put an encoder onto it and calibrate it to the cogging, you can't. It's simply a lack of information. \$\endgroup\$ – TimWescott Dec 5 '18 at 18:21
  • \$\begingroup\$ The usual thing to do in this situation, if it's an issue, is to choose a motor with lower cogging torque. There are motors built with skewed rotors that reduce cogging, and coreless motors that eliminate the static cogging (although the torque constant may still vary cyclically). \$\endgroup\$ – TimWescott Dec 5 '18 at 18:21

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.