1
\$\begingroup\$

In my experience, there are two ways to control motor speed; PWM and varying the voltage. I have found that there are at least two practical methods of varying voltage, some are less efficient (e.g. varying the gate voltage on a high side FET) and other methods seem to be more efficient (e.g. using a buck converter).

PWM seems to be the go-to approach for changing motor speed in many guides, but I noticed that it generates a 'whine' noise on some motors (I'm told due to magnetostriction), so I tend to avoid that approach.

Previously, I have used a buck converter to reduce the voltage to the motor. I'm aware that there is a stall current, and as long as you're above that, the motor seems to function without any obvious negative side effects. I guess you can also use a lower voltage power source such as a 12V battery on a 24V motor.

That brings me to the question: Are there any negative effects of running a DC motor at lower than the rated voltage? If so, does it depend on the type of DC motor?

Side note: There are off-the-shelf motor speed controllers available to purchase, but they often seem to have vague details; no datasheet or schematic, and vaguely mention PWM in the product description. I hope that these are simply controlled by a PWM input signal and have a smooth output (to prevent magnetostriction), but I suspect the output may be PWM, which I usually want to avoid.

\$\endgroup\$
6
  • \$\begingroup\$ PWM is effectively varying the voltage so as far as the motor is concerned there is no difference. Your average permanent magnet motor is thermally limited - if you put too much power into it, it will overheat. \$\endgroup\$
    – Kartman
    Aug 11, 2022 at 5:48
  • 2
    \$\begingroup\$ If the noise is bothersome you can raise the PWM frequency until it’s inaudible. I use ~29kHz in one of my motor drives: it’s this frequency that gets me the greatest bit depth without being audible. \$\endgroup\$
    – Bryan
    Aug 11, 2022 at 5:54
  • \$\begingroup\$ Of course, raise the frequency. Good thinking. \$\endgroup\$ Aug 11, 2022 at 6:03
  • 2
    \$\begingroup\$ @Bryan, "... without being audible" to us. A dog may disagree. I often wonder what discomfort we impose on them. \$\endgroup\$
    – Transistor
    Aug 11, 2022 at 6:07
  • \$\begingroup\$ In some cases, especially at much lower than rated voltage, I think a low frequency PWM may work better by providing a sort of "hammer drill" action where there may be a heavy load on the motor that may cause it to stall at low RPM. \$\endgroup\$
    – PStechPaul
    Aug 11, 2022 at 6:12

1 Answer 1

1
\$\begingroup\$
  1. In regard to the buck converter vs PWM - they are almost the same. The buck converter has PWM and some dedicated inductance built in, the bare PWM motor controller abuses the inductance of the motor coils for this purpose so it is cheaper. On the other hand, the motor coils are not always the best inductance for the purpose - e.g. they emit sound at low frequency (where they are efficient) and tend to develop core loss at high enough frequencies that make the sound inaudible.

Both of them also emit some electromagnetic "switching noise". The buck converter has most of it contained inside its package while the PWM is usually worse as it imposes square AC over the whole motor.

  1. In regard to running the motor at lower voltage - the only practical limitation is that most motors depend on their own rotation for their cooling. This is why running a motor at high torque (i.e. high current) and low RPM (i.e. low voltage) for an extended period (depending on the motor thermal mass and the cooling conditions) could make the motor hotter than comfortable. Of course, the fancy motor datasheet (if available) is your friend.

One can as well just oversize the motor and make it thermally safe from 0 to max rpm at the required max torque.

The above is generally true for DC (brushed) motors. AC motors have their own quirks, including the similar possibility to overheat them at lower voltage.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.