Let us expect we have a duty cycle of 50%. Is it possible to reach that duty cycle with different PWM Signals (waveform)? With different I mean variants in the number and lengths of the "active time". What are the expected impacts on noise and virbation resulting from these different PWM waveforms?

The context of this question is considering a 9V DC-Motor without any load where different virbation and noise should be generated. My motor is similiar to this one: https://www.alibaba.com/product-detail/25mm-battery-power-9v-12v-dc_60208528733.html?spm=a2700.7724857.normalList.41.3fe858e1928WCB

I hope the image explain what I mean: enter image description here

  • \$\begingroup\$ Well, it's true that one can change the frequency and amplitude of a PWM signal and maintain a 50% duty cycle. The effect on noise and vibration, I suspect, would be highly dependent on how you're generating it and what you're using it for. \$\endgroup\$ – Cristobol Polychronopolis Jul 26 '18 at 12:21
  • \$\begingroup\$ What is the context? I mean, what is the PWM being used for? And how are you generating it? Or have you not decided that yet? \$\endgroup\$ – piojo Jul 26 '18 at 12:21
  • \$\begingroup\$ Are you saying that the average duty-cycle over some period should still be 50%, but perhaps the individual duty-cycle for each cycle could vary? Like for example a mix of 45% and 55% averaging out to 50%? \$\endgroup\$ – brhans Jul 26 '18 at 12:26
  • \$\begingroup\$ @piojo PWM is used for controlling torque of a 9V DC-Motor. Vibration and noise is measured with sensor which are mounted on the motor. How to generate the PWM signal is not definded, probably a Pi or Andruino? \$\endgroup\$ – user3352632 Jul 26 '18 at 12:26
  • \$\begingroup\$ @brhans please take a look at the picture I uploaded \$\endgroup\$ – user3352632 Jul 26 '18 at 12:43

You can change the waveform and period, duty cycle is "the fraction of one period in which a signal or system is active".

As long as you have a period, you can determine a duty cycle.

One different type of PWM I have encountered is nonconsecutive PWM.

nonconsecutive pwm
(on a Siti DM163 LED controller)

| improve this answer | |
  • \$\begingroup\$ Thanks for your reply. Can you assert the impact on noise and virbation on a dc motor? \$\endgroup\$ – user3352632 Jul 26 '18 at 13:28
  • \$\begingroup\$ @user3352632 Considering the pwm will be "pseudo-random", the noise and vibration will be as well. \$\endgroup\$ – Jeroen3 Jul 26 '18 at 13:55
  • \$\begingroup\$ Ok, that means virbation and noise is in someway dependent from pwm signal / waveform. Do you have any source about this to read further ...? \$\endgroup\$ – user3352632 Jul 26 '18 at 14:54
  • \$\begingroup\$ I have heard an anecdotal report that one motor control manufacturer discovered that acoustical noise with a random or "pseudo-random" tone is less objectionable than a steady tone. \$\endgroup\$ – Charles Cowie Jul 26 '18 at 15:10

A given effective voltage or current can be provided using PWM waveforms with various carrier frequencies and patterns. Alterations in the waveform can avoid or mitigate problems with acoustical noise, vibration, torque pulsation, electromagnetic interference and other phenomena. Published literature describes various modulation strategies and their effects. Much of the literature describes research done with controllers for brushless DC motor and AC motors. That literature may be applicable to brush-commutator DC motors, but the extent of applicability may be difficult to determine. It may be difficult to find literature that is specific to very small and inexpensive brushed DC motors.

| improve this answer | |
  • \$\begingroup\$ Thanks for your reply, it is really helpful. Could you provide me a starting point (literature) for my specific issue (motor or nearly similiar ones)? \$\endgroup\$ – user3352632 Aug 14 '18 at 12:46
  • \$\begingroup\$ Most research papers in electrical engineering are found in the IEEE Xplore digital library, ieeexplore.ieee.org/Xplore/home.jsp. That is expensive to access unless you are an IEEE member or have privileges through a university library. There are other sources, but I am not familiar with them. \$\endgroup\$ – Charles Cowie Aug 14 '18 at 14:49

Your Answer

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

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