1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Reducing servo motor jitter noise

  1. Jun 17, 2016 #1
    I have a digital servo that makes a lot of jittering noise. The motor itself works great, however when you set it to a certain position while applying a force it will make a jittering sound as it tries to find the optimal position, which is pretty common for digital servos to do. My question is how do I reduce this noise?

    One method I tried that seemed to work was lowering the voltage. This is a 4.8 - 6v servo, reducing the voltage to 2.8v greatly reduced the noise while still maintaining a decent amount of torque.

    I've heard using capacitors for noise filtering on the wires might work as well.
    At this point I'm considering using a high frequency noise absorbing foam, since I really just want to get rid of those annoying high pitched sounds. The low pitch I can live with.

    Thanks in advance
  2. jcsd
  3. Jun 17, 2016 #2


    User Avatar
    Gold Member

    You don't want to reduce the voltage as the higher current needed to operate it, while initially offering more torque, will cause motor damage. The higher the voltage you run, the smaller the motor you can use; maybe that could be a solution? Field weakening would allow it to run faster, which may get it past the chatter point, but I don't know if there are any steppers that allow for this {EDIT and it would lower your torque END EDIT}. If what you have will run for a sustained time with the chatter, foam may be the way to go. Many medical appliances with stepper motors rely on the foam.
  4. Jun 17, 2016 #3
    True, in that case I would have to put in a regulator to make sure the servo can't draw too many amps at the lower voltage.

    At this point I'm just trying to find the best material/method for absorbing high frequency sounds. Acoustic foam so far is my best guess.
  5. Jun 17, 2016 #4


    User Avatar
    Science Advisor
    Gold Member

    I don't know a whole lot about your exact situation, but I once had some real problems with line noise with some servos I was using. I had to use cabling longer than the manufacturer recommended and as a result, it picked up some additional jittering due to noise those longer cables picked up. I solved that with careful shielding and ferrites. That certainly may not be an issue you are dealing with, but it's something to consider.
    Last edited: Jun 17, 2016
  6. Jun 17, 2016 #5


    User Avatar
    Science Advisor
    Gold Member

    Ideally you should fiddle with the drive electronics to sort this problem out but a common quick fix is to introduce a small amount of friction damping . An O ring inserted between fixed and moving parts is often adequate .
  7. Jun 17, 2016 #6
    An O ring sounds like a pretty good idea. Thankfully there are things I can try! Also gladly high frequencies should be easier to cancel out.
  8. Jun 17, 2016 #7
    I would like to try acoustic foam, but the foam itself is also a good insulator which would prevent the motor from cooling. Any way around this?
  9. Jun 17, 2016 #8


    User Avatar
    Science Advisor

    Three suggestions:
    Use a controller that supplies some holding current when the position is reached.
    If the position sensor is not mounted on the motor shaft, reduce backlash in the drive system.
    Implement a deadband in the positioning algorithm.
  10. Jun 18, 2016 #9


    User Avatar
    Science Advisor

    Is the annoying high frequency a PWM switched power from the H-bridge to the servo?
    If so you need to identify the part that is moving and resonant at that frequency. Consider worn motor bearings, lubrication, motor mounts, driveline etc.
    Maybe change the PWM frequency away from the sympathetic region.
  11. Jun 18, 2016 #10


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Servo controllers work by calculating the difference between the demand position and the actual position. The difference (the error signal) is used to drive the servo in the required direction. Usually small error signals are ignored to stop the servo hunting or buzzing. This is called the dead band. You could try increasing the dead band if it is under your control. The down side is reduced position accuracy. Also it may not stop the buzzing if the servo has to provide a holding torque eg the servo is under a constant load that would pull it out of position.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted