Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Audible noise from SMPS

  1. Sep 30, 2008 #1
    Hello, I'm a newbie for this forum, and I don't have much practical experience on SMPS too. I found this forum when I'm searching for my solution from internet.

    My Circuit:
    I'm using a simple buck converter:
    Vin: 24V
    Iout: 1.5A (dynamic load)

    The loading is running in PWM control, that means sometime it will draw around 1.5A and sometime it will draw around 0.05A, the PWM's frequency is about 250Hz.
    The converter is fine when I'm running it at fully ON (100% duty-cycle) and fully OFF (0% duty-cycle).
    But when I'm running it from 30% to 90% (duty-cycle), I can hear some audible noise (humming) clearly from the circuit.

    As I didn't hear the noise at fully OFF mode, so I think DCM is not the reason.

    I tried to remove the buck converter's inductor, and plug in a 3.3V at the converter output directly, then the noise disappeared, so I think the noise is from the converter circuit.

    I tried to replace the inductor with another brand and bigger size (higher current rating) inductor, the noise seems reduced slightly (the original inductor's current rating is about double of the converter's output current already), and both of them are shielded inductor with EPOXY potted already, so more EPOXY may not be helped.

    Anyone can share some experience knowledge with me on how to solve this issue?
    And any reply will be appreciated
    Last edited: Sep 30, 2008
  2. jcsd
  3. Sep 30, 2008 #2


    User Avatar
    Science Advisor
    Gold Member

    Just courious. Why do you need to get rid of the humm? Isn't it the 250Hz of the inverter you're hearing?
    BTW Welcome to PF.
  4. Sep 30, 2008 #3
    um... because it makes me feel unreliable on the circuit
    And I saw other similar circuit won't have this kind of humming

    and more information: I'm using 150kHz switching freq. and tried to change it to 250kHz, but the humming is still here.
  5. Oct 1, 2008 #4


    User Avatar
    Science Advisor
    Gold Member

    Well, in your first post you said Hz not kHz. That's why I was asking; since you can hear 150Hz.
  6. Oct 1, 2008 #5


    User Avatar

    Staff: Mentor

    The audible noise is most likely due to magnetostriction in the power magnetic element:


    You are not hearing the 150kHz directly, you are hearing effects of the PWM tracking for controlling the output voltage. You say you are using a shielded inductor with potting, so that's probably where the noise is coming from -- the interface between the two parts (inductor and shield).

    Try using an unshielded inductor to see if the noise goes away. If you need a shielded inductor for that part for other reasons, try using a toroidal inductor instead (it is self-shielding for generating external B-fields, not for being immune to external interfering B-fields).
  7. Oct 1, 2008 #6

    Thank you for your advice. I'm sorry that I didn't stay it clearly. What I mean is:
    1. The buck converter's switching frequency is 150kHz.
    2. The output's loading is driving by another switch and the switch is controlled by another PWM signal which is about 250Hz. That means the converter's circuit will be 1.5A, and then switch to 0.05A in 4ms, and then keep switching...


    Thank you for your opinion, I think I need sometime to study the "Magnetostriction". After I read wiki, I only can get that it can convert magnetic energy into kinetic energy.
    I tried to replace the inductor with WE-PD4-330,33uH, it still humming, and I can hear some higher frequency noise (weeeeeeeeeeeeee) at very low loading condition. I will check did I make anything wrong...

    (PS: The humming sounds like cry of cicadas)
    Last edited: Oct 2, 2008
  8. Oct 3, 2008 #7
    Today, I tried to add another output capacitor on my buck converter, so the output capacitance changed from 200uF to 2400uF, and the humming seems gone.
    Maybe 200uF is not enough to handle the fast switching from 1.5A to 0.05A...

    Of course, I can't use 2200uF capacitor on it, but I think it's 1 of the right ways for me to go, my next step is to find out the proper capacitance.

    Thank you for all the advices.
    Last edited: Oct 3, 2008
  9. Nov 11, 2008 #8
    Magnetostriction is (almost) never the source of noise in an inductor or transformer. Airgaps - also in the form of poorly assembled cores - are a common source, loose windings is another. And yes, I heard once a shielding vibrate.

    Capacitors sometimes make sounds. Quite unexpected. I heard it from a type-III ceramic one. But logical after all: it used BaTiO3 as a huge-K dielectric because it is ferromagnetic, and under polarisation it acts as a piezoelectric material and is commonly used in transducers. Type-I ceramic don't do that, plastic capacitors neither.

    The get a hum from a 150kHz switching frequency, you must have something flawed in your circuit: typically an oscillating regulation - consistent with the bigger capacitor stabilizing it. And by the way, 150kHz is already huge, I wouldn't go to 250kHz.

    The duty cycle can't be 100% nor 90% in an inductor buck circuit with 24V and 3.3V; and even with a transformer, designers stop at <90% duty cycle so that the transistor always switches.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook