atomicbob's SMPS Noise Nuke for HP amps with external SMPS

Discussion in 'Headphone Amplifier Measurements' started by atomicbob, Sep 26, 2017.

  1. Cspirou

    Cspirou They call me Sparky

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    4,956
    Dislikes Received:
    5
    Trophy Points:
    113
    Location:
    Northwest France
    Especially since its a 10 layer board.

    I think that's also why the average DIYer can create a pretty good Class AB amp and linear supply but need to resort to modules when it comes to SMPS and Class D.
     
    atomicbob likes this.
  2. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    3,004
    Dislikes Received:
    7
    Trophy Points:
    113
    Location:
    Laguna Hills, CA
    Home Page:
    @atomicbob Thank you kindly! The answer was about 9.8x as thorough as I was anticipating (because it came from you... for others, it's about 47x as thorough as I'd expect).

    @Cspirou The magic is not in the switching power supply, it's inside the chassis. I was reading about it earlier, mostly with regard to the ADI-2 DAC, and I figure the Pro has at least as much going on with the power inside as the DAC version.

    From the ADI-2 DAC manual (yes, I like manual pron):

     
    Thad E Ginathom likes this.
  3. highflyin9

    highflyin9 Almost "Made"

    Contributor
    Joined:
    Aug 3, 2017
    Likes Received:
    140
    Dislikes Received:
    0
    Trophy Points:
    33
    Location:
    DC/MD/VA Area
    Home Page:
    Very simple and interesting solution. Thinking it might be just what the doctor ordered for the Amp Camp with Meanwell PSU. Looks like Hammond makes a 1.5mH and 10mH inductor at 5A. I'm thinking I may try the 1.5mH (156B). There should be enough real-estate to mount the circuit right in the unit.

    Just to confirm to others as you can't see in the image, but the caps are in parallel, and adding a film bypass would be in parallel as well.
     
    Last edited: Jun 8, 2018
    rshuck likes this.
  4. bilboda

    bilboda Friend

    Friend
    Joined:
    Nov 20, 2016
    Likes Received:
    464
    Dislikes Received:
    3
    Trophy Points:
    63
    Location:
    Miama
    Just to confirm, caps wired in parallel, positive with choke lead to center pin tab of input jack. Negative to negative tab of the input jack, Output jack has the other choke lead to the center pin tab and a jumper wire between negative tabs on both jacks.
    Is this correct?
     
  5. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,309
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    The two 6800 uF capacitors are wired in parallel to achieve a total of 13600 uF.

    Here is a more pictorial type of diagram to complement the schematics in the original posts:

    atomicbobs noise nuke pseudo diagram.png
     
    Luckbad, quick, Torq and 3 others like this.
  6. bilboda

    bilboda Friend

    Friend
    Joined:
    Nov 20, 2016
    Likes Received:
    464
    Dislikes Received:
    3
    Trophy Points:
    63
    Location:
    Miama
    There , I think I got it. No shorts!! Making the out put cable now
    [​IMG]

    [​IMG]
    In action. Enjoying what I am listening to. File seems more dynamic I will have to put on some more familiar music for a good comparison.
     
    Last edited: Jul 15, 2018
  7. Wushuliu

    Wushuliu Acquaintance

    Joined:
    Jul 16, 2019
    Likes Received:
    54
    Dislikes Received:
    2
    Trophy Points:
    13
    Location:
    PNW
    Atomicbob, out of curiosity, what would happen if you wired up two sets in series (LCLC)?
     
  8. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,309
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    More filtering but diminishing returns.
     
  9. Wushuliu

    Wushuliu Acquaintance

    Joined:
    Jul 16, 2019
    Likes Received:
    54
    Dislikes Received:
    2
    Trophy Points:
    13
    Location:
    PNW
    Ah, ok. The thread inspired me to try LTSpice for the first time and I was able to model (I think) the circuit and yes, looks like not much improvement in series and only moderately so with increased inductance. Got some Hammonds on the way to try this out... Thanks!
     

Share This Page