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

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

  1. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,408
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    Is it sad that this thing has me more geeked than any other piece of audio gear right now?

    Would love a couple more internal shots to show where the wires are going.

    Thanks for putting this thing together. Really awesome piece of kit! Sell it to people at CA for $399 a pop.
     
  2. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    And Dino wins the Rock of Gibraltar.:D
     
  3. Chris45set

    Chris45set New

    Joined:
    Jul 25, 2017
    Likes Received:
    5
    Trophy Points:
    3
    Location:
    Omaha
    Thanks for this, it does look extremely interesting and (maybe) simple to implement.
    Could this be adapted for the 48V supply used by Horizon?
    And if so, what changes would need to be made?
    Chris
     
  4. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    The caps are rated to 63V and he’s using a heavy duty inductor. You can use it without any changes.
     
  5. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Look near the top of the first post for the words:
    It will work with any external SMPS or LPS between 1 and 48 VDC with current less than 2A.
     
  6. Thad E Ginathom

    Thad E Ginathom Friend

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    14,134
    Trophy Points:
    113
    Location:
    India
    Way not-sad! An audiophile gadget that is cheap and measurably useful? This is revolutionary!
     
  7. crazychile

    crazychile Eastern Iowa's Spiciest Pepper

    Pyrate BWC
    Joined:
    Jan 19, 2016
    Likes Received:
    2,512
    Trophy Points:
    93
    Location:
    Eastern Iowa
    I coulda sworn that was from Zappa’s “Thing Fish”......
     
  8. Armaegis

    Armaegis Friend

    Pyrate BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    7,464
    Trophy Points:
    113
    Location:
    Winnipeg
    So if it's so easy to make a noise nuke, why don't the manufacturer's do it?
    purely a cost/size/weight thing? are there downsides?
     
  9. k4rstar

    k4rstar Britney fan club president

    Pyrate BWC
    Joined:
    Jun 11, 2016
    Likes Received:
    6,941
    Trophy Points:
    113
    Location:
    Toronto, Canada
    Because going overboard to filter noise like this provides nice looking measurements but has detrimental subjective effects. You may experience a loss in dynamics, immediacy, and low-level information, even if you notice improvements in macro-detail retrieval and transient cleanliness. There is a sweet spot when it comes to filtering, and usually less is more.
     
  10. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,408
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    It's also not particularly inexpensive.
     
  11. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    This is basically for the regular off the shelf SMPS for charging laptops. The purpose built SMPS made for audio has all sorts of extra filtering.

    Plus you see the size of the inductor AtomicBob uses? It’s pretty big.
     
  12. 93EXCivic

    93EXCivic New

    Joined:
    Oct 10, 2017
    Likes Received:
    15
    Trophy Points:
    8
    Location:
    Alabama
    Very cool. I am definitely building one of these for my modified Bravo V2 hybrid.
     
  13. winders

    winders boomer

    Banned
    Joined:
    Feb 13, 2017
    Likes Received:
    1,596
    Trophy Points:
    113
    Location:
    San Martin, CA
    @atomicbob

    Have you tried grounding the negative line output of an SMPS to see what effect that might have? I ask because there has been some discussion about this on a another forum. John Swenson, who by all appearances seems to be pretty smart, has suggested doing this as a way to reduce SMPS noise injection. I don't have the equipment to test something like this so I am unsure as to the value of the suggestion.
     
  14. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    When making measurements, I am careful to have a single ground reference for the device under test and the measurement system. Ground loops often introduce additional noise. Each setup undergoes several different points of connection to define that single ground reference and the one that yields the best results is used for obtaining data. In this case, the negative side of the PS-III amplifier is connected internally to the shield of all RCA connectors, both input and output on the amp. A ground was established between the negative side of the PS-III and ground lug on the dScope to yield best performance.
     
  15. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,408
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    While I'd already owned a couple of Garage1217 amplifiers prior, @atomicbob turned me onto the Sunrise III and Horizon III with his measurements of the Sunrise here a couple of years ago. I found it to be excellent--far exceeding the cost--and superior even to G1217's flagship (Ember II).

    Later on, after many PMs (sorry Bob), I eventually bit the bullet and purchased a linear power supply (Jay's Audio) for my Sunrise III. It self-destructed after some damage to the power supply during shipping. A few months later, I bought an HP / Agilent / Keysight E3617A linear power supply for my Horizon III.

    The difference seemed subjectively significant (in audiophile terms, mind you. A few %). Cleaner, faster, more of everything good, less of everything bad. It wasn't until a few months ago that atomicbob posted objective measurements to back up just how much impact a good LPSU has on the Sunrise III. It's a pretty astonishing objective change that does seem to be reflected audibly.

    Enough preamble...

    After seeing the atomicbob Noise Nuke's objective measurements and simple design, I knew I had to build one and give it a try.

    My parts list was a little different from atomicbob's:
    • Choke: Hammond 155B 6mH 2A
    • Capacitors: Nichicon FW 6800uf 63V (KW were not in stock anywhere)
    • Enclosure: Mammoth 4S1590BBT w/ black texture powdercoat
    • DC Jacks: Switchcraft PC722A
    • Odds & Ends: A pair of 1" cable tie mounts, cable ties, heavy duty velcro, a bit of wire, solder, some padding and electrical tape for the choke to stay snug and comfortable.
    The current listening station:

    Lynx AES16e → Neutrik AES/BNC transformer → Schiit Bifrost Multibit → Garage1217 Project Sunrise III (w/ Noise Nuke) → Sennheser HD650 (w/ phase reversal trick)

    Some glamour shots of the atomicbob Noise Nuke paired with the Garage1217 Project Sunrise III, cleaning up the power from a Mean Well switching power supply (they mean well, but it's still an SMPS... har har...):

    luckbad_noisenuke_01_external.jpg luckbad_noisenuke_01_internal.jpg luckbad_noisenuke_01_sunrise.jpg

    I've been living with it for several days now, and I must say it's an incredible device. Subjectively, it does all of the same things as a top tier linear power supply.

    Upon initial listen, I wasn't sure I could hear a difference between it and the Keysight LPSU. I could definitely tell the difference between it and just the SMPS. Blacker background/lower noise floor seems to be the main objective improvement (I hear hiss at a lower volume without it--way above listening levels and only with sensitive headphones).

    After a few days, I think I can tell a difference between the Noise Nuke and my linear power supply. The differences are far fewer between those two than the switching power supply and either one.

    With either the atomicbob Noise Nuke or a nice linear power supply and a neutral/clean tube, the Garage1217 Project Sunrise III sounds a lot like a Cavalli Liquid Crimson Lite to me.

    If you have an amp or DAC that is DC powered and can use the atomicbob Noise Nuke, build yourself one. The DIY chops requirement is minimal and parts cost is under $60 shipped.

    Future experimentation will eventually follow. @johnjen suggested bypass capacitors. I'll probably pick up a few and pop them in. After using it for a while, I'll probably build a second, possibly with the in/out jacks on opposite sides of the enclosure.
     
    Last edited: Feb 14, 2019
  16. skank

    skank Friend

    Pyrate
    Joined:
    Sep 30, 2015
    Likes Received:
    69
    Trophy Points:
    18
    Location:
    Florida
    ^^^ also appeared in FZ's Thingfish (one of my favorite FZ albums). I also saw FZ in concert shortly after that album came out. Chad Wackerman is one of my favorite drummers. I think his playing on Thingfish is just incredible (although I could do without the electronic drum kit).

    Ummm, perhaps Bozio played on the album although Wackerman played the concert in 1984 that I attended.
     
    Last edited: Dec 27, 2017
  17. Luckbad

    Luckbad Traded in a unicorn for a Corolla

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    3,408
    Trophy Points:
    113
    Location:
    Holly Springs, NC
    @atomicbob have you tried the Noise Nuke with the RME Adi-2 Pro? I know they using a switching power supply, but I think it cleans up the noise internally. I imagine the nuke would still help some.
     
  18. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    RME ADI-2 Pro power supply noise is exceptionally low. To find any 60 Hz component it was necessary to use a high resolution FFT and average. Even then the 60 Hz noise measured -141 dBu which is approximately 69 nVrms (nanovolts). When used with HD6x0 300R headphones which have a sensitivity of 96 dB SPL / mW, the result is -42 dB SPL. Yes, way below the threshold of human hearing.

    Here is a very high resolution 262K FFT. No 60 Hz discernable. Only a miniscule amount of 3rd harmonic distortion mingled with the noise of the FFT measurement.
    Reference 0 dBFS = +19 dBu
    20180107-13 ADI-2 Pro Bal 40 Hz THD THD+N 262K FFT- ASIO -30 dBFS - HpW.PNG


    Here is a ludicrous mode resolution 8M FFT (Yes 8,388,608 FFT). 60 Hz is a barely visible dot above -160 dBFS.
    3rd harmonic distortion is more visible at -155 dBFS.
    Reference 0 dBFS = +19 dBu
    20180107-15 ADI-2 Pro Bal 40 Hz THD THD+N 8M FFT- ASIO -30 dBFS - HpW.PNG

    This would suggest that my Noise Nuke isn't going to make any improvements worth chasing on the ADI-2 Pro.

    Ludicrous mode resolution of 8M FFT with 4 averages takes 12.8 minutes to complete a single measurement. Not something one does frivolously.
     
    • Like Like x 5
    • Miss Information Miss Information x 1
    • List
  19. Cspirou

    Cspirou They call me Sparky

    Pyrate
    Joined:
    Sep 27, 2015
    Likes Received:
    8,200
    Trophy Points:
    113
    Location:
    Northwest France
    Daaaaaaamn. Im wondering how much replacement power supplies from RME cost.
     
  20. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    I suspect a good portion of the noise management is inside the ADI-2 Pro along with VERY CAREFUL ATTENTION to PCB layout. RME has Real Engineers(TM) as part of their design team.
     

Share This Page