What is balanced? What are balanced amps? Do they sound better? Discussion for noobs and boobs.

Discussion in 'Amplifiers and Combo (DAC/Amp) Units' started by Marvey, Dec 5, 2015.

  1. johnjen

    johnjen Friend

    Friend
    Joined:
    Sep 29, 2015
    Likes Received:
    213
    Dislikes Received:
    0
    Trophy Points:
    43
    Location:
    Well Grounded
    My 2¢
    Balanced refers to the signal being sent down the cable using 2 'hot' wires which are used in a differential mode that is independent from ground (ie. floating rather than than being 'tied' to the system ground).
    Whereas unbalanced is a single 'hot' signal carrying wire with direct reference to system ground (the shield).

    While a balanced cable can be made to run 'unbalanced' just by reconfiguring the wires themselves, an unbalanced configuration can't 'make' a balanced signal, not without added components (transformer, active circuit, etc).

    Running in a balanced mode means using 2 signal wires and 1 ground wire.
    Although using the ground wire is not always mandatory since it solely acts as path for the shield of the 2 differential signal wires, and it can be connected at only one end of the cable, this can be used to 'break' (or not create in the first place) a ground loop.

    Running unbalanced means 1 signal wire and 1 ground wire both of which are required.
    Running unbalanced also means the shield wire performs triple duty.
    1. Provides the shield a path to ground for any RF or noise.
    2. Provides the 'return' for the signal (hot) wire.
    3. Connects the up and down stream gear's 2 chassis grounds together, which can lead to a ground loop.

    Balanced was originally designed to effectively transmit signals over much longer distances than unbalanced (single ended) wiring schemes during the rise of the telegraph system.

    JJ
     

  2. aufmerksam

    aufmerksam Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    1,004
    Dislikes Received:
    0
    Trophy Points:
    93
    Location:
    Madison, WI
    See above edit to avoid confusion.
     
  3. Armaegis

    Armaegis Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    1,474
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Winnipeg
    Balanced interconnects also means each leg of the differential pair have equal impedance to the ground in order to maximize noise rejection at the receiving end (essentially by ensuring that whatever noise is picked up will be the same in both legs and thus cancel out). Whether or not one of those legs is sending a "zero" volt signal is inconsequential.

    Balanced amplification is a bit fuzzier. Many balanced amps are actually one of these:
    1) input a balanced signal, reduce it to single ended, amplify that
    2) as above but also derive a second inverted signal and amp that, then bridge the outputs
     
  4. 8man

    8man Rando

    Joined:
    Dec 7, 2015
    Likes Received:
    14
    Dislikes Received:
    0
    Trophy Points:
    8
    @Armaegis are those just main archetypes as a matter of choice, or do they differ in SQ ? Well I guess since sound is subjective, I'm referring to whether there is a better structure between the two. Thanks in advance !
     
  5. Armaegis

    Armaegis Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    1,474
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Winnipeg
    Sound quality is really just a crapshoot which has very little bearing on the general topology used.

    In a happy fuzzy idealized theoretical world of spherical cows*, a fully balanced not-just-bridged amplifier topology would be best as it would have the lowest rejection of voodoo and gremlins. Circlotron is probably the best known**. A bridged amp could sort of get there if everything were matched perfectly.

    *it's a thing, look it up

    ** can any of the guys here in the know name any other topologies other than circlotron which are inherently balanced?
     
  6. NekoAudio

    NekoAudio Acquaintance

    Joined:
    Apr 18, 2016
    Likes Received:
    38
    Dislikes Received:
    0
    Trophy Points:
    18
    Location:
    San Jose, California
    Home Page:
    spwath, maybe a picture would help.

    Look at the photo half-way down the page on http://hometheaterhifi.com/volume_5_3/balancedaudiovk500amplifier.html with labels 1,2,3.

    #2 is your ground wire. #1 and #3 are your signal wires, and they have signals opposite each other. Notice those little nicks in the signals on #1 and #3? Because the nicks happened in both signals, when you do #1 - #3, the nicks cancel out. Those nicks represent noise or other interference that was picked up.

    If you only have two wires, say #1 and #2, then those nicks can't be cancelled out. As far as you know, those nicks were supposed to be there in the original signal. And so you end up with the wrong music.
     
  7. 8man

    8man Rando

    Joined:
    Dec 7, 2015
    Likes Received:
    14
    Dislikes Received:
    0
    Trophy Points:
    8
    Cheers for the simplistic analogy, appreciate it when people help us understand this way !
     
  8. k1arg

    k1arg Acquaintance

    Joined:
    Oct 12, 2015
    Likes Received:
    29
    Dislikes Received:
    0
    Trophy Points:
    13
    Location:
    London, UK
    Does anyone know what the story is with the balanced versus single-ended frequency response graphs at HeadRoom? It seems that there are significant differences for HD 650 and HD 800. I'm assuming that this is just down to differences in measurement rig or procedure, but wonder if anyone has other information.
     
    Last edited: Aug 30, 2016
  9. Armaegis

    Armaegis Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    1,474
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Winnipeg
    Those look close enough to me that I'd chalk it up to product/measurement variance.
     
  10. zonto

    zonto Friend

    Friend
    Joined:
    Sep 30, 2015
    Likes Received:
    1,357
    Dislikes Received:
    5
    Trophy Points:
    93
    Location:
    Boston, MA
    Audio Myth - Balanced Headphone Amplifiers Are Better - John Siau, Benchmark Audio, September 15, 2016

    Abstract

     
  11. Armaegis

    Armaegis Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    1,474
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Winnipeg
    That article doesn't actually say very much though... and only talks in generalities that support their own point.
     
  12. Marvey

    Marvey Loves sex and records

    Staff Member Friend
    Joined:
    Sep 24, 2015
    Likes Received:
    16,839
    Dislikes Received:
    12
    Trophy Points:
    113
    Location:
    Calabasas, CA
    Balanced (bridged) requires x2 parts. A better argument would be that one could make a better single-ended amplifier for lower cost than balanced, and that re-terminating all of your headphones is a pain-in-the-ass (let's not even talk about new cables for people who don't know how to use a soldering iron).

    I'm generally not a big fan of balanced (bridged) for the above reasons, but that article seems rather self-serving.
     
    Psalmanazar, ultrabike and Sam Lord like this.
  13. PoochZag

    PoochZag The Shadow knows - Friend

    Friend
    Joined:
    Oct 7, 2015
    Likes Received:
    791
    Dislikes Received:
    0
    Trophy Points:
    93
    Location:
    Indiana USA
    Mjolnir 2's "basically required" balanced output caused me to buy a bunch of soldering equipment and learn the entire cable making process. I view this as a good thing because I enjoy it and have gotten semi-good at it. But it's a huge time sink (especially in the beginning) and I can't imagine paying for several cables from the typical third party cable manufactures, so I can definitely see the draw to sticking with single-ended for that reason alone.

    That said, why more headphone manufactures don't include an XLR cable with an adapter to 1/4" on high end headphones blows my mind, as it's the much more versatile 1 cable solution (Hifiman got this right way back with the HE-6). Heck Sennheiser could have done this instead of TWO cables for the 800S
     
    Last edited: Sep 16, 2016
    pluvious and JK47 like this.
  14. johnjen

    johnjen Friend

    Friend
    Joined:
    Sep 29, 2015
    Likes Received:
    213
    Dislikes Received:
    0
    Trophy Points:
    43
    Location:
    Well Grounded
    It strikes me as that write up is more like marketing material than a 'balanced' :) explanation of the strengths/weaknesses of SE vs Balanced.

    This quote alone (and there are others) sounds like marketing material to me…
    "In many cases, voltage-balanced headphone amplifiers will produce more noise and more distortion than single-ended amplifiers of an identical design."
    Um, a balanced circuit and a SE circuit simply can't be "of an identical design" unless you use the same circuit twice, which isn't identical.
    This statement (as well as others) makes many hidden assumptions which don't apply in all cases.
    And the writeup doesn't even bring up the grounding issue among other factors which are as relevant as those he does make.

    Just my 2¢

    JJ
     
    Sam Lord likes this.
  15. Madaboutaudio

    Madaboutaudio Friend

    Friend
    Joined:
    Sep 28, 2015
    Likes Received:
    542
    Dislikes Received:
    8
    Trophy Points:
    93
    Location:
    Singapore
  16. Armaegis

    Armaegis Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    1,474
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Winnipeg
    There's circlotron, super symmetry, pivot point... what others are there?
     
  17. johnjen

    johnjen Friend

    Friend
    Joined:
    Sep 29, 2015
    Likes Received:
    213
    Dislikes Received:
    0
    Trophy Points:
    43
    Location:
    Well Grounded
    How about the double down circuit topology?

    JJ :)
     
  18. Sam Lord

    Sam Lord Rando

    Banned
    Joined:
    Oct 11, 2015
    Likes Received:
    0
    Dislikes Received:
    2
    Trophy Points:
    1
    Thanks to Purrin (Marvey) and Johnjen for a fine discussion.

    Really? That much more hardware? I just had a heated discussion with a fellow at HeadFi over the general HP amp bal vs SE discussion, and as usual I wrote before thinking enough. I used to upgrade and redesign loudspeaker amps which were basically like Self's blameless amps: RCA inputs to diff input pair, global feedback with DC servo, class A driver stage, a typical AB output stage, excellent device matching everywhere, and good degeneration.

    So it was just a nice AB amp...fully balanced would nearly double that. So making my amp drive phones (as some folks drive HE6s) would require 4 lines minimum, no ground needed (except to shield the wires)....so how about that kind of circuit? That amp had a differential signal through the driver stage, but the outputs were just push-pull AB. Our speaker negatives went the the secondary return (like most SS amps), not a whole inverting amp. So really double the parts?

    I haven't looked at schematics of *good* SE headphone amps. The negative terminals are tied to ground, but is the driving circuitry so different? Perhaps no push-pull output stage? Capacitor-coupled? Anyway, then I thought about the problems with tying both channels' negative leads to chassis ground in SE HP amps. Johnjen and another feller here made the points I did about the drawbacks of using a chassis ground as a shared side of each channel.

    Continued below JJ's quote....

    ....Then I made my opponent's point, conceding that for low-current cans it shouldn't matter as much, but as current increases the cans act more like loudspeakers: that chassis ground's impedance is higher than an active circuit's, leading to greater crosstalk as well as distortion from underdamping.

    I had looked up Benchmark's site to hear their reasoning (cited earlier), and agree that they tell just half the story. They make NO standalone HP amps! All their HP amps reside in the same chassis with sources, either DACs, micpres, or ADCs. So changes in ground potential don't bother their circuit, the reference sits (probably) very close to the source reference. Now in that case, the common ground makes vastly more sense. And otherwise Benchmark might not be so snarky about balanced HP amps. Thoughts? Thanks folks.
     
  19. Armaegis

    Armaegis Friend

    Friend
    Joined:
    Sep 27, 2015
    Likes Received:
    1,474
    Dislikes Received:
    0
    Trophy Points:
    113
    Location:
    Winnipeg

Share This Page