[ahamric]

All I know about balanced and unbalanced is that generally balanced equipment uses XLR connectors and unbalanced uses everything else - miniplug, RCA, etc.

Can someone tell me why something (phones, amp, source, etc.) that is "balanced" is better than "unbalanced"?

Please help me cure my ignorance!

 

[ericj]

The terminology is confusing because outside of the realm of component to component interconnection, the term "balanced" is meaningless to the point of misleading.

What they're referring to is differential signalling.

Generally headphones are single-ended; you have a signal wire and a ground wire. The signal is AC, and generally swings both positive and negative of the ground reference.

"Balanced" is what they say when they mean that the signal is differential -- Rather than a signal referenced to ground, both wires carry half the signal, with what was formerly the ground wire now carrying a negative mirror of the positive signal.

So, you have twice the amplifier components, twice the gain, and twice the current capability, at a cost of only 2x what a single-ended setup costs.

This is actually how many - possibly even most speaker amps are designed. This configuration used to be called push-pull.

Since there isn't a single ground reference anymore, the three-conductor connector has to go. You could use a single four-pin connector (and in fact they used to use a single 5 pin connector for both single-ended and differential driving of headphones in europe), but in a stylish holdover from the studio signalling history of this type of connection, people generally use two full-size three-pin XLR connectors.

There is additional confusion with regard to the way sometimes the amp is really just dual mono, with two separate single-ended signals. This isn't nearly the same thing, even though you hook it up the same way.

In component-to-component interconnection, 'balanced' signalling has the advantage of 6db better noise rejection. Headphones can't take advantage of this.

 

[vcoheda]

read this. a good intro.

http://www.headphone.com/products/fa...ed-headphones/

 

[pfloding]

Push-pull is just a way to arrange output devices (normally transistors) so that one swithes off as the other swithes on -pushing and pulling the output voltage towards the plus and minus rail alternatingly. It's the standard configuration unless you use a single ended topology (unusual).

 

[zilch0md]

The terminology is confusing because outside of the realm of component to component interconnection, the term "balanced" is meaningless to the point of misleading.

What they're referring to is differential signalling.

Generally headphones are single-ended; you have a signal wire and a ground wire. The signal is AC, and generally swings both positive and negative of the ground reference.

"Balanced" is what they say when they mean that the signal is differential -- Rather than a signal referenced to ground, both wires carry half the signal, with what was formerly the ground wire now carrying a negative mirror of the positive signal.

So, you have twice the amplifier components, twice the gain, and twice the current capability, at a cost of only 2x what a single-ended setup costs.

This is actually how many - possibly even most speaker amps are designed. This configuration used to be called push-pull.

Since there isn't a single ground reference anymore, the three-conductor connector has to go. You could use a single four-pin connector (and in fact they used to use a single 5 pin connector for both single-ended and differential driving of headphones in europe), but in a stylish holdover from the studio signalling history of this type of connection, people generally use two full-size three-pin XLR connectors.

There is additional confusion with regard to the way sometimes the amp is really just dual mono, with two separate single-ended signals. This isn't nearly the same thing, even though you hook it up the same way.

In component-to-component interconnection, 'balanced' signalling has the advantage of 6db better noise rejection. Headphones can't take advantage of this.

 
Resurrecting this eight year old thread (instead of starting a new one)...
 
@erickj   I really like the way you've explained this - especially that last sentence!  I've long thought there's no way that headphones can enjoy any benefit of common mode noise rejection in such short cable lengths at the typically higher voltages than one sees with microphone cables, for example.
 
In my opinion, unless you've got a really long cable run between DAC and amp, where common mode noise rejection can be useful, the short interconnects between stacked components just don't warrant running balanced. And my only interest in having balanced cables for my headphones is when I'm using an amp that offers more power to a balanced 4-pin jack than to a TRS jack - as with many portable and desktop push-pull amps, and even then, only if the headphone is sufficiently inefficient to enjoy the extra power offered by the 4-pin connection.  
 
With the OPPO HA-1, for example, the 500 mW rms into 32-Ohms provided at the TRS jack is enough power to maximize the performance of the efficient OPPO PM-1 or PM-2, in terms of headroom related traits (bass control, dynamics, etc.).  I'm not alone in saying that I cannot hear any improvement, even with the power increase to 2000 mW rms, when using balanced cables with the PM-1 plugged into the HA-1's 4-Pin XLR jack. The Audeze LCD-2 rev.1, on the other hand, responds nicely to the additional power available at the 4-Pin jack, with better dynamics and bass control, because it's too inefficient to reach its best performance at the lower wattage of the TRS jack.
 
There are even some people out there who have incorrectly stated that the PM-1 doesn't "scale" at all. I say, they need to compare using the PM-1 with a Sansa Clip vs. the OPPO HA-1's TRS jack! The PM-1 "scales" just as well as any other planar magnetic. It simply maxes out its performance at a much lower wattage.
 
I am convinced 99% of balanced headphone enthusiasts are listening for and hearing differences in dynamics, bass control and extension, etc. - the traits most affected by differences in an amp's power output. They are not listening for, nor are they hearing, any differences one could attribute to factors such as common mode noise rejection or cross-channel impedance fluctuations on a shared ground, of which Jan Meier writes - whatever those allegedly audible differences are, given that he doesn't describe them. That's the thing - I greatly respect Jan Meier's designs - they are relatively complicated creations and I believe their sophistication goes unappreciated, with no feature less appreciated than his "Active Balance" solution for dealing with the alleged, unsavory, audible impacts of having a shared ground, while maintaining the convenience of a TRS connector instead of requiring balanced cables, but not even Jan Meier has explained what one would actually hear in the absence of his "Active Balance" technology.
 
And consider this: There are many very expensive headphone amps that do not offer balanced output and many very experienced buyers who are perfectly content with using the TRS jacks on these amps. And I have never - and I mean never - read any owner comments or professional reviews of those expensive unbalanced amps where the author has written anything like this:  "Gee, I can hear <specific negative traits> that are no doubt due to this otherwise awesome amplifier having only an unbalanced unbalanced output instead of balanced output.  What a shame the designer didn't give this amp balanced output to eliminate these <specific negative traits> I'm hearing."
 
Seriously... I've never seen anything akin to that when reading reviews of a TRS-only amp - whether it's crazy expensive or not. Oh, you can find plenty of reviews where people say they wish the amp had balanced output, but there are no reviews to be found where the author is identifying and describing specific negative traits (other than headroom deficiencies for a given inefficient headphone, perhaps) that he explicitly attributes to the amp's unbalanced output, stating that those traits would not be suffered with balanced output. Find me that review. It doesn't exist.
 
I used to believe that if I could negate the headroom differences between a TRS jack vs. the 4-Pin jack on the same amp, there might be a chance of hearing a difference, but I no longer believe that - because I've since tested this myself, extensively, to no avail, using a sufficiently efficient headphone. Using the OPPO PM-1, which enjoys no headroom-related audible improvements when moved from the lower-powered unbalanced connections of either the iBasso PB2 or the OPPO HA-1, to their respective higher-powered balanced connections... and using my SPL meter and a white noise file to volume match to 85 dB, then toggling back and forth between balanced and unbalanced connections, while inserting or removing a 4-Pin to TRS pigtail adapter as I adjust the volume to tape marks and switch between jacks...  I cannot hear any difference at all - precisely because I am using a headphone that is sufficiently efficient to enjoy no headroom-related improvements with the added power available at the balanced connector.  This requirement for making a valid comparison excludes my using the Audeze LCD-2 Rev.1, for example, to perform the A/B testing, as the LCD-2 Rev.1 can "scale" almost endlessly to however much more power someone is capable of giving it - certainly up to 1500 mW, but unlikely all the way to its maximum power handling capacity of 13,000 mW.  
 
And back to the subject of running "fully balanced":
 
Quoting this 6Moons interview of ALO's Ken Ball:
 

"To obtain a balanced working signal, a non-inverting amplifier buffers the input signal. The output of that amplifier is inverted with a unity-gain amp. The balanced signal is then taken from the outputs of both amps which are contained within the same IC. The worst-case mismatch between inverting and non-inverting outputs is 0.02dB. Going in single-ended [as input to the balanced Rx-Mk-3-B] is thus no worse than balanced. The real advantage of the RxMk3-b is its fully balanced amplification circuit and ability to drive all headphones in balanced mode."

 
Ken Ball is saying that nothing is accomplished feeding his balanced amp with a balanced output DAC vs. a single-ended output DAC. Ironically, he later succumbed to market demand by offering an amp with balanced inputs in addition to balanced outputs!  

  But I seriously doubt he had a change in his convictions, and I really have to wonder if he even values balanced output, for any reason other than satisfying the market craze.
 
I say, the only reason for going to a balanced output amp is for the dynamics and control offered with the additional power that normally accompanies balanced connector of an amp that also offers a less-powerful TRS connector - but only when driving headphones that cannot be driven sufficiently with less power.  And the only reason for going to a balanced output DAC is for noise reduction on long cable runs.
 
Still, running balanced can't hurt, except perhaps your wallet. 
 
(There, I got all that off my chest!)  

 
Mike
 
UPDATE:  Fixed some typos.