[JaZZ]

Quote:

Originally Posted by Bullseye /img/forum/go_quote.gif If you did a proper matched leveled comparison then you wouldn't be able to discern either of them. I had some data and information about that but I would need to search for it and don't have time right now to do so. Will come back with the explanation for it once I have time.

.
Quote:

Originally Posted by Bullseye /img/forum/go_quote.gif Hello hydrogen people, This is going to be my first post here. I come from Head-Fi and I am fed up with all the voodoo in that site. Getting some kind of objective info is too much to ask. ... Even so, if you know, Head-Fi is built upon myths and hype. On Head-Fi, loads of people say that electrostatic headphones are the holy grail. They say that it is the true "high-end" path. I, obviously as I am writing here, don't believe it.

Just for the fun...

 

[haloxt]

Quote:

Originally Posted by Oedipus Rex /img/forum/go_quote.gif So the question is simple: if I were to blind test balanced and non-balanced system with, say, HD-650, would I be able to tell them apart?

You should try it someday, I think you'll like it. The amount of control my balanced amp phoenix has over my headphones would alarm social workers if headphones had human rights.

 

[jcx]

bridged drive does change some operating conditions of the amplifiers - the drivers themselves can't respond to anything except the difference V across their terminals

for identical output stages bridged/balanced drive requires only 1/2 the signal V swing from each at the same current as required in SE drive

the different operating conditions of the output stages could result in different distortion generation

in bridged/balanced drive any equal even order distortion components will cancel - as will some parasitic power supply effects

internal to the bridged/balanced amp the headphone drive current doesn’t enter the signal ground path - perhaps measurable as improved channel crosstalk numbers - below any meaningful threshold - in Class B operation the signal gnd coupling could introduce some distortion as well

the above should seldom result in audible differences given the readily achievable low distortion levels of headphone amps not being obviously audible in themselves

any series impedance is doubled - possible audible significance but the "fair" comparison would be to trim the SE and balanced to the same total series Z

the only "headphone" change is the cabling - bridged/balanced requires 4 wires and separate connectors for each
TRS SE cables bring the 4 driver wires to the TRS plug where they share the gnd connection - some R/L crosstalk is introduced by the TRS jack/plug resistance in the common gnd connection
the numbers on this crosstalk suggest it can't be audibly significant - certainly not with open cans - and its still typically well below the interaural coupling of closed cans - only IEMs might give enough interaural sound isolation for the SE common gnd to be numerically significant but still not likely to be audible

my guess would be that the extra +6 dB of drive with balanced means you can listen louder without clipping - and even slightly louder is usually reported as "better sounding"

other than the headroom the only technical improvements from bridged/balanced operation are in audibly insignificant numbers - only heroically bad amps should sound better bridged/balanced

 

[JaZZ]

Quote:

Originally Posted by jcx /img/forum/go_quote.gif ...only heroically bad amps should sound better bridged/balanced.

Does that mean you haven't tested it yourself yet? Well, after all it's always good to know what something should be like according to a well thought-out theory...

.

 

[Steve Eddy]

What jcx said.

Er, mostly.

I do have a couple of quibbles.



Quote:

Originally Posted by jcx internal to the bridged/balanced amp the headphone drive current doesn’t enter the signal ground path...

I don't see how the headphone drive current inherently enters the signal ground path any more in an unbalanced amp than in a balanced amp.

Whether balanced or unbalanced, all of the headphone drive current enters and exits the ground reference node, the same node referenced by the input signal.

Quote:

the only "headphone" change is the cabling - bridged/balanced requires 4 wires and separate connectors for each

That's required for bridged, but not inherently required for balanced. If your outputs are transformer coupled, you can have balanced drive using just the TRS plug.

k

 

[royalcrown]

K: wouldn't you still need a TRS plug for each channel (versus one TRS for SE)? I don't see how you can have balanced drive without additional cabling.

 

[jcx]

Quote:

Originally Posted by Koyaan I. Sqatsi /img/forum/go_quote.gif What jcx said. ... I don't see how the headphone drive current inherently enters the signal ground path any more in an unbalanced amp than in a balanced amp. Whether balanced or unbalanced, all of the headphone drive current enters and exits the ground reference node, the same node referenced by the input signal. k

not an inherent flaw - just a common one with bad layouts: many of the chip amp layouts posted on the diyaudio forum have +/- ps rails on opposite edges of boards, power entry gnd is not often properly star gnded with RCA signal input/feedback and TRS output "gnd"
it would be nice if most designers had at least read up to the level of Doug Self's site/book

Quote:

That's required for bridged, but not inherently required for balanced. If your outputs are transformer coupled, you can have balanced drive using just the TRS plug.

you could do that - then youv'e got the common impedance crosstalk again

Meier built a "3-channel" amp that drives the active gnd with a derived signal to give ~ bridged drive for signal common to both channels



Quote:

Originally Posted by JaZZ /img/forum/go_quote.gif Does that mean you haven't tested it yourself yet? Well, after all it's always good to know what something should be like according to a well thought-out theory... .

why would anyone buy/build 2 bad amps just to "prove" the point? - the theory really works very well in the hands of experienced engineers

 

[JaZZ]

Quote:

Originally Posted by jcx /img/forum/go_quote.gif why would anyone buy/build 2 bad amps just to "prove" the point?

Why bad amps? That's your preconception which is still to be confirmed – not for «proof», just for own experience. Just give yourself a kick and audition some balanced amps – most if not all of them allow for single-ended mode as well, so you get the idea what effect balanced drive has or can have. By all means it won't hurt.

Quote:

...the theory really works very well in the hands of experienced engineers.

I'm not an engineer, and I don't base my audio evaluations on theories. And in most cases I have personal listening experience with the components I talk and write about. I think that's a reasonable attitude.
.

 

[Bullseye]

jcx and K explained it well enough.

Just for some literature you can read the following link, so you get another type of explanation.

Bridged and paralleled amplifiers - Wikipedia, the free encyclopedia

And yeah, we get back again to that louder = better... (+6dB drive is extremely necessary when using headphones)


(NOT)

 

[Steve Eddy]

Quote:

Originally Posted by royalcrown /img/forum/go_quote.gif K: wouldn't you still need a TRS plug for each channel (versus one TRS for SE)? I don't see how you can have balanced drive without additional cabling.

No, just a single TRS for both channels, as is normal.

Tying together the bottoms of two separate transformers (which would be necessary to connect to a stereo TRS), does nothing to effect balance.

k

 

[Steve Eddy]

Quote:

Originally Posted by jcx /img/forum/go_quote.gif not an inherent flaw - just a common one with bad layouts: many of the chip amp layouts posted on the diyaudio forum have +/- ps rails on opposite edges of boards, power entry gnd is not often properly star gnded with RCA signal input/feedback and TRS output "gnd" it would be nice if most designers had at least read up to the level of Doug Self's site/book

Right. It's not an inherent flaw, just a flaw in the layout. But you gave no hint of this in your original post which could lead some to erroneous conclusions.

Quote:

you could do that - then youv'e got the common impedance crosstalk again

Yes, but ultimately, what does it amount to? When done properly, all you're dealing with is the resistance of a couple of solder joints and the contact resistance of the TRS plug/jack combination.

How much crosstalk could you reasonably expect given the rather miniscule currents that headphones typically operate with?

Quote:

Meier built a "3-channel" amp that drives the active gnd with a derived signal to give ~ bridged drive for signal common to both channels

Cool. But perhaps just a bit of overkill?

k

 

[jcx]

I tend towards the opinion that many good amplifiers could be equalized/impedance padded to be indistinguishable in controlled blind testing - with the caveat that some amp/headphone combos simply don't have enough dynamic headroom for realistic music reproduction

and, similar to "k's" opinion above other considerations such as pleasing your own biases, demonstrating mastery of "audiophile arcana", pride of ownership could all be excused as "conceptual art" - which people do willingly pay for

 

[cafe zeenuts]

Quote:

Originally Posted by Oedipus Rex /img/forum/go_quote.gif Hi, I continue my questions of trying to figure out what audiophile concepts are the real deal and what not so. So the question is simple: if I were to blind test balanced and non-balanced system with, say, HD-650, would I be able to tell them apart?

The ability to tell the difference depends on how familiar you’re with the given setup. So, how well do you know the sound signature of singled end HD650? What about balanced HD650 sound signature, with the same source and cable as the single ended setup only difference being the amp? Without being familiar with the characteristic of each setup, the blind test will be meaningless.

 

[b0dhi]

Quote:

Originally Posted by JaZZ /img/forum/go_quote.gif And show me where I stated that you were referring to HeadRoom graphs! And show me which graphs you actually meant! .

My post was the only one that could conceivably have anything to do with what you said. Either that, or you were rebutting the air. Anyway I'm seeing a pattern here.

Moving on: one of the graphs is this one: http://www.head-fi.org/forums/4205745-post33.html. You can see the rest of the thread for further measurement details.

Can't track down the others at the moment (non-headfi).

 

[glitch39]

this is not a scientific explanation - but subjective and repeatable.

on my setup, I am able to switch between SE and balanced while keeping the same cans, cables, IC's, source and amp. Even when I volume match between the two, there is a DEFINITE change in soundstage width and sense of space/separation, in favor of balanced.

I use the same tracks for testing. I also used the "virtual barbershop" track for testing. I can clearly hear the change in channel separation between SE and balanced. no amount of volume matching will fix that. See my sig for the equipment used.