[kwkarth]

Quote:

Ok, what terms should I disambiguate?
 
se

"balanced" as it applies to headphone amplifiers.

 

[Steve Eddy]

Quote:

"balanced" as it applies to headphone amplifiers.

"Balanced" as it applies to headphone amplifiers, or "balanced" as it has been MISapplied to headphone amplifiers?

 
Actually, "balanced" with regard to their OUTPUT, I don't have a problem with per se. Since "balanced" has always been associated with a balance of impedances vis a vis common mode rejection, everything's fine on the output side of things. The output is balanced and the headphones are also balanced as well as differential so there would be common mode rejection.
 
My quibble is on the input side, where bridged amps are technically "balanced" from an impedance standpoint, but not differential so you don't get any common mode rejection.
 
Ah well. I'll shut up now.

 
se
 

 

[kevin gilmore]

all electrostatic transducers both speaker and heaphone with one exception
are 3 wire devices normally considered to be dual push pull. All of my
electrostatic amps both tube and solid state, and all stax amps are also
dual push pull, bridged amps if you want to look at it that way. They would
have no trouble driving any high impedance load with the bias disconnected.
In fact, many of the electrets are driven the same way.
 
The beveridge speaker is the only exception. Single bias, single diaphram,
single stator. Driven by an OTL tube amp with active tube current source.
Wonderful things if you have a huge room.
 
The crown power amps are all bridge amps made with only NPN transistors.
In non bridged mode, they sound absolutely horrible. In bridged mode they
sound wonderful. The difference is the almost exact canceling of all 2nd
and 3rd harmonic distortions.

 

[sbradley02]

Interesting information on the Crowns, as this is something we have been speculating in the thread as to why it could be that some designs sound better in bridged mode (other than the obvious increases in power, etc.). I have never listened to one myself, and mainly know them from the Stereophile review. This is the only detailed test measurements I have been able to find on a Crown:
http://www.stereophile.com/content/crown-macro-reference-power-amplifier-measurements
This shows higher THD bridged compared to non-bridged. Could you please send me a pointer to a reference that shows lower THD when bridged? Possibly the above test is not valid because it only shows a simple total measurement and not distortion spectra.
BTW while Stereophile didn't try the amp out bridged in listening, they did have some rather nice things to say about it depending on the reviewer.
 
Quote:

all electrostatic transducers both speaker and heaphone with one exception
are 3 wire devices normally considered to be dual push pull. All of my
electrostatic amps both tube and solid state, and all stax amps are also
dual push pull, bridged amps if you want to look at it that way. They would
have no trouble driving any high impedance load with the bias disconnected.
In fact, many of the electrets are driven the same way.
 
The beveridge speaker is the only exception. Single bias, single diaphram,
single stator. Driven by an OTL tube amp with active tube current source.
Wonderful things if you have a huge room.
 
The crown power amps are all bridge amps made with only NPN transistors.
In non bridged mode, they sound absolutely horrible. In bridged mode they
sound wonderful. The difference is the almost exact canceling of all 2nd
and 3rd harmonic distortions.

 

[SP Wild]

Quote:

"Balanced" as it applies to headphone amplifiers, or "balanced" as it has been MISapplied to headphone amplifiers?

 
Actually, "balanced" with regard to their OUTPUT, I don't have a problem with per se. Since "balanced" has always been associated with a balance of impedances vis a vis common mode rejection, everything's fine on the output side of things. The output is balanced and the headphones are also balanced as well as differential so there would be common mode rejection.
 
My quibble is on the input side, where bridged amps are technically "balanced" from an impedance standpoint, but not differential so you don't get any common mode rejection.
 
Ah well. I'll shut up now.

 
se
 

Ahhh, but Steve, it is not necessary for the signal to be differential to benefit from common mode noise rejection.  As long as the transmission lines have equal impedance and both have an equal impedance to ground - if these two conditions are met, even a single ended signal will benefit from common mode noise rejection.
 

 

[nullstring]

Great read guys. =)

 

[Steve Eddy]

Quote:

Ahhh, but Steve, it is not necessary for the signal to be differential to benefit from common mode noise rejection.  As long as the transmission lines have equal impedance and both have an equal impedance to ground - if these two conditions are met, even a single ended signal will benefit from common mode noise rejection.
 

What you say is absolutely true.
 
However I wasn't speaking of the signal. I was speaking of the input itself.
 
If you take two amplifiers and simply bridge them together to make a "balanced" amp, the resulting input is not differential. Any common-mode noise will simply be amplified and passed on to the output.
 
se
 
 
 
 

 

[nikongod]

 
Quote:

What you say is absolutely true.
 
However I wasn't speaking of the signal. I was speaking of the input itself.
 
If you take two amplifiers and simply bridge them together to make a "balanced" amp, the resulting input is not differential. Any common-mode noise will simply be amplified and passed on to the output.
 
se
 

 
As long as the noise stays common-mode through the amp(s) it cancels across the load. 

 

[jcx]

 
Quote:

 
 
As long as the noise stays common-mode through the amp(s) it cancels across the load. 

but eats into dynamic range - if you have 2 V common mode signal at the output you're going to clip 2 V earlier than if you had a true differential input amplifier that removed the common mode
 
due to the headphone drivers only responding to the differential signal you get some rejection of input signal common mode noise
although many amps are built with 1% gain resistors, limiting the rejection to 30-40 dB while most purpose engineered differential receivers greatly exceed 30-40 dB common mode rejection

 

[Steve Eddy]

Quote:

 
As long as the noise stays common-mode through the amp(s) it cancels across the load. 

 
Yeah. After it's been amplified.
 
"Common mode rejection? Hell, our amps don't reject common mode noise. They amplify it!"
 
There's a real selling point for ya.
 
se
 
 
 

 
 

 

[sbradley02]

And I believe (correct me if I am wrong) that cancellation at the load depends on having essentially perfectly matched amplifiers for the two phases.
You also get substantially more inherent THD and noise in a bridged configuration, at least with the one example I found, see my post of the Stereophile review of the Crown amplifier previously in this thread.
 
I would love to see similar specs on a headphone amplifier but haven't been able to find any. Anyone have something like this available? It would be extremely useful to see what is actually going on in a real world example.
 
Quote:

 
Yeah. After it's been amplified.
 
"Common mode rejection? Hell, our amps don't reject common mode noise. They amplify it!"
 
There's a real selling point for ya.
 
se
 
 
 

 
 

 

 

[nikongod]

I agree that the input noise eats dynamic range, and that there are better ways to put a balanced amp together than dropping 4 boards in a box like is commonly done. OTOH, there is still *something* to be gained by doing it half-assed.
 
What sucks about the whole thing is that doing a better job does not really require much more effort or expense than what you normally see. 

 

[sbradley02]

We definitely are in agreement on this point.
 
Thanks
 
Quote:

I agree that the input noise eats dynamic range, and that there are better ways to put a balanced amp together than dropping 4 boards in a box like is commonly done. OTOH, there is still *something* to be gained by doing it half-assed.
 
What sucks about the whole thing is that doing a better job does not really require much more effort or expense than what you normally see. 

 

 

[nikongod]

I made 4 points there. 
 
Edited:
I originally said I made 3 points, I actually made 4.

 

[sbradley02]

And I agree with all of them.
To restate (correct me if I have misstated), there are advantages to dropping 4 boards in a box, but the same or better results can be obtained at the same or lower cost with a similar level of effort by designing it in a more optimal fashion.
 
Quote:

I made 4 points there. 
 
Edited:
I originally said I made 3 points, I actually made 4.