[exsion]

I was wondering if anyone could help me understand the signal that is sent in both unbalanced and balanced cables.

As I understand it, an unbalanced cable uses a two wires to send separate mono signals while using a common ground through the shielding as a return.
But a balanced cable uses four wires to send the approriate left and right signals along with their inverted signals and can use a grounded shield.

Single-ended is termed for the ground return path.
But what is the return path for balanced?

Any help is much appreciated.

 

[Meetite]

The way balanced audio works is it has (usually) 1+2n pins/conductors (n being however many audio channels the device has). One is ground, and for each audio channel you have there are two additional pins; the main signal and an inverted signal. When the signal is sent between devices the main audio signal is sent on one conductor, while the polarity is switched (flipping the signal) and then sent down the other conductor. As EMI (electro-magnetic interference) affects the wires slightly, causing noise, it will affect both signals the same. When the signal gets to the end device the polarity of the second signal is then switched back and combined with the original signal, this cancels out nearly 100% of electrical noise. This is especially important when you have really long cable runs of 25 feet or more (think of a theater, cable needs to be routed all the way from the stage for a microphone and then all the way up to wherever the theater's A/V booth is without tons of noise.). Balanced most commonly uses XLR connectors.
On the other hand, unbalanced audio is cheaper and allows for more compact connectors and requires fewer conductors (1+n, n being the amount of audio channels that your device has). This is most common on headphones, mobile devices, and short-distance audio equipment. Due to not having the additional conductor it doesn't get the noise negation that balanced has, but over a short distance it isn't that big a deal. Unbalanced most commonly uses TRS connectors.

On the topic of single ended and return path, I don't know off the top of my head.

I may not be 100% right on everything, but this is my understanding of it.

Also take a look at this blog post, it should answer most your questions:
http://www.aviom.com/blog/balanced-vs-unbalanced/

 

[Speedskater]

An ungrounded interconnect system has:
a] Send
b] Return/Shield

An grounded interconnect system has:
a] Send
b] Return
c] Shield

There is no need for the word 'ground' to enter the conversation.

 

[pinnahertz]

The way balanced audio works is it has (usually) 1+2n pins/conductors (n being however many audio channels the device has). One is ground, and for each audio channel you have there are two additional pins; the main signal and an inverted signal. When the signal is sent between devices the main audio signal is sent on one conductor, while the polarity is switched (flipping the signal) and then sent down the other conductor. As EMI (electro-magnetic interference) affects the wires slightly, causing noise, it will affect both signals the same. When the signal gets to the end device the polarity of the second signal is then switched back and combined with the original signal, this cancels out nearly 100% of electrical noise. This is especially important when you have really long cable runs of 25 feet or more (think of a theater, cable needs to be routed all the way from the stage for a microphone and then all the way up to wherever the theater's A/V booth is without tons of noise.). Balanced most commonly uses XLR connectors.
On the other hand, unbalanced audio is cheaper and allows for more compact connectors and requires fewer conductors (1+n, n being the amount of audio channels that your device has). This is most common on headphones, mobile devices, and short-distance audio equipment. Due to not having the additional conductor it doesn't get the noise negation that balanced has, but over a short distance it isn't that big a deal. Unbalanced most commonly uses TRS connectors.

On the topic of single ended and return path, I don't know off the top of my head.

I may not be 100% right on everything, but this is my understanding of it.

Also take a look at this blog post, it should answer most your questions:
http://www.aviom.com/blog/balanced-vs-unbalanced/

Petty well said, overall. Just a couple of small details. A TRS connector would be used for unbalanced headphones only, otherwise in audio it could be used for a single balanced connection. The typical unbalanced consumer audio connector is an RCA plug/jack. In semi-pro/music systems, it's a TS .25" connector.

In an unbalanced interconnect the "return" is via the shield, and is ground referenced.

The 25' rule...well, there's a lot more to it than that. Longer runs sometimes work, short runs sometimes don't, it's more a function of what you're interconnecting and what "ground" actually is.

The thing to remember about a balanced interconnect is the wires are driven differentially, there's really no "main" and "inverted" signal. The driver might actually be a ground-referenced unbalanced amplifier, and another one with signal inverted, but you can also drive a balanced line with a transformer, the output of which is not ground referenced, but still diferential. The receive side must receive differential signals only and reject those in "common mode", equally on both signal wires. That can be a differential amplifier or a transformer.

One thing to be aware of, true balanced interconnects will require more driving and receiving circuitry to accomplish their noise immunity. Technically, a single-ended system could be lower noise and lower distortion, assuming no interfering signals.

 

[Whazzzup]

balanced amps topography can produce twice the voltage of SE amps and thus power cans that demand more watts a little more effectively. the advantage can go to demanding cans that have higher ohm requirements, power via balanced cables, xlr 4 pin as an example. I only know this from my system, no science engineer here.

 

[castleofargh]

balanced amps topography can produce twice the voltage of SE amps and thus power cans that demand more watts a little more effectively. the advantage can go to demanding cans that have higher ohm requirements, power via balanced cables, xlr 4 pin as an example. I only know this from my system, no science engineer here.

it's kind of a fake argument, if we wished to have twice the voltage in SE, the amp would be designed that way. it makes a point only for the same amp that has both SE and balanced output, and only for the rare case of using a headphone that just wouldn't be loud enough in SE but would be with +6dB or less... talk about luck. ^_^

 

[pinnahertz]

balanced amps topography can produce twice the voltage of SE amps and thus power cans that demand more watts a little more effectively.

No, that is a myth. A SE amp can be designed (and more easily) to produce any voltage and power required.

the advantage can go to demanding cans that have higher ohm requirements, power via balanced cables, xlr 4 pin as an example.

Headphone loads can be demanding in two different ways. One is in that they may require a higher voltage swing to produce the desired SPL (typically these are higher impedance models), the other is they present a lower impedance, but then require less voltage swing. Either can be driven effectively SE or balanced. A good amp will have some adjustment to compensate for headphones requiring less or more voltage.

I only know this from my system, no science engineer here.

 

[Cold Train]

No, that is a myth. A SE amp can be designed (and more easily) to produce any voltage and power required.
Headphone loads can be demanding in two different ways. One is in that they may require a higher voltage swing to produce the desired SPL (typically these are higher impedance models), the other is they present a lower impedance, but then require less voltage swing. Either can be driven effectively SE or balanced. A good amp will have some adjustment to compensate for headphones requiring less or more voltage.

Ok, explanations can be found all over the web to differenciate SE from Balanced, I don't think that the fact of knowing the number of poles and grounds can bring light to what newbies understand !

The questions are : What are the real sonic changes between SE and Balanced? What are the REAL pros and cons of choosing balanced instead of SE?

As I use both since years, I have my statements ready for these questions, but I will like to read from a guy who knows what he's talking about and compared to what I think I know !

 

[pinnahertz]

The questions are : What are the real sonic changes between SE and Balanced?

For interconnects, mostly, none. Balanced has the potential of lower external noise pickup, so if there's external noise, there could be a big improvement. Balanced drivers and receivers have higher internal circuit noise and distortion. Inaudibly possibly, but higher.

What are the REAL pros and cons of choosing balanced instead of SE?

For interconnects, the pros of balanced is only noise immunity. If you don't have noise in that situation with SE, then balanced won't improve anything. The cons are additional expense, possibly incompatibility with other gear, and frankly, some balanced inputs lack the expected high noise immunity attributed to balanced interconnects. Balanced interfaces are more than twice as complex in terms of circuitry. I've already mentioned noise and distortion.

For headphones, balanced connections add complexity and incompatibility issues. There is nothing to be gained sonically that could not also be gained with a well designed SE amp and a properly sized and configured headphone cable. However, going to a balanced headphone system often replaced cabling and amplification in one swoop, so it's perceived as somehow better, when in fact, the same results could be obtained SE.

As I use both since years, I have my statements ready for these questions, but I will like to read from a guy who knows what he's talking about and compared to what I think I know !

Ok, so I took the bait.

 

[castleofargh]

I doubt that the clear benefits of balanced designs can be seen on playback devices. but above all, a DAC or an amplifier can't possibly be fully defined by that alone. trying to generalize too much can only become a fools errand giving birth to ludicrous myths about the consistent superiority of one design.
I avoid balanced output for headphones because I did that once and recabling ended up being much more expensive and bothering than I had imagined. but I do run my powered speakers balanced so I guess I haven't turned racist yet, like I did with tubes, NOS DACs, Apple and Celine Dion.

 

[jagwap]

I posted this semi technical reply on this elsewhere on this forum:

There are several differences between single ended and differential systems:

Differential, if the same amplifier as single ended and ignoring losses, has twice the voltage and four times the power output into the same load

Because of the above, each amplifier in differential is loaded with half the impedance of single ended.

The signal in differential no longer touches the ground, which can make design easier. The relatively high currents from a headphone can disrupt the signal integrity of the around low current parts of the circuit if the layout isn't perfect.

The above point also means the left and right do not share a return path in differential, either in the cable or the active circuitry.

The output impedance, unless specially increased in single ended, is often double in differential.

Any even order distortions in single ended can be cancelled in differential, while odd order remain similar.

However as some distortions can be load dependant, as the load in differential is half, those distortions can rise.

Most high end DACs have differential outputs, and is has benefits staying differential all the way to the transducer.

Differential is more noisy, but not double due to some noise being cancelled, and others being uncorrelated.

Differential costs more.

I could go on, but it is a mixed bag of features, which are more involved than a bit of output impedance and fancy cable.

No, that is a myth. A SE amp can be designed (and more easily) to produce any voltage and power required.
Headphone loads can be demanding in two different ways. One is in that they may require a higher voltage swing to produce the desired SPL (typically these are higher impedance models), the other is they present a lower impedance, but then require less voltage swing. Either can be driven effectively SE or balanced. A good amp will have some adjustment to compensate for headphones requiring less or more voltage.

With speaker poser amps there is a voltage threshold where you end up going balanced due to secondary breakdown when using bipolar output transistors. Generally above 350W into 8ohms.

 

[pinnahertz]

With speaker poser amps there is a voltage threshold where you end up going balanced due to secondary breakdown when using bipolar output transistors. Generally above 350W into 8ohms.

Last I checked this was a headphone forum. Last I checked headphones didn't require 350W, and very few are 8 ohms.

 

[jagwap]

It is a typo. It should say speaker power amps. But I think you know that. You do don't you...

 

[pinnahertz]

It is a typo. It should say speaker power amps. But I think you know that. You do don't you...

Yes, I managed to surmount my extreme stupidity and decode that you were talking about speaker power amps. However, the rest of us are clearly not, which was my point. You did know that, didn't you?

BTW, this: "The signal in differential no longer touches the ground, " is partially incorrect. A floating (transformer output) differential signal is not ground-referenced, but the typical active differential output is most often just two single-ended outputs, ground referenced, out of phase.

 

[TYATYA]

Balance transmition use 3 wires : 1 hot (normal), 1 cold (invert), 1 ground.

Typical jack is XLR3 in which pin 1 is ground, 2 is normal, 3 is invert.

For output 2 channels XLR4 is use, pin 1 is L-, 2 is L+, 3 is R+ and 4 is R-
Base on that you can diy any interconnect cable balance or unbalance. Ex 6.3mm TRS to XLR4 or 2.5 ttrs to dual XLR3