Balanced versus single ended

Then there is the argument that the AES48, as an application is a sonic limitation in itself, and is part and parcel and core to the idea that all AES48 applications have little noted differences in their sonic signatures.

You might try it. But keep in mind that if such were really the case the Golden Age of HiFi in the 1950s would never have happened. Basically what you seem to be suggesting is that the cables are too colored to be able to hear a difference, which is oxymoronic. Coloration is exactly what the balanced line system solves.

But this is a condition that is vertically nonexistent in high end audio.

 

The reality is that the quality of the interconnect usually trumps single ended versus balanced for the vast majority of actual systems. I have never heard a system where the single ended / balanced was a significant factor.

Just for the purposes of discussion, what if you used an AES48 compliant cable between compliant devices, but used such poor quality wire that both signal wires  in a cable negatively affected the sound in an equivalent fashion. 

@zlone @ghdprentice One of the benefits of balanced line operation is that the 'sound' of the interconnect is vastly reduced or eliminated (in particular if the low impedance aspect of balanced operation is supported). This benefit accrues even if the connection is only 6 inches long. If you've ever auditioned RCA cables and heard a difference, this is the 'sound' to which I refer. Imagine a technology where that problem of having to choose cables to match the system is eliminated. I've used cables that were 50 years old, with rubber insulation and tin-plated wire and no ill effects at all.

In the home the benefit is being able to place your amps next to the speakers, thus minimizing the coloration of the speaker cables because you can run a shorter speaker cable, plus never having angst about upgrading your interconnects. And of course phono cartridges are balanced sources- if there's anywhere in a system where the cable has to get it right, its the cable between the tonearm and preamp- if you can run that balanced you can get closer to real neutrality and the cable need not be expensive.

People often say how balanced is more expensive; in high end audio it isn't. Its often cheaper since you're not paying for expensive interconnects that become white elephants over time.

Interesting. Are you saying that both amplifiers and cables that are billed as balanced might not be to standard? 

Absolutely yes for real. I've been harping on this fact for years. I've not sorted out if high end audio manufacturers choose to ignore this issue or are simply ignorant. But its common in 'high end audio' for some reason.

With those who report here they have fully balanced audio systems input to output and yet they hear very distinct differences among balanced cable brand/models. Are there alternative or proprietary balanced audio circuits by these manufacturers or are they not true balanced circuits? 

Is this due to balance audio via use of  transformers or deviating  fully differential circuit approaches?

@charles1dad You can have a fully differential balanced preamp that does not support the balanced standard. In tube preamps this is common because coupling caps are used at the output. That coupling cap is usually part of a cathode follower circuit, and a cathode follower produces its output with respect to ground.

So that means you have two cathode followers, one for each phase, each referencing ground. The nice thing about this is that one of those can be employed as a single-ended output, so seeing RCAs and XLRs on such preamps is common. When two independent outputs are used like this, Common Mode Rejection Ratio is reduced in the system- it can never be truly balanced.

The balanced standard ignores ground, IOW the output of a device that supports AES48 does not generate either phase with respect to ground. Usually the way to do this is with an output transformer, whose secondary is connected to pin 2 and pin 3 of the XLR output and nothing else.

We patented an method of getting doing this without an output transformer, because as you know, we're the OTL guys. IOW you can support AES48 by using a Circlotron output, which is at the core of both of our patents in this area. It is possible to use semiconductors to support the standard.

One clue that the standard is supported is the 6dB issue- if the volume increases 6dB when running the preamp balanced as opposed to single-ended, that means it does not support the standard. The reason why is explained in the first paragraph above.

I think there is the idea that a transformer is a Bad Idea and so most high end audio manufacturers avoid using them (Backert Labs is an example of one that uses a transformer so they gear supports the standard). So as you point out, you get really variable results and people hear all kinds of differences in cables that simply shouldn't be there.

If the manufacturer supports AES48 they should say that- if not, ask them. If they don't know the answer, then they probably do not 😉

@holmz To be clear, black is the only color you get...

They both sounded good, but given the price difference, I would buy Ralph’s cables.

To be clear, 'our' cables are simply Mogami Neglex with Neutrik connectors.

 This is why XLR is especially good for long runs. The additional line for the single path is also responsible for the 6 dB gain

If your setup supports the balanced standard there will be no 6dB gain difference between balanced and single-ended.

I'd not considered this point before, but given that the signal is duplicated then reversed then transported then recombined, doesn't that mean that any coloration or other effect by the interconnect is simply negated?  So, you don't really need fancy balanced cables?

@jji666 

If your equipment supports the balanced standard then you don't need fancy cables. If it does not support the standard, you'll find yourself auditioning cables to find the one that sounds 'right'. Personally I find this a bit silly, since the balanced line system is a technology that eliminates cable colorations.

n the conversations with him, I mentioned that, in my mind, for balanced to work at it’s best, it should be designed in layout (from the active transmission end and at the active receive end) with an RF design and build mindset, where the field effects are a major consideration, down to the board mounting points and any local potential of the chassis and circuit boards in having any additional field effect interference. Just for the sake of the last little bits of attainable perfection in actual gear. Also, that these active aspects should be mirror imaged against one another and that includes a localized short run mirrored power supply for said mirrored circuit halves.

@teo_audio Honestly you don't have to do any of that! As a sort of proof, look at the studio gear used to make LPs and CDs- none of that involves any such practice. As you recall, we've been doing balanced line longer than anyone else in high end audio, and I've yet to see where any of what you suggest would be a thing. Look at it this way: a lot of studio gear employs transformers to execute the balanced operation and those transformers don't have nearly that kind of bandwidth nor do they need to.

Since the balanced line system is supposed to ignore ground, you really don't have an issue of 'any local potential of the chassis and circuit boards in having any additional field effect interference'. Such might become an issue if AES48 is ignored in the design.

I recall replacing Silver Surfer XLRs with WireWorld Platinum Eclipse 8 - HUGE difference in quality. Im not inclined to A/B them at this point, though, it might eat at me. However, I used a sample Nano 45 graphene contact enhancer on the WWs, and can’t find the rest, so I assume that would skew the A/B

FWIW you can have an entirely differential balanced preamp and it may not support the balanced standard. If you hear differences like you describe above it probably doesn't.

Back in the 1950s in the Golden Age of Stereo a great number of remarkable recordings were made. They still sound great today- and the better our systems get the better these recordings sound. They have the kind of neutrality that they do largely because the balanced line system is an exotic cable technology- but instead of spending the money on the cables, the money is spent on driving them and handling them correctly so colorations are eliminated. 

Cable manufacturers hate it when I talk about this! That's because if the standard is supported in your equipment, its very difficult for them to make a sale if you listen to what your ears are telling you. 

That said, I do find differences in mfg XLR cables (including length of identical cables), which could be the XLRs themselves, and/or quality of solder/ joints, as well as the wire, shielding, jacketing...

@tweak1 This is why I harp about AES48 and low impedance operation! If your gear is neither (for example any balanced ARC preamp and I'm not picking on them in particular) then you will hear cable differences. That's not supposed to happen. But too much 'high end audio' equipment does not support the balanced standard even though it really is otherwise balanced.

Hi @atmosphere, can you discuss what you just said in context with my Audionet Max amps and Pre G2 Preamp?  Thanks!

No- there is too little information on their website.

You absolutely convinced me that I did not need an expensive IC when using your amps and preamp. Has that changed now that I am using my AGD amps?

@mglik  I don't believe so.

@teo_audio 

Beyond that, it is a degeneration in sound quality, due to how it functions, how it works, how it is designed. (twisted pair in reflection and differential)

If one is not using it for long microphone lines, or hum reduction along said long line of very fine signals... then it has no place in home audio and is inferior to single ended.

and that’s a fact.

The supposed 6db lift comes at a price of loss of quality of micro signal aspects.

This is entirely false. The only reason its not used in the home is balanced line tends to be more expensive, especially if transformers are used to implement it. But these days transformers are not needed- for example our tube preamps have a patented direct-coupled output, and balanced outputs with semiconductors have been available for decades.

It appears you are laboring with a misconception. If the balanced line is properly implemented, there is no '6dB lift'. The balanced output is pin 2 and pin 3; the signal of pin 2 being generated with respect to pin 3 and vice versa, rather than ground. Imagine a simple output transformer secondary with one side tied to pin 2 and the other side to pin 3 and no ground connection, and you have the idea. Obviously if one side were to get grounded to drive an RCA input, the signal voltage would be the same.

If the signals are generated with respect to ground in a a balanced connection, then AES48 is not supported and you will get that '6dB lift'. 

I can't make sense of this 'twisted pair in reflection and differential' bit in your post.

BTW that 6dB lift you mention can only happen if there are two single-ended outputs that are out of phase with each other. That is not how a balanced line is implemented, and I can see problems with that if one side is slightly different gain from the other. That would mess with the input of an amplifier if it had a lower CMRR value, and it would increase the probability of ground loops, which the balanced system is supposed to lack!

The benefit of a properly set up balanced line is there even if the interconnection is only 6 inches; the length really has nothing to do with it. The rejection of noise including that of the cable itself is. IME once you hear a properly set up balanced line there is no going back to single ended.

Anyways, my question is: has anyone ever thought that single ended sounded better?given the 2 options. Im only referring to a truly balanced connection.

I ask, because a manufacturer who makes tube amps, recommends single over balanced connection. Is there something else involved in this decision, additional parts or labor complexity? Is the signal path extended?

The issue you are up against when asking this question is that there is a standard for balanced line operation that is rarely supported in high end audio.

The standard (AES48) reduces the ’sound’ of interconnect cables dramatically and thus also allows longer distances. It also prevents ground loops.

If the standard is not observed these benefits go away.

Balanced line however (especially if the equipment itself is also balanced) is inherently lower distortion, since even ordered harmonics get cancelled.

But otherwise when the standard is not observed, how its going to sound is anyone’s guess.

I am also a manufacturer, and in this case the one that introduced balanced line operation into home audio equipment back in the late 1980s. So now you also know a tube amp manufacturer that says that if you hear a balanced line system that is properly set up, there’s no going back to single-ended.

If you do it right, although there may be more parts, the actual signal path may well be less complex if you go balanced (inside the equipment itself). For example our amplifiers have only a single stage of gain.

One advantage of balanced operation is that the interconnect cables need not be expensive to sound right! Quite literally the balanced line system is a technology meant to minimize any artifacts associated with the interconnect cable. It was used in all the classic stereo recordings. Sometimes the microphone signal in those recordings had to travel up to 200 feet to arrive at the microphone preamp; if there were cable problems this simply would not have worked!