Why is not all audio equipment using "balanced"?

[DonH56]

Done end-to-end, including between components, and assuming the components are fully differential internally, I have no doubt there's often a measurable positive difference in signal quality due to CMMR. Is it audible? Is the measurable difference worth the added cost in circuitry and cables? I don't know the answer to either one of those questions. I do know that a lot of people who really don't know what they're talking about think the terms "differential", "fully differential", "balanced", "quad balanced", "dual differential", etc., sound cool, sound special, and sound like they must make an audible difference in component design. Add in XLR connectors with their big metal construction and heavy weight, and you have an instant recipe for audiophile or HT nerd preference.

1. I have said many times that, for consumer audio, except in the rare case of a high-noise environment, the biggest benefit is the ability to eliminate a ground loop that is quite audible (and annoying). That is the main reason my rear subs use XLR runs from my AVP.
2. CMRR may be a consideration if routing long runs through walls near power lines or bundled with power cords behind a rack or console (typically for aesthetics).
3. You gain CMR for all the schemes, but it ranges from very low (typically 6 dB'ish for impedance matching) to pretty high (60 to 80 dB for well-designed differential links).
4. I do know what those terms mean, at least technically (from an engineer's point of view). So do many on this site.
5. Most of my experience with XLR and various flavors of "balanced" connections are for pro audio, including live sound and studio recording. Made a big difference there, and some home environments have similar issues, though rarely as long of runs as live sound systems.

Audiophile or not, being "cool" is not something people usually associate with me...

 

[blueone]

1. I have said many times that, for consumer audio, except in the rare case of a high-noise environment, the biggest benefit is the ability to eliminate a ground loop that is quite audible (and annoying). That is the main reason my rear subs use XLR runs from my AVP.
2. CMRR may be a consideration if routing long runs through walls near power lines or bundled with power cords behind a rack or console (typically for aesthetics).
3. You gain CMR for all the schemes, but it ranges from very low (typically 6 dB'ish for impedance matching) to pretty high (60 to 80 dB for well-designed differential links).
4. I do know what those terms mean, at least technically (from an engineer's point of view). So do many on this site.
5. Most of my experience with XLR and various flavors of "balanced" connections are for pro audio, including live sound and studio recording. Made a big difference there, and some home environments have similar issues, though rarely as long of runs as live sound systems.

Audiophile or not, being "cool" is not something people usually associate with me...

I wasn’t referring to you, Don. You obviously know what you‘re talking about. I was talking about people who don’t know, and think end-to-end balanced must be a great thing, just because.

 

[Kal Rubinson]

I always make my own custom lengths and extensions.

Build your own cables with high quality connectors.

I know, I know but I don't do much of that these days and, yes, the problems have been from cable restraint boots mis-matched to the cable jackets. When I repair them, my repaired ones seem durable. Still, I've never encountered such issues with standard XLR cables.

 

[escksu]

I know, I know but I don't do much of that these days and, yes, the problems have been from cable restraint boots mis-matched to the cable jackets. When I repair them, my repaired ones seem durable. Still, I've never encountered such issues with standard XLR cables.

Oh, this I can understand, the mis-match part. Frankly, me too, I have never encountered such issues with XLR.

 

[richard12511]

like they would "degrade soundstage" or some bullshit.

Also degrades the PRAT.

 

[MaxBuck]

Also degrades the PRAT.

Are you saying that the PRAT falls?

Tip your waitstaff. Try the veal.

 

[Snoopy]

Didn't expect that so many people would actually respond ! Awesome community.

I'm just surprised that unbalanced connections never got fully replaced by balanced. Is price really That much of a issue?

If I take a look at my audio equipment I got XLR out on my DAC/Streamer. XLR in on my amplifier and headphone amp.

And some "decent" balanced cables didn't cost me more than RCA cables of similar quality.

Same for the XLR headphone cable. Not that much more expensive than a good quality unbalanced cable.

 

[Yasuo]

For the same reason why not everybody is shooting raw.

 

[Atanasi]

Sound connectors are more conservative than computer connectors. In computer applications, balanced standards (USB, HDMI, Ethernet, DisplayPort) have all but replaced single-ended standards (RS232, PS/2, VGA).

 

[Kal Rubinson]

I'm just surprised that unbalanced connections never got fully replaced by balanced. Is price really That much of a issue?

Yes but not just for the cables which are simple passive hardware. To implement balanced inputs/outputs in the majority of audio electronics would require not only additional connectors at minor cost but, also, additional circuit elements to adapt/derive the signals for the generally unbalanced internal circuitry.

 

[Snoopy]

Yes but not just for the cables which are simple passive hardware. To implement balanced inputs/outputs in the majority of audio electronics would require not only additional connectors at minor cost but, also, additional circuit elements to adapt/derive the signals for the generally unbalanced internal circuitry.

So why not make the thing fully balanced and leave the unbalanced stuff out completely?

 

[Trell]

For the same reason why not everybody is shooting raw.

Not really as here we are talking about reproducing audio as distinct from creating it in the first place.

 

[Trell]

So why not make the thing fully balanced and leave the unbalanced stuff out completely?

Cost? Legacy? Not relevant in many contexts?

 

[DonH56]

So why not make the thing fully balanced and leave the unbalanced stuff out completely?

Manufacturers count cost in pennies and weigh the sales benefit of each added feature. Almost all conventional gain paths use op-amps and almost all of those have single-ended outputs. You'd have to buy or build differential gain stages throughout the component, double the feedback components for gain and filter stages, and double the number of traces (signal lines) used for each and every signal path, more than doubling the board area needed since there has to be a little space between each side of the single pair plus space to the next pair. Simple things like a volume control now need four (or five for shield) connections instead of two, and so forth. You'd also be adding power for the extra active circuitry. The overall added cost and size is (very) large for minimal benefit to a few consumers. It is simply not needed for the vast majority of audio gear in consumer use.

Much of my career was spent on RF systems with far greater bandwidth and dynamic range than a stereo system and yet differential interconnects were almost completely unknown. Doubling the power in a radar or communications system is just not feasible for most applications, and single-ended designs are more than adequate.

While I am a proponent of differential connections, they are overkill for most home systems, and I would not want to push a large price increase on the majority of folk who would gain no benefit. There are plenty of higher-end components featuring differential connections if I really need them.

IME/IMO - Don

 

[q3cpma]

I'd like a proof that this would increase cost significantly for the buyer, with economy of scale in the balance.

For the OP: inertia and "good enough being the worst enemy of better"; same reason MS Singularity/Plan 9 From Bell Labs didn't replace Windows/UNIX.

 

[egellings]

Long cables benefit from balanced connection for sure. Short interconnects used in a typical home stereo setting do not. Won't hurt, and won't help, either. I have a home system with short (3 to 6 foot) unbalanced cabling, and it is quiet. I'm not in a particularly electrically noisy area. Of course with most if not all non-pro home stereo equipment being equipped with RCA type connectors, there's not much of a choice anyway.

 

[DonH56]

I'd like a proof that this would increase cost significantly for the buyer, with economy of scale in the balance.

For the OP: inertia and "good enough being the worst enemy of better"; same reason MS Singularity/Plan 9 From Bell Labs didn't replace Windows/UNIX.

My experience is from IC design and PCB manufacturing. Look up the cost of op-amps and such used in a typical AVR and compare single-ended vs. differential signal chains, and the cost per square inch of a PCB (or the cost to add layers for the additional routing). Manufacturing is not really my area of expertise so I don't have cost basis on hand and no real incentive to try to generate a cost analysis when the question is just "will it add cost?" -- it will. Economy of scale is not the issue since it increases the cost basis of every unit, unless your argument is that higher per-unit cost will offset the manufacturing cost? So consumers will pay more? The industry does not usually head that direction... I have not dealt with the audio industry in years so maybe something has changed, but everything I have seen indicates otherwise (in favor of the status-quo, at least as far as adding balanced I/O to consumer products is concerned).

 

[KSTR]

A balanced output costs nothing (1 added resistor plus a 1/4" TRS instead of the RCA).

A basic balanced input with a single OpAmp also costs nothing (again swapping RCA for TRS) and that already provides the main benefit of a balanced connection, which is a separate conductor for the GND connection that can take power supply leakage currents etc without affecting the signal. The common-mode signal rejection is far less important but 20dB...40dB can be achieved which already goes a long way. The drawback is a slight noise penalty and not so high input impedance (10k'ish).
A two- or three-OpAmp "instrumentation" balanced input removes those drawbacks at little added cost.

 

[Cbdb2]

As soon as you put an XLR connector on a piece of gear you have to spec if its -10dbm and will work with consumer gear or if its what it should be, +4dbu and mates with pro gear. It should have a switch, but that will add even more expense. This will confuse most home stereo buyers.

Just the expense and space (the back of my AVR would have to be 3 times the size for all xlrs) of putting xlrs on an enclosure would put the gear in the next price bracket.

 

[preload]

I'd like a proof that this would increase cost significantly for the buyer, with economy of scale in the balance.

For starters you can read the post right above yours.