Why or why not use balanced connections

[Krunok]

Back to the D10 vs RME/Oppo/etc discussion, perhaps given recent posts in the D10 thread about USB input noise, this is not in fact the best choice for a competitor? Perhaps the D50? But that would bring us up to the $250 mark for our candidate for least-expensive transparent DAC.

IMHO, it is too early to judge the D10 by this isolated case, especially as even that one has not been properly analyised to see what is the cause of the problem.

 

[pos]

I want to know what source(s) are you using where you have four AES outputs? Is it a pro card? I am looking for an AES output for my PC but have yet to find one that is not a USB-to AES variety. I'm looking for something not USB. Preferrably just a dedicated two channel AES output than one with a massive breakout cable and relatively cheap.

Hello
I am using a miniDSP UDIO-8 8x8 192kHz soundcard.
I am afraid this is an USB soundcard, and it also comes with a breakout cables
The cool thing about it is that it is USB class compliant (ie works on linux), and as it uses independant ASRCs on each input (4 x AES3), making it practically plug and play for any situation. So far I have not encountered any stability issue.

If you are using windows I think there are some good options for PCI(E) soundcards from RME and Lynx.

 

[andreasmaaan]

Hello
I am using a miniDSP UDIO-8 8x8 192kHz soundcard.
I am afraid this is an USB soundcard, and it also comes with a breakout cables
The cool thing about it is that it is USB class compliant (ie works on linux), and as it uses independant ASRCs on each input (4 x AES3), making it practically plug and play for any situation.

If you are using windows I think there are some good options for PCI(E) soundcards from RME and Lynx.

In terms of USB interfaces, the RME Babyface for example has pretty clean measurements for a soundcard with a breakout cable and 12 channels, if I remember correctly.

 

[SIY]

If you're doing stuff yourself or customizing gear, I highly recommend the REAN Tiny XLR. They're inexpensive, not much bigger than an RCA, and have the mechanical integrity of XLR (not to mention balanced capability if you want it). Only major limitation is male-only for chassis connector.

 

[cjfrbw]

First rule of Fight Club: never, never discuss cables unless you are really, really in the mood for a fight.

 

[Fitzcaraldo215]

If they did, then something's wrong. A balanced output has at least one extra opamp or a transformer, so should measure no better, hopefully no worse, but certainly not better.

What I HATE about the RCA connector is that the signal makes contact before the screen, and the reverse on disconnection. Whoever designed that should be taken out and soundly whipped. For unbalanced use, a BNC is much preferred, although those are hard to connect and disconnect in tight spaces. The XLR is such a good connector, that I use it whenever I can, even for unbalanced circuits in stuff I make for my own use.

S

In true, symmetrically balanced, there is no extra op amp or transformer. But, agreed, that is becoming somewhat rarer, as it is expensive, due to the duplication of circuitry. But, my impression is that use of an op amp or transformer these days imposes negligible performance penalties yet still provides the advantages of noise rejection in balanced transmission.

I seem to recall some better DAC measurements by Amir via balanced outputs. There was lower noise/distortion in some cases, which makes sense because of higher output levels via balanced. But, there may be questions of how audible these improved measurements were. Also, I seem to recall in a few cases, Schiit maybe, better rejection of power supply noise products, possibly due to better grounding via balanced.

My understanding is that RCA engineers wanted the exposed signal lead so they could do circuit continuity tests without other instruments just by hearing the hum in touching the tip to ground. Bad, bad idea.

Agreed, XLR is a great connector.

 

[DonH56]

I have not read the thread (busy at work etc.) This has been discussed before, many times, here and elsewhere.

I prefer balanced but there are so many variations that it is hard to tell the "true" differential designs from the quasi-differential designs that offer varying degrees of balanced "goodness".

Pros for balanced: Positive connector capture is good. Common-mode noise rejection is good (which includes greater power supply noise rejection). Having the shield isolated from the signal path is good (helps with both noise immunity and makes it easy to break ground loops). Correctly implemented balanced offers 3 dB better SNR and lower distortion (all even-order products are suppressed).

Cons: It costs more at the component and cable level. Circuitry is more complex. Output impedance is often a little higher (not really a concern but for completeness). Fewer components have balanced connections and they tend to be at the high end so are pricey (consumer, not pro -- most pro equipment has some sort of balanced connections).

Most consumers do not need balanced but in some cases (like the runs to my rear subs) it can really be a life-saver.

FWIWFM - Don

 

[pos]

I have not read the thread (busy at work etc.) This has been discussed before, many times, here and elsewhere.

Same here, I lost track after a few posts and next thing I knew there were 5 pages of discussion...

Correctly implemented balanced offers 3 dB better SNR

A good illustration of this is the Topping DX7s :

• unbalanced: noise = 1.7uVRMS, signal = 2VRMS -> 121.1dB calculated S/N
• balanced: noise = 2.3uVRMS, signal = 4VRMS ->124.8dB calculated S/N

Balanced connection gives a 6dB boost to the signal (2 voltage gain) while giving only 3dB boost to noise (sqrt(2) voltage gain) because it is not correlated, resulting in a 3dB improvement in dynamic.
I hope the SMSL SU-8 delivers! (although I would have preferred AES3 inputs )

 

[Wombat]

Uh I use it at home, typically? And so does my friend. Also I forgot to mention so does my other friend! Oh and my fourth audio buddy also does, but less often.

4 out of ...? Silly sample.

 

[Wombat]

I can't believe this basic, well established/known topic has run to so many posts today. Some posters need to acquire some knowledge on the subject before spouting 'audiophile' opinion.

I thought this forum was established to get away from uninformed prejudices. As a learning tool this forum doesn't seem to work for some. Same old, same old, false beliefs being repeatedly posted and indignation when corrected.

 

[Krunok]

Most consumers do not need balanced but in some cases (like the runs to my rear subs) it can really be a life-saver.

I fully agree, although I think the best solution to connect active subs in home would be using Wifi. Unfortunately no standard yet exists but hopefully that will happen soon.

Look for example at the Denon Heos HomeCinema, you can send music to it via UPnP over WiFi from your Android phone and it sends signal to sub via WiFi. I have it for some time now and it works flawlessly and unlike most of the sounbars I heared this one plays music pretty well too.

https://usa.denon.com/us/heos-homecinema-soundbar-with-wireless-subwoofer

 

[sergeauckland]

In true, symmetrically balanced, there is no extra op amp or transformer. But, agreed, that is becoming somewhat rarer, as it is expensive, due to the duplication of circuitry. But, my impression is that use of an op amp or transformer these days imposes negligible performance penalties yet still provides the advantages of noise rejection in balanced transmission.

I seem to recall some better DAC measurements by Amir via balanced outputs. There was lower noise/distortion in some cases, which makes sense because of higher output levels via balanced. But, there may be questions of how audible these improved measurements were. Also, I seem to recall in a few cases, Schiit maybe, better rejection of power supply noise products, possibly due to better grounding via balanced.

My understanding is that RCA engineers wanted the exposed signal lead so they could do circuit continuity tests without other instruments just by hearing the hum in touching the tip to ground. Bad, bad idea.

Agreed, XLR is a great connector.

I'm confused by the first sentence highlighted. An unbalanced output will have a single op-amp (or other device like an emitter follower) as an output buffer. A balanced, differential output will have another opamp or other buffer for the second leg, interconnected in one of several different ways, or in some equipment a transformer, which can be fully floating or centre-tapped. I don't know of any way of creating a balanced differential output without a second device (or a transformer). An impedance balanced output can indeed avoid a second device, but then the second leg isn't driven, so whilst being balanced, it's not differential.

The result is, as mentioned above, a 6dB increase in output, which improves the S/N ratio by 3dB, but that's the only improvement I can think of, as frequency response is unaffected, and distortion should be unaffected. Consequently, the measured performance of a balanced output is only better in one, relatively minor, parameter, and no worse (but equally, no better) elsewhere. Ideally, all domestic equipment would be correctly and consistently wired such that ground loops are not created, but in the real world, where manufacturers all do things differently, then balanced, differential I/Os avoid problems. I use a mixture of balanced and unbalanced equipment at home, as I expect many of us do, and balanced is certainly a lot easier to deal with.

S.

 

[JBNY]

I use balanced xlr from my preamp to amp only because the amp is located about 15 feet from the preamp. But I have tested going from the source to the preamp with both rca and xlr I can not hear any difference whatsoever, so I use rca solely because I have better cables that I bought a while ago in RCA format. Sure XLR is the better solution but for home use, with a short run, no one will hear any difference at all so use what you like.

 

[Krunok]

I hope the SMSL SU-8 delivers! (although I would have preferred AES3 inputs )

From what I can tell you don't seem to understand that digital transfer doesn't really benefit from balanced connestion. Try to read carefully this article before advocating for AES3 without really understanding how things work:

https://benchmarkmedia.com/blogs/application_notes/balanced-vs-unbalanced-analog-interfaces

THE MYTH OF "BALANCED" AES DIGITAL INTERFACES
In the early days of digital audio, the Audio Engineering Society (AES) decided that it would be handy to use existing analog XLR cabling to carry digital audio. In my opinion, this was a really bad idea!

Balanced connections provide no advantage with digital audio signals. Digital signals provide substantial immunity to noise. The data format used to carry digital audio was designed so that it would have no spectral content at audio frequencies. This feature allows the use of a simple high-pass filter to remove AC line-frequency interference.

Digital pulses produce substantial energy at RF frequencies. The shape of these pulses is only preserved when the cable has a controlled impedance and is terminated with resistive loads that match the cable impedance. Existing analog XLR cables had a variety of impedances and these impedances were not well controlled. Analog cables proved to be completely unreliable for digital signals and special digital XLR cables had to be created. So much for using existing cables! We now have analog and digital audio cables that look nearly identical. Digital cables are acceptable for analog audio, but analog cables cannot be used for digital audio.

The AES initially gave us a standard (AES3) for digital audio using XLR connectors and special 110-Ohm cable. But, it has been shown that coaxial cables provide better signal integrity over long transmission distances. Coaxial cables support cable runs as long as 1000 m while the 110-Ohm cable is limited to about 100 m. The video industry created a separate standard for digital audio over 75-ohm coax. As a result, the AES3 standard was updated to include digital audio over coaxial cable.

Given a choice, we would strongly recommend using unbalanced coaxial digital connections instead of balanced XLR digital connections when making long cable runs (over about 50 m). Some professional products use BNC coaxial connectors instead of RCA connectors. Consumer and professional digital audio formats are designed to talk to each other. Simple adapters can be used to connect RCA and BNC connectors. Transformers are required when adapting between balanced and unbalanced digital audio connectors.

 

[pos]

This is going in circle, back to post #2...

 

[mindbomb]

Dacs will often not provide transformer isolation on the spdif inputs. So I feel that theoretically, there could be benefits from the CMRR in aes3 from a noisy source, reflected in jitter results.

 

[DonH56]

I didn't read the article but digital signals definitely can benefit from differential signals. We could argue if it matters for audio signal rates, but the vast majority of high-speed digital links are differential. Ethernet, PCIe, SAS/SATA, USB, etc. What was quoted seems to be more about the cable than the fundamental characteristics of differential signals. And for the record there are plenty of differential cables based on coaxial technology.

 

[SIY]

Dacs will often not provide transformer isolation on the spdif inputs. So I feel that theoretically, there could be benefits from the CMRR in aes3 from a noisy source, reflected in jitter results.

But no data on DAC analog outputs from anyone claiming that this is an actual problem. "Feel" and "theoretically" and "could" don't cut it any more so than unsupported assertion, data does.

 

[Thomas savage]

I’m struggling to see any value in this ‘debate ‘going on further so I’m closing the thread.

If anyone actually has the kind of data @SIY mentioned let me know and we can readdress the issue.

Cheers