RCA to XLR trouble

[rcstevensonaz]

Maybe a misunderstanding/language problem here.

Agreed. Thanks for including the diagram since that is worth a thousand (misunderstood) words.

 

[TehMowat]

This wrong, as mentioned in above post, the working principle of balanced connections is to sense the signal directly at the source side with two connections (one hot and one cold wire), it doesn't matter if that signal is symmetrical or not and to which GND it is referenced. The difference operation to extract the playload signal is then performed in the receiver but the foundation of this is where and how the input signal is sensed.

How am I wrong? There is no way to turn an unbalanced signal into a balanced signal by simply buying a passive adaptor, such as an RCA to XLR. RCA is an UNBALANCED signal. Rewiring it to make it work with an XLR doesn't magically make the signal balanced. I stand by everything I said. You aren't fully or properly describing how balanced audio works. The audio signal is sent down the positive, and inverted signal is sent down the negative. At the receiving end, the negative is inverted again to match the positive, and any interference is then cancelled out. This is an incredibly simplistic explanation. There is no way to passively convert a balanced signal, you are just cutting out the negative with these passive adaptors. Or in the case of going from RCA to XLR, you aren't creating the balanced signal, you are just tying the ground to the negative. The "proper" way to turn balanced into unbalanced can vary depending on the receiving device. Some want the negative signal tied to the ground, some don't. My point stands, converting balanced to unbalanced with a simple passive adaptor isn't ideal. I would advise making different gear choices.

 

[sergeauckland]

How am I wrong? There is no way to turn an unbalanced signal into a balanced signal by simply buying a passive adaptor, such as an RCA to XLR. RCA is an UNBALANCED signal. Rewiring it to make it work with an XLR doesn't magically make the signal balanced. I stand by everything I said. You aren't fully or properly describing how balanced audio works. The audio signal is sent down the positive, and inverted signal is sent down the negative. At the receiving end, the negative is inverted again to match the positive, and any interference is then cancelled out. This is an incredibly simplistic explanation. There is no way to passively convert a balanced signal, you are just cutting out the negative with these passive adaptors. Or in the case of going from RCA to XLR, you aren't creating the balanced signal, you are just tying the ground to the negative. The "proper" way to turn balanced into unbalanced can vary depending on the receiving device. Some want the negative signal tied to the ground, some don't. My point stands, converting balanced to unbalanced with a simple passive adaptor isn't ideal. I would advise making different gear choices.

This, and the fact that there are several types of balanced outputs. Fully floating, centre-tapped (earth referenced) and impedance balanced, then of those, there are transformer and electronic versions, so to unbalance a balanced output, one needs to know firstly what type it is.

Going the other way, from unbalanced to balanced is much easier, as there's no chance of damaging the output or input. By the way, there's a distinction to be made between balanced and differential...

S.

 

[DVDdoug]

How am I wrong? There is no way to turn an unbalanced signal into a balanced signal by simply buying a passive adaptor, such as an RCA to XLR. RCA is an UNBALANCED signal. Rewiring it to make it work with an XLR doesn't magically make the signal balanced.

True, the SIGNAL isn't balanced. BUT the input IS differential and common mode noise is eliminated which gives you most of the benefits of a balanced connection.



...I solved a difficult noise problem by building a differential buffer with an RCA input which has no "ground" on the input. (Of course the minus input of the differential amplifier is connected to the ground of the connected device but the buffer isn't "ground-referenced" and the noise problem was solved.)

 

[KSTR]

How am I wrong? There is no way to turn an unbalanced signal into a balanced signal by simply buying a passive adaptor, such as an RCA to XLR. RCA is an UNBALANCED signal. Rewiring it to make it work with an XLR doesn't magically make the signal balanced.

The whole idea of a balanced input is exactly that it doesn't matter if the source signal is signal-balanced or not or whatever. The balanced input takes the difference of two signals and re-references that difference to the local ground (or whatever reference potential it may use). It doesn't matter what the individual signals look like as long as their difference is the payload signal.

 

[rcstevensonaz]

The whole idea of a balanced input is exactly that it doesn't matter if the source signal is signal-balanced or not or whatever. The balanced input takes the difference of two signals and re-references that difference to the local ground (or whatever reference potential it may use). It doesn't matter what the individual signals look like as long as their difference is the payload signal.

And moreover, that is the reason for using 2-wire STP (or quad) since the hot and cold wires are (nearly) impedance matched from the view of the balanced receiver. Balanced means both wire are same impedance; and the noise reduction (CMRR) is accomplished by the receiver — exactly as @KSTR notes — to eliminate any noise that the lines picked up.

In the case of the RCA to 2-wire conversion diagram (shown earlier by @KSTR), the cold wire’s impedance will not be an exact match. Which is why people say it will provide “most” of the CMRR benefit.

That is also why some diagrams you will see show adding a resistor on one of the wires to more closely match the impedance of the cold wire to the hot wire. However, a lot of people on this forum argue that is not worth the added complexity for the minimal incremental benefit.

 

[sergeauckland]

And moreover, that is the reason for using 2-wire STP (or quad) since the hot and cold wires are (nearly) impedance matched from the view of the balanced receiver. Balanced means both wire are same impedance; and the noise reduction (CMRR) is accomplished by the receiver — exactly as @KSTR notes — to eliminate any noise that the lines picked up.

In the case of the RCA to 2-wire conversion diagram (shown earlier by @KSTR), the cold wire’s impedance will not be an exact match. Which is why people say it will provide “most” of the CMRR benefit.

That is also why some diagrams you will see show adding a resistor on one of the wires to more closely match the impedance of the cold wire to the hot wire. However, a lot of people on this forum argue that is not worth the added complexity for the minimal benefit.

One further thing to add to the whole discussion about balanced ins and outs, is that these were done in studios, where cable runs were many tens or even hundreds of metres, especially with low level microphone circuits, or even between studios, where cable runs of many kilometres were normal. They were never intended to be used in home audio, where cable runs are of a few metres, and the electrical environment is generally benign. Unbalanced operation, if earths are sensibly arranged, is perfectly adequate.

However, consumer audio manufacturers started putting balanced ins and outs on their 'high-end' equipment, presumably to appeal to buyers, with a false narrative of 'professional' usage.

This has caused all the issues we get regularly on here about how to unbalance ins and outs.

My preferred choice would be to stick to unbalanced or balanced connections and not try to mix them, but have failed in my own system, which is a mixture, but I've largely got round it by using transformers and/or ground lift switching to avoid noise. Apropos ground lifting, it's my view that all equipment should be safety grounded, and use a ground lift on the audio, but of course that's another story, and will never happen...

S.

 

[KSTR]

In my world as an analog and audio systems designer all equipment should have simple balanced inputs and balanced cabling. Impedance-balanced outputs are sufficient. XLR is a far superior connector than RCA or 1/4" phone plug in every regard. 5-pin DIN connectors with balanced wiring would also do well. RCA was never intended to be a user-accessible connector, neither was the (tele-)phone connector.
We'd see zero threads dealing with "ground-loop" hum and buzz issues.

 

[rcstevensonaz]

… However, consumer audio manufacturers started putting balanced ins and outs on their 'high-end' equipment, presumably to appeal to buyers, with a false narrative of 'professional' usage.

I don’t disagree.

However, I think the real issue is that many on ASR realize that profession audio equipment is often superior to the “audiophile” equipment you are referencing.

So we wind up with XLR/TRS components because we buy RME, Genelec, Neumann, Ashly, etc.

Then, we face the issue that our knowledge base is still that of a consumer, not an audio professional. So we need to move up the learning curve (because of the equipment we chose); while then also weeding out fact from fiction… such as the fiction that balanced lines have a positive and inverted negative signal (even though that coincidentally is the case in most — but not all — professional audio gear.

 

[sergeauckland]

I don’t disagree.

However, I think the real issue is that many on ASR realize that profession audio equipment is often superior to the “audiophile” equipment you are referencing.

So we wind up with XLR/TRS components because we buy RME, Genelec, Neumann, Ashly, etc.

Then, we face the issue that our knowledge base is still that of a consumer, not an audio professional. So we need to move up the learning curve (because of the equipment we chose); while then also weeding out fact from fiction… such as the fiction that balanced lines have a positive and inverted negative signal (even though that coincidentally is the case in most — but not all — professional audio gear.

I think that 'proper' professional gear does have balanced and differential signals, it's the rise of the 'home-studio' and semi-pro gear where single-ended but balanced I/O exist. I personally regret the passing of transformer balanced I/O as that has so many advantages over all forms of electronic balancing with the exception of cost, (especially if LF distortion is to be low) and it's that last criterion, that of cost, that dominates everything these days.

S.

 

[rcstevensonaz]

Unbalanced operation, if earths are sensibly arranged, is perfectly adequate.

This is another truth I learned from ASR.

Except I wouldn’t use the word “adequate”, since that seems to be understating the reality for that vast majority of people.

But if you do have balanced input devices, you might as well use a proper RCA to XLR/TRS cable.

 

[somebodyelse]

Unbalanced operation, if earths are sensibly arranged, is perfectly adequate.

"if earths are sensibly arranged" is doing some heavy lifting here. Most of the time unbalanced interconnects are good enough for home use, but we still get a stream of questions about ground-related noise issues, most of which only exist because of unbalanced interconnects. Of course there are a few involving balanced interconnects where some component isn't following AES48 :-(

 

[Sokel]

but we still get a stream of questions about ground-related noise issues, most of which only exist because of unbalanced interconnects.

No, the main stream of questions and problems come from the mix of class I and class II gear by far with the second most often problem the USB connections and the CPU/GPU/fan/etc noise it carries, despite balanced or unbalanced gear.

The balanced OUTPUT gear is maybe third but even that comes mainly from mixed HC-music systems.

The above are my favorit poll, i gather them all over the forums (pro as well! Specially the USB problem)

 

[Speedskater]

But the RCA inputs on components often don't follow the "Pin 1 Problem) rule.
All cable shields should be attached to the metal chassis at their connector.

Components are sometimes powered by different AC circuits.

One problem with balanced interconnects is that XLR connectors are way big.

 

[G|force]

I prefer TRS to XLR because they can rotate and are bidirectional, but I use both out of necessity.

 

[Speedskater]

I have the opposite view of 1/4 inch TRS connectors.
The chassis connectors are troublesome.
The cable plugs are exposed to dirt and damage.

 

[LTig]

One problem with balanced interconnects is that XLR connectors are way big.

That makes them easy to solder, and is one reason why I love them.

 

[TehMowat]

True, the SIGNAL isn't balanced. BUT the input IS differential and common mode noise is eliminated which gives you most of the benefits of a balanced connection.



...I solved a difficult noise problem by building a differential buffer with an RCA input which has no "ground" on the input. (Of course the minus input of the differential amplifier is connected to the ground of the connected device but the buffer isn't "ground-referenced" and the noise problem was solved.)

This answer make no sense. Are you implying that by converting an RCA to an XLR you largely get the same benefit of a fully balanced connection all the way through? If you are, you are wrong. You having to create a "differential buffer" to fix a noise issue just illustrates the benefits of a proper balanced signal.

 

[TehMowat]

The whole idea of a balanced input is exactly that it doesn't matter if the source signal is signal-balanced or not or whatever. The balanced input takes the difference of two signals and re-references that difference to the local ground (or whatever reference potential it may use). It doesn't matter what the individual signals look like as long as their difference is the payload signal.

I'm not sure this is relevant to my point, or even accurate. The situation being discussed is taking an un-balanced signal and wiring it into an XLR. My entire point is that this isn't ideal, you get no benefit from it and may even have some issues. In my opinion, and that of anyone who works with and services these systems would say, the OP should just stick with all balanced or unbalanced gear.

 

[rcstevensonaz]

This answer make no sense. Are you implying that by converting an RCA to an XLR you largely get the same benefit of a fully balanced connection all the way through? If you are, you are wrong. You having to create a "differential buffer" to fix a noise issue just illustrates the benefits of a proper balanced signal.

You have conveniently changed the context of the discussion to make a straw-man's point. The conversation is about the wiring from an unbalanced output device to a balanced input device — NOT about end-to-end signal flow.

Any noise picked up in the system before that RCA connector will be transmitted without any CMRR benefit; only the noise picked up on the interconnect from the RCA output connector to the balanced input bridge.

And yes, we are saying that in this situation, you will get "most of the benefit" of that connector [i.e., unbalanced to balanced] being wired in the way described. The only reason you do not get all of the benefit is because while the hot and cold wires in this situation will have similar impedance, they are not exactly the same impedance (and thus the interconnect is not a true balanced connection).

Also, this conversation is not about whether to use all balanced, all unbalanced, or mixed. It is about how to wire an interconnect from an unbalanced output to a balanced input.