Monitor hookup

[boingoman]

A balanced audio connection has two wires, one of inverted polarity to the other. (For instance, in an XLR connector, pin 2 carries the signal with normal polarity, and pin 3 carries an upside-down version of the same signal.) However, an XLR plug also carries a third connection - pin 1 is used as an earth to shield the other two. The received signal is the difference between the two signal lines. This signal recombination can be implemented with a differential amplifier where the negated signal is tied to the negative terminal of the operational amplifier. A balun may also be used instead of an active differential amplifier device. Much of the noise induced in the cable is induced equally in both signal lines, so this noise can be easily rejected - the noise received in the second, negated line is applied against the first, upright signal, and cancelled out when the two signals are added together.

The separate shield of a balanced audio connection also yields a noise rejection advantage over a typical two-conductor arrangement such as used on domestic hi-fi where the shield is actually one of the two signal wires and is not really a shield at all, but relies on its low, but in practice not zero, impedance to signal ground. Any noise currents induced into a balanced audio shield will not therefore be directly modulated onto the signal, whereas in a two-conductor system they will be. This also prevents ground loops.

All that makes sense to may, and always has. Whatever it means to you, that's wonderful too.

Note that balancing does not require an out of polarity signal to work, and that it is the gear, not the wire, that has the polarity inversion.

 

[Reggie]

and that it is the gear, not the wire, that has the polarity inversion.

Exactly; this is what our friend RAK doesn't understand. Unfortunately Wikapedia isn't the most reliable source, and there appears to be some miswording:

A balanced audio connection has two wires, one of inverted polarity to the other. (For instance, in an XLR connector, pin 2 carries the signal with normal polarity, and pin 3 carries an upside-down version of the same signal.)

This would leave you to believe that the wires are actually inverting the polarity, which is not the case.

This is not proving your case:

I said "if a TRS cable carries a mono signal"

I still don't think this is correct if I catch your meaning. I'll try to finish the thought: If a TRS cable carries a mono signal, then there are two common ways it will do so-- coming from a balanced source, the +polarity will travel through the wire connected to the tip and the -polarity signal will travel through the wire connected to the ring. If the ring is connected to ground (such as when interfacing with unbalanced gear) at either source or destination, then the cable does not carry a balanced signal and all benefit is lost.
This is the second way a TRS cable will typically carry a mono signal; with an unbalanced signal going through the wire connected to the tip; and the wire connected to the ring simply being connected to ground (sleeve) at some point.

And this is just talking in circles as far as I can tell:

Clearly, if the cable carries this signal, then it is capable of carrying the signal, because it is carrying the signal.

Now I'm not trying to be belligerent and I'm not saying I expect you to know how this all works (unless you are the guy that engineers the mics), but I think it is a good thing for even home recordists to have an understanding of the way these things work.

 

[RAK]

I actually do understand how it works, and I agree with everything you're saying, and I even understand what you're saying. Look, I've studied this stuff, I always want to continue to learn (that's why I'm part of this community) And I was working full time professionally as an engineer in all areas of the industry before I got my day job, so I'm just going on what I've studied and how others I've worked with talk. And while I may only have a minor in electrical engineering (and I'm trying to say that means anything). I don't really appreciate you telling me what you expect me to understand or not. I certainly don't know anything about your background and would never assume anything or comment on it.

I think all that Wikipedia meant was that Pin 2 is hot (+) and Pin 3 is cold (-). That's audio wiring 101 right?

The talking in circles was a joke. My point was, if something is carrying a signal, then by definition it is capable of carrying the signal, because how could it do something it wasn't capable of?

I understand what you're getting at. But I think to the real world, an XLR cable will still be a balanced cable. A TRS cable will still be balanced. A TS cable will be an unbalanced cable that is prone to hum/noise because it's not balanced.

I think if you hold an XLR cable up in the air not connected to anything, and ask, is this balanced or unbalanced, not too many people would say "neither, until you plug it into a balanced or unbalanced source" I might be wrong about that though.

I apologize for using Wikipedia agian, but they had a nice little chart about TRS cables:

Unbalanced Balanced Stereo
Tip Signal Positive/"Hot" Left channel
Ring Not present Negative/"Cold" Right channel
Sleeve Ground/Return Ground Ground

Now this certianly shows that a TRS cable can carry an unbalanced or balanced signal from an unbalanced or balanced source. I've always agreed with you on that, and knew that before we came here. I realize now I may have given the impression otherwise. And of course many jacks are now wired to accept both TS and TRS plugs.
I guess my point is, I will always call a TRS cable a balanced cable. You got to call it something.

 

[Treeline]

WHOA, NELLIE!


boingoman has it right. Wikipedia is sort of right, but poorly written. Let's go back to the beginning here.



1. A cable is a cable. Different designs, differing qualities. A TRS cable does NOT a balanced signal make.


2. The term "balanced" refers to the nature of the signal being tranmitted through the cable. Three conductor cables, such as TRS and XLR cables, facilitate the transmission of balanced signals by virtue of their design. They do not "create" a balanced signal and if encountering an "unbalanced" signal, will faithfully transmit it as an unbalanced signal.

Everybody with me so far?

So: you will not have a balanced signal if using TS cabling; TRS or its equivalent is necessary. But the cable does not make the signal.



3. What is a balanced signal? Glad you asked.

A balanced signal is a signal split at its source (usually at or near the output jack) into two hot signals with respect to common neutral: there's the three thingies.

Again at the source, the polarity of *one* hot signal is reversed.

At the receiving (input) jack at the other end of the cable - again, not in the cable but in the "balanced" input jack, the polarity of *one* hot signal is again reversed. (I expect the standard is to reverse the same all the time but I don't know that for sure).

So that's pretty neat, but what does it accomplish? The answer is : plenty.

A cable is an antenna. Pretty good one in fact, and the longer the cable is, the better it is at picking up static, noise, hum and Radio Moscow. Unbalanced cables rely on shielding only - and although it helps a lttle, there's still a real problem with long cable runs. Not so much with one or two foot runs, or patch cords, as short cable is not a great antenna.

All cables pick up electromagnetic signal. Balanced cables pick it up just like any other. But here are the two really neat things that happen next:

Really Neat Thing #1:

The hot signals are transmitted on the two hot conductors in reverse polarity and are summed to one polarity at the input jack. The static and interference signals are transmitted on the two hot conductors in the same polarity and are summed to reverse polarity at the input jack.

Really Neat Thing #2:

1 + (-1) = 0.​

A signal encountering it's "polar opposite" will be cancelled out. If the signals are exactly polar opposites, they will cancel out exactly, leaving: complete silence. And the only part of the signal that encounters this particular experience at the input is the....static interference.

And that's how balanced cables reduce / eliminate noise in the signal chain, why they are imperative for mic cables, and why balanced outputs on mixers are so much preferred to unbalanced outputs.

Are unbalanced cables useless in your shop? Of course not. Patch cords are frequently unbalanced with no discernable degradation. But long signal runs will be noisier than if they incorporated balanced signals.

 

[RAK]

See this link for all I was ever talking about refering to. Call it the "court of public opinion if you will" This was my point about there being balanced cables, because (I thought) it was a generally accepted term.

http://www.hosatech.com/hosa/products/audio.html

To me, balanced always meant that third wire that acted as the noise/hum cancellation. Granted a "stereo" cable acts in a different way even though the cable itself is the same

But I completely agree/understand your explanations.

I'm sure I need to stop now.

 

[Reggie]

RIght, I think we are past the point that cables get classified by balanced and unbalanced with respect to how many conductors are inside and what connectors are used. And perhaps you even understand what the three of us are trying to say; but I look at this post as public information and I am just doing my best to clarify some fuzzy-worded logic as it appears in some of your posts for the greater good of whoever else is reading this and trying to interpret it. This is how myths and misperceptions become rampant on webboards. People find a little nugget of knowledge, take it out of context, and make incorrect assumptions. So yes, maybe we are arguing semantics.

And again with the slightly misleading posts:

To me, balanced always meant that third wire that acted as the noise/hum cancellation.

THat third wire doesn't do jack other than conduct electricity. Only when the + and - signals get recombined in a piece of gear that has a balanced input does the noise cancellation take place.

So anyway, we are past the "figuring out cable classification" stage; and we are now trying to get at the meat of the underlying concepts of what actually goes through the cables and why and whynot.

 

[RAK]

RIght, I think we are past the point that cables get classified by balanced and unbalanced with respect to how many conductors are inside and what connectors are used.

That's all I was ever trying to get across, ever.

THat third wire doesn't do jack other than conduct electricity. Only when the + and - signals get recombined in a piece of gear that has a balanced input does the noise cancellation take place.

Yes, I'm sorry, I should have said, it's the cold (-) signal in relationship to the hot (+) signal. I took that for granted as being understood and I shouldn't have, since without the third wire you don't get that cancellation. I was trying not to get into the nitty gritty. To me, that noise cancellation is the main thing to take away in terms of real world use.

So anyway, we are past the "figuring out cable classification" stage; and we are now trying to get at the meat of the underlying concepts of what actually goes through the cables and why and whynot.

And that's great, and I agree with everything you're saying. Again, all I was ever trying to get across was what was said up top. It seemed like I was getting a lot of resistance (say about 12 Ohms) to that. You said we're past the point, but that was the only point I was ever trying to make.

 

[boingoman]

And again with the slightly misleading posts:


THat third wire doesn't do jack other than conduct electricity. Only when the + and - signals get recombined in a piece of gear that has a balanced input does the noise cancellation take place.

Well, really, the third wire is connected to the shield, and conducts noise to ground. It also provides the ground connection between the two chasses. So it does a little more than "jack".

The signal is not necessarily split, that's not a requirement for balancing. Some schemes use it, some don't. What is required is that both legs have equal impedance to ground. That is the definition of balanced. The equal impedance on both legs lets them be equally susceptible to noise, which is important for rejection at the inputs.

Also, the signals aren't "recombined". They are fed to the inputs of a differential amplifer, which senses differences in voltage across at it's inputs.

People get this "recombined" idea from the fact that addition and subtraction are used to describe what happens. The values can be added and subtracted to see what happens, but the signals themselves don't recombine. The input senses the voltage difference, and outputs a corresponding signal. Common-mode noise, which presents zero total voltage difference after inversion, never gets past the inputs. Signal, which presents a voltage difference, gets passed along.

 

[Reggie]

Well, really, the third wire is connected to the shield, and conducts noise to ground. It also provides the ground connection between the two chasses. So it does a little more than "jack".

I was referring to the (-) polarity wire as the third wire, because I believe that was what RAK was referring to as the third wire about 4 posts back. The wire that separates a balanced cable from an unbalanced cable.

 

[boingoman]

I was referring to the (-) polarity wire as the third wire, because I believe that was what RAK was referring to as the third wire about 4 posts back. The wire that separates a balanced cable from an unbalanced cable.

Gotcha................

 

[Reggie]

The signal is not necessarily split, that's not a requirement for balancing. Some schemes use it, some don't. What is required is that both legs have equal impedance to ground. That is the definition of balanced. The equal impedance on both legs lets them be equally susceptible to noise, which is important for rejection at the inputs.

Also, the signals aren't "recombined". They are fed to the inputs of a differential amplifer, which senses differences in voltage across at it's inputs.

People get this "recombined" idea from the fact that addition and subtraction are used to describe what happens. The values can be added and subtracted to see what happens, but the signals themselves don't recombine. The input senses the voltage difference, and outputs a corresponding signal. Common-mode noise, which presents zero total voltage difference after inversion, never gets past the inputs. Signal, which presents a voltage difference, gets passed along.

Interesting info. I honestly never knew there is balanced gear that doesn't do the reverse polarity thing. We are getting right past the edge of my limited knowledge with some of this stuff.

So what about transformer balanced inputs. What are the different ways those are used to handle a balanced signal?
Thx

 

[boingoman]

Interesting info. I honestly never knew there is balanced gear that doesn't do the reverse polarity thing. We are getting right past the edge of my limited knowledge with some of this stuff.

So what about transformer balanced inputs. What are the different ways those are used to handle a balanced signal?
Thx

Transformers by nature are balanced differential inputs. They only react to the difference in voltage at their inputs, and their inputs have closely matched impedances. For a balanced output that uses the reverse polarity scheme, the transformer would have one side positive while the other is negative.

So if your output was feeding the transformer a +1 volt signal and a -1 volt signal, the transformer would have a two volt potential across it. Noise, which will present the same polarity voltage at both ends of the transformer, will end up with a net of 0 volts across the transformer, and will not pass to the transformer output.

That's why the impedance of the legs to ground needs to be equal. Different impedance in the conductors means the noise will have different voltages on each leg. When the noise hits the input, it won't add to 0, and you will get noise.

There's more, gotta go to work. I'll hit you with some links later.

 

[Reggie]

Also, the signals aren't "recombined". They are fed to the inputs of a differential amplifer, which senses differences in voltage across at it's inputs.

People get this "recombined" idea from the fact that addition and subtraction are used to describe what happens. The values can be added and subtracted to see what happens, but the signals themselves don't recombine. The input senses the voltage difference, and outputs a corresponding signal. Common-mode noise, which presents zero total voltage difference after inversion, never gets past the inputs. Signal, which presents a voltage difference, gets passed along.

I may need you to expound on this area for me, because I have a hard time understanding opamp theory (and a lot of other things), but I figure now's a good time to sort it out!

I was with you on the impedance needing to be matched to give it the term "balanced," BTW.

And I guess recombine is a bad choice of wording on what happens to the + and - signals; would convert be the more accurate wording?

Oh wait, something just clicked. For some reason I had been trying to picture noise being different on the + and - signals for some reason, and the opamp ditching anything that didn't cancel out while inverting and combining the 2 clean signals. I think I need to see a sketch, but let me know if I am on the right track here: the common-mode noise follows the same polarity on both + and - signals, so if you were to look at it the + and - signals would look exactly opposite but the noise would have the same peaks and valleys on each signal. So inverting the - signal would cancel out any noise shared between the hot and cold signal.......but what comes back out of the opamp? When the cold signal is inverted it doesn't meet with the hot signal in there? I guess I just have no idea what really goes on inside those dang opamps!

Perhaps you could also elaborate on this?:

The signal is not necessarily split, that's not a requirement for balancing. Some schemes use it, some don't.

Sorry if this is hijacking the thread, but I think it is still kind of related to the topic at hand.
Thanks

 

[Treeline]

This is getting into cool stuff. Keep it going! I'm taking notes at this stage...