how is noise cancelled in a balanced signal?

[Farview]

Actually, Sonixx explanation seems totally different than the "polarity, OpAmp"
explanation to me?? A "difference" not a "Summing" explanation?

difference is summing a negative and a positive. 3 minus 2 is one. 3 plus negative 2 is one. It's the same thing.

Balanced INPUTS & Balanced OUTPUTS on the equipment.
so a Balanced signal could be sent "from" unit#1 to unit#2 "to" a Balanced Input. OK, this is the ideal setup.

So what if one piece of equipment in the chain isn't Balanced???

If both sending and receiving equipment is not balanced, neither is your signal. Both the from and to units have to have balanced i/o.

......and this has nothing to do with Chasis NOISE or AC Line noise...does it?
Which is probably more common a problem.[/B]

No it only has to do with noise that might be picked up in that cable itself. It isn't a common problem with short runs of cable unless they are run in a high interference area.

 

[bennychico11]

difference is summing a negative and a positive. 3 minus 2 is one. 3 plus negative 2 is one. It's the same thing.

THANK YOU!!!!!
5th grade math

If both sending and receiving equipment is not balanced, neither is your signal. Both the from and to units have to have balanced i/o.

i'd like to modify that a bit and say "if sending and receiving equipments are NOT BOTH balanced at the same time in the chain, then neither is the signal."

 

[Sonixx]

geeze, where to start...

Summing and Difference (Differential) are not the same thing...

Summers add... Differential subtracts. Difference is not summing a negative and a positive. Summing a negative and a positive is summing. Subtracting is difference. This is not a distinction without a difference

OP amps are differential, not summing.

Sure, if you invert one of the inputs, you'll end up with the same thing. The reason one signal of a Balanced Source is inverted is because the Receiver calculates the difference. If it summed, both signals would have to be positive or negative. Again, noise canceling is achieved regardless of the presence of a signal. A differetial signal accommodates the Receivers needs. Noise canceling in this type of design is called Common-Mode Rejection. Common-Mode Rejection gives an indication of how well signals common on both lines are rejected.



On to interconnecting Balanced and Single-Ended...

Both source and receiver do not have to be balanced... with properly designed equipment, both are compatible, but with some loss of functionality. This loss of functionality is different based on which end is not balanced.

A Balanced source can drive a Single-Ended Receiver. The result is a 6db reduction in signal level, based against a balanced receiver, and no interconnect noise canceling

A Single-Ended Source can drive a Balanced Receiver. Result is a 6db reduction in signal level. With a specially crafted cable, the Receiver can see near equivalent resistances and significant noise reduction can be acheived, but not as good as in using a Balanced Source.

 

[Sonixx]

THANK YOU!!!!!
5th grade math




i'd like to modify that a bit and say "if sending and receiving equipments are NOT BOTH balanced at the same time in the chain, then neither is the signal."

sorry man, it's not the same thing...

 

[travelin travis]

Again, noise canceling is achieved regardless of the presence of a signal.

ok, you lost me here. if there is no signal present, there is no noise introduced into a signal. how can noise cancelling be achieved if there is no noise?

 

[travelin travis]

ok, maybe i do understand..........

are you saying that the difference in impedance on the hot and cold lines is what is calculated? not the voltage? i'm still unclear on exactly what the noise is being introduced into, if not a signal.

 

[Sonixx]

ok, you lost me here. if there is no signal present, there is no noise introduced into a signal. how can noise cancelling be achieved if there is no noise?

I think you are getting the noise present in the Source confused with noise that is induced into the interconnect (cable) by external sources. A Balanced interconnect design is not a Single-Ended noise reduction system, say as in a Hiss remover. A Balanced design minimizes external noise (electrical and magnetic) interference. Whatever noise levels are output by the Source remain.

 

[Sonixx]

ok, maybe i do understand..........

are you saying that the difference in impedance on the hot and cold lines is what is calculated? not the voltage? i'm still unclear on exactly what the noise is being introduced into, if not a signal.

no not quite...

In a Balanced design both signals have the same impedance (Both Source drivers present the same impedance level, the Cable is considered negligible (almost), the Receiver presents two loads each with the same impedance). This is a must for the Differential input at the Receiver to work correctly. The Common-Mode Rejection (see above) of the Receiver is what cancels the noise, because the same noise signal (i.e. level) is induced equally in both conductors and since the Receiver is differential, it subtracts the common-mode signal which ends up equal to ZERO. But, equal line impedance is required. The Source signal is differential, not common mode, so Subtracting it gives you 2 Times the signal level on either line.

Also, don't think of the signals as Hot and Cold or Plus or Minus. They're Differential.

I hope this helped... if not I'll try again.

-keith-

 

[Samantha C.]

I'm curious why it is that I often hear of mastering engineers using unbalanced cables in their setups rather than balanced cables (or if that's true at all)?

What's the the reasoning behind this?

Samantha

 

[Sonixx]

I'm curious why it is that I often hear of mastering engineers using unbalanced cables in their setups rather than balanced cables (or if that's true at all)?

What's the the reasoning behind this?

Samantha

Because a very good single-ended design has better specs and is less expensive to design than balanced. Also, if your interconnects are short, single-ended is less complex and avoids the issues with creating/receiving two signals. Balanced is better for longer runs due to its noise reduction.

Inside each unit, all signals are single-ended anyways. Balanced interconnects are a means for solving several problems such as externally induced noise and ground loops.

 

[Farview]

Both source and receiver do not have to be balanced... with properly designed equipment, both are compatible, but with some loss of functionality. This loss of functionality is different based on which end is not balanced.

A Balanced source can drive a Single-Ended Receiver. The result is a 6db reduction in signal level, based against a balanced receiver, and no interconnect noise canceling

A Single-Ended Source can drive a Balanced Receiver. Result is a 6db reduction in signal level. With a specially crafted cable, the Receiver can see near equivalent resistances and significant noise reduction can be acheived, but not as good as in using a Balanced Source.

Balanced and single ended are compatable, the point I was making is if one is balanced and the other is single ended, you don't get the benefits of a balanced signal.

And yes, there is a difference between a summing circuit and a differential circuit but as far as the math involved, difference is the sum of a positive and a negative number.

 

[travelin travis]

no not quite...

In a Balanced design both signals have the same impedance (Both Source drivers present the same impedance level, the Cable is considered negligible (almost), the Receiver presents two loads each with the same impedance). This is a must for the Differential input at the Receiver to work correctly. The Common-Mode Rejection (see above) of the Receiver is what cancels the noise, because the same noise signal (i.e. level) is induced equally in both conductors and since the Receiver is differential, it subtracts the common-mode signal which ends up equal to ZERO. But, equal line impedance is required. The Source signal is differential, not common mode, so Subtracting it gives you 2 Times the signal level on either line.

Also, don't think of the signals as Hot and Cold or Plus or Minus. They're Differential.

I hope this helped... if not I'll try again.

-keith-

ok, i aplogize ahead of time. i am not completely at fault but a big ass can of worms has been opened up here and now i have a lot more questions going thru my head and i'm not comfortable with not knowing the answers. you guys can tell me to shut the hell up at any time. ok, here goes:

i really do appreciate the effort but I am confused. i don't understand what common mode is. can you explain?

 

[travelin travis]

Because a very good single-ended design has better specs and is less expensive to design than balanced. Also, if your interconnects are short, single-ended is less complex and avoids the issues with creating/receiving two signals. Balanced is better for longer runs due to its noise reduction.

Inside each unit, all signals are single-ended anyways. Balanced interconnects are a means for solving several problems such as externally induced noise and ground loops.

how is a unbalanced short cable better than an balanced long cable? what exactly is considered short cable and long cable? I've heard that anything under 20 feet, concerning mic cable, is'nt really suceptible to line noise, interferance. is this true? is this also true for interconncts and speaker cables?

what is lost over a long cable, balanced and unbalanced? dynamic range? high frequencies?

 

[travelin travis]

Balanced and single ended are compatable, the point I was making is if one is balanced and the other is single ended, you don't get the benefits of a balanced signal.

And yes, there is a difference between a summing circuit and a differential circuit but as far as the math involved, difference is the sum of a positive and a negative number.

so, the definition of single ended is a single conductor? i'm not exactly sure what single ended means. excuse my ignorance but that's why I'm here, to learn.

 

[travelin travis]

Because a very good single-ended design has better specs and is less expensive to design than balanced. Also, if your interconnects are short, single-ended is less complex and avoids the issues with creating/receiving two signals. Balanced is better for longer runs due to its noise reduction.

Inside each unit, all signals are single-ended anyways. Balanced interconnects are a means for solving several problems such as externally induced noise and ground loops.

two questions.

1. how can a balanced cable solve a ground loop? (maybe three questions) ok, i know that in an unbalanced cable, the ground is also the shield. in a balanced cable, the ground is isolated and the two hot wires, or hot and cold, however you like to think of it, carry an equal signal of opposite voltages. ..........damn, another question still yet, how exactly does a negative voltage exist? .........ok, back on track. is a ground loop caused by a difference in voltage? what exactly is a ground, technically? a voltage of zero?

ok, i know i sound really ignorant here. i have read, multiple times, the basics of electricity and electronics over the years. i understand electron flow (i think i do anyway), ohm's law, inductance, the basics of electronic components and their functions in a circuit, but I really do not know how both a negative and a positive voltage can exist.

2. is something lost when an extra signal is "created" or "received"? how exactly is an exact opposite signal created?

sorry, that's alot more than 2 questions.

disclaimer: damnit, my name is not Walters!

 

[Sonixx]

Balanced and single ended are compatable, the point I was making is if one is balanced and the other is single ended, you don't get the benefits of a balanced signal.

A Balanced Receiver can benefit even with a Single-Ended Xmitter using a specially crafted cable. The goal is to supply the Receiver with essentially the same resistance in both conductors. It's basically a Psuedo Balanced design. The Common-Mode is not as high but it's better than than totally Single-Ended.

And yes, there is a difference between a summing circuit and a differential circuit but as far as the math involved, difference is the sum of a positive and a negative number.

Yeah, but they're not the same because the input is differential not summing. A summing input in this case would have no Common-Mode Rejection.

Let's lay this to rest... a Balanced Receiver does not sum the inputs, it develops an output that is the difference of the two iinputs.

 

[Sonixx]

how is a unbalanced short cable better than an balanced long cable? what exactly is considered short cable and long cable? I've heard that anything under 20 feet, concerning mic cable, is'nt really suceptible to line noise, interferance. is this true? is this also true for interconncts and speaker cables?

what is lost over a long cable, balanced and unbalanced? dynamic range? high frequencies?

In this case, it's not that a short single-ended cable is better than a long Balanced. Really, it's that balanced have a function in audio and so do Single-ended. The issues and benefits of single-ended versus balanced is a bit more than I can write on right now. If I can get some time soon I'll will. If both xmitter and receiver are balanced, then use balanced... likewise for single-ended. Balanced has the benefit of canceling noise that is induced into the cable from external sources like transformers and other current carriers such as power cords.

If you have to lay mic/guitar/other audio cables next to/around/over power cords, try to keep them far away as possible and crossing at 90 degrees. This will minimize coupling.

Speakers versus low-level signal...
Speakers are low impedance high current and thus are not nearly as subject to external interference as are high impedance low level signal designs. It takes a much much srtonger field to couple with the low impedance of the speaker and to actually move the coil, so interference is usually not an issue. The physics is different. To help minimize induced noise, one design goal in small signal interconnects is to maintain a load impedance that is significantly lower than the source.

 

[Sonixx]

so, the definition of single ended is a single conductor? i'm not exactly sure what single ended means. excuse my ignorance but that's why I'm here, to learn.

Single-ended is unbalanced... a guitar cord... any interconnect with a 1/4 inch Tip/Sleeve design is single-ended or unbalanced. For small signal audio like a guitar, it's a center conductor with a conducting shield

 

[Sonixx]

two questions.

1. how can a balanced cable solve a ground loop? (maybe three questions) ok, i know that in an unbalanced cable, the ground is also the shield. in a balanced cable, the ground is isolated and the two hot wires, or hot and cold, however you like to think of it, carry an equal signal of opposite voltages. ..........damn, another question still yet, how exactly does a negative voltage exist? .........ok, back on track. is a ground loop caused by a difference in voltage? what exactly is a ground, technically? a voltage of zero?

ok, i know i sound really ignorant here. i have read, multiple times, the basics of electricity and electronics over the years. i understand electron flow (i think i do anyway), ohm's law, inductance, the basics of electronic components and their functions in a circuit, but I really do not know how both a negative and a positive voltage can exist.

2. is something lost when an extra signal is "created" or "received"? how exactly is an exact opposite signal created?

sorry, that's alot more than 2 questions.

disclaimer: damnit, my name is not Walters!

wow, you're asking a lot here

Voltage measurements are always referenced to some other location in the circuit (i.e. a reference point). Probably most of the time it's to ground. So Ground is assumed to be Zero Voltage reference. Now based on the circuit design, voltages can be above the reference or below it. Remember that voltage measurements are always relative to some other point. If you flip the leads on your VOM, the Voltage reading will read the negative.

Ground loops (GL)...
A GL is when two devices that operate similarly have different ground references and are connected. Here's an example that you can prove...

Power one piece of equipment (cassette deck) directly from your wall socket...

Power another piece of equipment (the amp) using a long extension cable. Now connect these two... 60Hz buzzolla...

The reason is significant 60Hz current is flowing between the two devices because they do not share a common ground but expect too. The cassette deck to amp interconnect is trying to bring the two devices to a common ground point. Not a good thing... The extension cable acts like a resistor and you have a significant voltage drop in the extension cable which changes the ground reference point for the amp.

How can a balanced design help this...

First, the balanced cable ground lead is connected only at one end in the cable, so no current flows in the ground lead. There's a preference to connect the ground at the Xmitter or Receiver, but I don't recall

Next the balanced receiver only looks at the difference between the two signal carriers and not the difference between the signal and ground... thus the ground connection is taken out of the circuit... voila no ground loop


How is a balanced signal created...
It's common to use two opamps with one inverting and the other not. But the design requires very high component tolerance (i.e. resisitors, caps and layout.) The goal is to create two signals with exactly the same gain and output impedance. This goal is never totally realized... but a very well designed Xmitter presents a better common-mode signal to the receiver.


What is Common-Mode Rejection...
It's the ability of the receiver to reject signals that are common on both conductors. This is the key... Remember, the receiver develops a signal that is the difference of the two input signals, so assuming the induced noise signal is the same on both conductors, (i.e. it is common on both conductors), the receiver will reject it by subtracting the two (not summing). Now, since the Xmitted signal is differential (i.e. one is positive and the other is negative) subtracting these two we end up with two times the signal level.

 

[CoolCat]

yeah, Common mode rejection...heard that one, sometime ago.
the Balanced does use the OpAmps you say...
so now it looks more and more as Southside had the +( , -)
..drawing...this is more of an accumulative article mentioning the actual application of the OpAmp and circuit id.

a great thread. XLR, Stereo 1/4" plugs, 3 conductor= balanced.
Single 1/4" and Rca type wiring = Unbalanced.

I was so excited i went thru all my wires late last night.
In coming is cool.

but my Output is RCA plugs going to my Tascam amp (amp has balanced inputs available).

has Monitor Out 1/4", RCA jacks and SPDIF Outs..
i'd love those AIR DynAudio..$4,000 monitors for SPDIF, that'd be sweeeett! just gearlust