Why does a TRS cable minimize noise?

  • Thread starter Thread starter lirvala
  • Start date Start date
L

lirvala

New member
Can someone explain to me why a TRS cable minimizes noise??

From what I understand, is that a TRS is 3 cupper wires; one is ground and the other two are for transmitting the signal, except that one is 180 degrees out of phase from the other. When the signal is recieved, the two signals are added and the noise is cancelled out.

I don't understand how a two signals being 180 degrees out of phase will minimize the noise??!
 
lirvala said:
Can someone explain to me why a TRS cable minimizes noise??

From what I understand, is that a TRS is 3 cupper wires; one is ground and the other two are for transmitting the signal, except that one is 180 degrees out of phase from the other. When the signal is recieved, the two signals are added and the noise is cancelled out.

I don't understand how a two signals being 180 degrees out of phase will minimize the noise??!
Click to expand...

Just do a little math. Let's say you have a signal at a given instant of 2. If you divide that into in and out of phase sides, you have 2 and -2. Let's say the cable gets interference of .5. You now have 2.5 (2+.5) and -1.5 (-2+.5). If you invert the out-of-phase signal, and add to the original signal, that's (2.5-(-1.5))/2=2, which is the same as the original signal of (2-(-2))/2=2.
 
it has to do with the principles of sound. we know that when two signals are out of phase the signals cancel. now with the cable, what happens is the signals are put out of phase before entering the cable. The in-phase signal traveling down one wire while the out-of-phase one traveling down the other wire. During their trip to the other end, noise in form of electrical static interferance, radio freq., etc. can be introduced to the signal. Now this will be introduced to both the hot and cold wires....the in-phase and out-of-phase wires. When it gets to the other end the phase is then reversed on the out-of-phase cable only, putting it back in-phase. At the same time, this reverses the phase of the noise signal on that wire...and what happens then? Well, that noise is out of phase with the other one creating cancellation of the noise!!! At the same time when we add our two good signals after we reversed them back, we get more amplitude in our waveform (adding two waveforms together raises the signal). So balanced=good! I know this seems like such a simple procedure now, but I still think it's one of the coolest ideas someone came up with in the audio world. So hats off to that guy :cool:
 
Hey, I kinda screwed up. This post is how out of phase signals cancel. The next is about balanced cable noise. Sorry.


This will be easier to understand if you think of things in terms of waves. In audio we are basically concerned with sine waves. Knowing what they are and how they interact is gonna be important to getting this whole idea.

Here is a picture of a full cycle of a sine wave. As you see it goes from zero, up to a positive value, to zero, to a negative value, and back to zero. Sound waves and elecricity do this. How many times this happens in a second is the frequency of the signal. If a voltage or airwave does this 20,000 times a second, you get a 20k signal. A signal that is 180 degress out of phase with the sine wave below starts at zero, goes negative first, to zero, then positive, then to zero.

Most of our gear (and ears) work in part on relative differences. In the case of our two sine waves, or sound waves, or rf noise in a cable, or whatever, as one wave goes positive, the other goes negative. They add to zero.

Here is an applet that you can play with. Set the two waves to equal wavelength and amplitude, and play with the phase to see the effect of adding the two waves together.

Sine waves
 

Attachments

  • sine.webp
    sine.webp
    690 bytes · Views: 297
Last edited:
Whoops- didn't quite answer your question. The above post merely explains about how two out of phase signals cancel out. How balanced lines reduce noise is another thing. MSHilarious got it. It's not a question of anything being out of phase, really. It's just simple addition and subtraction.

There are a couple of kinds of ways balancing works.
One is called common mode rejection. The piece of gear receiving the signal can only see voltage differences between the two signal wires, ie a line signal at 1volt. one wire is plus, one minus. The gear senses the voltage differential, as it called, between the two wires, and passes this along. (hopefully your awesome vocal track or best guitar solo)

When noise enters the cable, it enters both conductors at equal levels. it travels along towards your gear. When it gets there, it is present on both wires at the same level. Your gear doesn't see any voltage difference, therefore the noise does not get through.

Benny provides a great explanation of the out of phase type of noise reduction. In this case the common signal goes down both wires. As he said, in this kind of balancing the same signal is on both conductors, but out of phase. One input on the recieving gear is inverting. Any noise that gets in gets that side reversed, and adds with the non-inverted noise, and cancels out.
 
Last edited:
I'd like to add that it's an actual circuit called the balancing circuit that does the work (along with a differential amplifier at the beginning to help us). The balancing circuit is the unit that is inverting the out of phase wire, though. I don't think it's just a simple matter of passing on the + and - signals. Because those signals will be still out of phase. You need to switch them back. Yeah, a lot of it is a matter of addition and subtraction, but you need to keep in mind that when the two signals on the wire combine at the end, they're adding another number...3dB. You're doubling the original signal when flipping it back. Here's some extra reading for you:
http://homerecording.about.com/library/weekly/aa082697.htm
http://www.whirlwindusa.com/tech01.html
 
Yeah, that's true. The signal gets inverted and passed down pin 3. Balanced outputs and inputs on gear are usually active and use this scheme.

But dynamic mics, for instance, do not use an active balanced output. They are balanced by definition, ie identical impedance between pins 2 and ground and 3 and ground. That's all balanced means, really. Both conductors have the same reference to ground(zero), allowing a balanced input to perform it's common mode rejection function. The noise voltage presents no voltage differential at the input device and therefore cancels. There is no active inverter in the sending unit(microphone). The negative polarity signal from the mic is a result of the function of the transformer.
 
Great! Thanks for everyone's help. I think the text I was reading missed the part about flipping the cold signal back so that the noise is cancelled.


One more question.......

is it bad for the signal to be doubled when we do this?? I dunno, I kinda got the impression from benny that adding 3dB is not desirable. Any comments on this??
 
no it's definitely okay. it'd be just like adding another speaker at the same volume. doubling the signal gives us 3dB naturally without the aid of a preamp. This means after the circuit balances back our signal, we use less gain in the preamp which also results in less extra noise. Of course the noise that we're losing can barely be heard if even at all. And our signal before it's doubled is very faint....but it's still a good thing! :cool:
 
by doubling i mean another speaker at the same volume....for example, when you listen in stereo. If you have something in the L channel and you add the same thing to the R channel the volume doubles. This is about a 3dB increase.
 
I beleive to double the acoustic output you need a 6dB increase.... stupid logarithmic rules!!! 3dB will get you an increase of 1.41
 
I think you are right. In terms of sound pressure level (volume from a speaker) +3db is noticeable, +6db is moderate, +10db is double. I think +6db signal level out of a mixer would get that volume doubling out of an amp+speakers if you've got the headroom.

Hmmm.... I need a db and logarithm refresher.
 
boingoman said:
I think you are right. In terms of sound pressure level (volume from a speaker) +3db is noticeable, +6db is moderate, +10db is double. I think +6db signal level out of a mixer would get that volume doubling out of an amp+speakers if you've got the headroom.

Hmmm.... I need a db and logarithm refresher.
Click to expand...

True...it all matters on the person. Some people notice a difference at 3dB...some more. That's why I said 3dB. I think it's actually a standard on certain gear as well. I know the SSL 9000J, when recording and panned dead center, you're actually 3dB less than what the person is really playing. This is because they automatically compensate for you when you are listening with two monitors. Therefore you won't notice a difference. This is why you have to do an odd/even pan when recording so that you aren't losing 3dB when going to tape.
 
mshilarious said:
Just do a little math. Let's say you have a signal at a given instant of 2. If you divide that into in and out of phase sides, you have 2 and -2. Let's say the cable gets interference of .5. You now have 2.5 (2+.5) and -1.5 (-2+.5). If you invert the out-of-phase signal, and add to the original signal, that's (2.5-(-1.5))/2=2, which is the same as the original signal of (2-(-2))/2=2.
Click to expand...

Is there anyway to do that in pst production in Sound Forge cause i like where this is going!
 
Thorguitarist said:
Is there anyway to do that in pst production in Sound Forge cause i like where this is going!
Click to expand...

That's a toughie; once the noise is in your signal, it ain't that easy to get out. The idea is to reverse phase before the noise hits the signal.

Post production noise reduction is usually accomplished with gates or very narrow parametric EQ cuts. If you had a source of pure noise then a reverse phase on that might help but I doubt it would work that well.
 
I have a really basic question to tack on to this thread:
I am just starting a bedroom studio and have been struggling with the differences between balanced and unbalanced gear and cabling. should you try to use DI boxes on everything you have that sends an unbalanced signal before it hits your mixer? Or can you run an unbalanced line into a balanced mixer with no problems? How compatible are the two? Thanks in advance for any help.
 
staen said:
I have a really basic question to tack on to this thread:
I am just starting a bedroom studio and have been struggling with the differences between balanced and unbalanced gear and cabling. should you try to use DI boxes on everything you have that sends an unbalanced signal before it hits your mixer? Or can you run an unbalanced line into a balanced mixer with no problems? How compatible are the two? Thanks in advance for any help.
Click to expand...

Could you list your equipment? It depends a lot. Many (most) mixers can take unbalanced inputs with no problems. If your cable runs are short, balanced doesn't matter too much. You should be able to track bass and keys and even guitar if you do that direct unbalanced with too much trouble if you keep your cable runs short.
 
My equipment: I have two synths, a sampler, a laptop (will be picking up a firewire 410 soon) and some effects boxes. I want to buy a set of studio monitors and a 12-16 channel mixer within the next week. A couple more questions: can I use balanced cable to connect unbalanced gear to a balanced mixer? what about connecting unbalanced to unbalanced? should I never use unbalanced cable if balanced is better? When all is said and done, I dont want to negatively affect my gear or the sound quality. Additionally, I would like to start making my own cables when I understand which type I need to be making. does anyone have some good links with instructions on how to get started soldering? thanks for the help.
 
staen said:
My equipment: I have two synths, a sampler, a laptop (will be picking up a firewire 410 soon) and some effects boxes. I want to buy a set of studio monitors and a 12-16 channel mixer within the next week. A couple more questions: can I use balanced cable to connect unbalanced gear to a balanced mixer? what about connecting unbalanced to unbalanced? should I never use unbalanced cable if balanced is better? When all is said and done, I dont want to negatively affect my gear or the sound quality. Additionally, I would like to start making my own cables when I understand which type I need to be making. does anyone have some good links with instructions on how to get started soldering? thanks for the help.
Click to expand...

OK then if you are mixer shopping, you will note that input channels generally have unbalanced (1/4" TS) and balanced inputs (XLR) on the input channels. It probably will have balanced (XLR or TRS) and/or unbalanced outputs as well.

If a piece of gear, like a synth, has an unbalanced output, then using balanced cable doesn't make a difference. If your soundcard/interface has balanced input and your mixer has balanced output, then you can use a balanced cable. However, if your mixer is right next to your interface, then you probably won't notice a big difference if you use a short unbalanced run.

Effects sends on a mixer are unbalanced send and return (using a TRS cable in the channel insert), or might be unbalanced or balanced using aux sends. Your effects boxes might be unbalanced or balanced input, and probably have a stereo return which could also be either one. Usually the boxes are labeled, but if not, XLR is balanced, whereas TRS is either mono balanced or stereo unbalanced. XLR could be stereo unbalanced but that's not typical.

My studio used to be all unbalanced ('cept for mics of course), but is now all balanced, I can't tell the difference--all the cable runs are less than 3'.

As far as soldering, look for a thread on the Studio Building board called like "Who Knows How to Solder". I like soldering TS & TRS a little better than XLR, but none of them are that hard.
 
You must log in or register to reply here.
Back
Top