What is the correct wiring for using shielded cat6 for unbalanced audio? I'm using these connectors for audio out from a Sony TV to a Sonos Amp, about 20' away. Should I connect the shield to ground on both sides or just one? I'm current using three pairs for L,R and ground with the shield tied with the ground as well on both ends. Client states she occasionally gets loud noise on one speaker; never noticed it myself. I am guessing it is from the LED drivers and it probably happens only at night. Will test it today with all the lights on.

[Link: snapav.com]

[Link: snapav.com]

I'm assuming from the way you wrote this that your source is unbalanced, meaning it has only a hot and a ground. If you mean that since the Sonos's input is unbalanced but your source is balanced, so you need to know how to hook that up, the answer will be different.

Pick a pair from the CAT6. Always wire CAT cables in pairs, as that takes advantage of the noise-rejecting properties of the twisting.

These are your audio connections. Your shield is the third wire. I'll pick the blue pair for no particular reason.

The Sonos Connect's input connectors are RCAs, which have only hot and ground. Connect blue/white to the hot at the source and to the hot at the Connect;
connect white/blue and the shield together and to the ground at the source;
connect the white/blue to the ground at the Connect;
don't connect the shield at the Connect end.

The sense of this is that the two wires conduct the signal, and any noise picked up by the shield is shunted to ground at the source end. The source is likely to be a lower impedance than the following stage, so the noise has less of an effect when grounded there; and connecting the ground at the cable output end would directly feed the noise into the following stage.

(Back when Monster Cable introduced "directional" audio cables, that's how they were wired. I've seen studio audio cables to be used in audio patch bays as old as from the fifties that were wired this way, too.)

eeek.....

just use a balun...[Link: snapav.com]

Dude,
it's not a mouse.

It's a line level connection that will work perfectly when done the way I described it. A balun is not needed unless there are ground potential differences between the two locations, which is unlikely (see Posts 1, 6, 7, 9, 16, 17, etc at [Link: remotecentral.com]).

Don't spend the money on a balun if you don't need one. Don't assume you need one if you haven't tried this. And yes, the way I describe wiring it IS the way Monster did all their audio cables, and they worked successfully. The few cases that had problems often turned out to be subwoofer connections where the sub and the system were on different phases of the power. It's been a long time since I've looked, but at least some subwoofer ground connections are pretty squirrely -- for instance humming when there's no connection but not humming as soon as connected to a cable!

Thanks Ernie,

Just to be clear, you mean the solid blue to for the hot and blue/white for the ground, right? Yes, the signal is unbalanced, just the analog audio out from the Sony XBR on a 3.5mm jack.

Since I have left and right, i'll connect the Blue for the right and the Green for the left, blue/white and green/white together for the ground and the shield wire for the ground only at the TV.

Yes.

Right.
I try to keep color coding as simple as possible so I won't have to try to remember what I did five years ago when I see my work again. I'd use blue and orange just because those are pairs 1 and 2 of the four pairs. If I was trying to use that cable for a phone connection, too, I might use blue and orange for that, then green and brown for the audio.

When I use CAT for IR sensor wiring, I use orange for power since it's the closest color to red....

I don't know of any ways the four colors are used such that blue and green are the first two pairs to choose, so i wouldn't do that. There's no technical reason to choose or not choose any color pair.

By the way, you gave no indication that you were going to use that connector for both channels; I assumed you were going to use the three-terminal connector for three wires of one channel and you wanted to know how to do that. As always, more details in the description and question help to pin down the answer.

20' and you suggest a balun?

Anybody who drives a 3 Pedal Mini should be brave enough to go twenty feet without a balun. Even if not, I don't understand trying it out first without the balun. If it works, you've saved yourself or your client money and time ( = more money).

This is like deja vu, all over again.

highfigh,
Yes, these things do come up over and over. You'd think the other thread would have given all the answers needed and this thread would be unnecessary, right?

You've never heard of TIA568B?

hdsystems, you don't state your point. Yes, I've heard of it. Now I'll respond to what I am guessing you wanted to say.

568A names the pairs 1 - 4 as blue, orange, green, and brown. This is identical to the way the pairs are numbered for phone cable use.

568B is the same except exchanges the positions of the orange and green pairs. So, yes, I've heard of it. Or, really, them.

568, whether A or B, defines the placements of the so that everybody can do it the same way and techs can tell at a glance which standard is being used. This is called a convention, meaning we've decided we're going to do it that way because we can't afford to figure out on every damn installation just how the previous people decided to wire it. This is not a technical reason to choose colors, and I said there's no technical reason for certain colors to be in certain positions.

To me, a technical reason would be, let's say, if blue always had more twists to the inch than the other pairs, or some other fact that made the pairs perform differently from one another. You might find that the CAT5 or 6 or E you're using ALWAYS has more twists per inch in the green pair, so you'd want to use that pair for some special thing.

The thing is, different pairs have different twist rates to lower crosstalk between pairs but the spec does not call out which colors have which rates of twist. Another model or brand of the same wire you're now using could have the most twists on the brown wire. All that's required is that the four pairs not have the same twist rates as one another. So you can't spec a particular color pair for a particular technical result.

So... is there a technical reason for choosing one color pair over another?

By the way, all the details in this discussion are so far beyond the needs of the thread that they don't really even need to be discussed.

Mostly 568A and 568B pertain to preventing technician "A" and technician "B" from terminating the ends of the same cable differently. However, I have seen a few specialty applications (some baluns) specifically specifying that pair 1&2 not be split (ie., both wires of each pair are adjacent pins on the connectors) I haven't seen any numbers, but there will be a slight difference in the crosstalk characteristic at the ends vs standard 568A or 568B.

Left on my own I'll use 568A, but I always check a takeover house and use the prevailing convention.

|This is identical to the way the pairs are numbered for phone cable use.

It is, but you also mentioned using the blue as the positive and POTS installers don't- they use the white/colored tracer for the Tip and the solid color for the sleeve.

This is not a technical reason to choose colors, and I said there's no technical reason for certain colors to be in certain positions.

In POTS or networking, two configurations are used and for any other use, the only thing that matters is that the ends are the same (assuming that's necessary). There's no standard for using Cat5/6 when some kind of phone plug is needed.

Less twist results in less noise cancellation so yes, there's a reason if the signal dictates this. Brown always has the lowest twist rate.


And yet, here we are and thanks for unnecessarily commenting on my 'deja vu' post.

That's totally true. However, the spec defines crosstalk levels, which can be reduced by using different twist rates. It does not say which color gets which amount, nor even more/less, of twist.

As long as you verify before wiring to be sure you're using the highest twist rate wire for the most critical signal, you'll be good. Since the twist spec does not state which colors should have the highest amount of twist, any such choice depends on using the same brand and model, and also depends on nobody at the factory deciding to make a random color change... since it's not part of the spec.

I have the impression that the placement of pairs around a spline in CAT6 is identical from cable to cable but I don't know if that's a spec, or just a practice.


If you're lucky. Just don't depend on it.



*and someone at the factory doesn't decide to make a change, which they could do, since which color is which doesn't matter