• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Would increasing power minimize noise on unbalanced cables?

Lambda

Addicted to Fun and Learning
Joined
Mar 22, 2020
Messages
841
Likes
656
XLR interconnect system does not have a ground conductor. Pin #1 is the shield.
That's semantics again.
Shield can be connected to ground and at one or both ends and be used as ground reference potential.

Good luck concocting equipment that's powered from 2 pin power plugs with XLR interconnect where shield is only connected to one side.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
5,672
Likes
10,130
Location
Monument, CO
That's semantics again.
Shield can be connected to ground and at one or both ends and be used as ground reference potential.

"Can be", but normally is not. Fully differential circuits do not use ground as a reference; it is only a leakage path.

Good luck concocting equipment that's powered from 2 pin power plugs with XLR interconnect where shield is only connected to one side.

??? Happens all the time and usually works fine, though without the safety ground/shield/chassis connection you'll lose some common-mode noise rejection.
 

mansr

Major Contributor
Joined
Oct 5, 2018
Messages
3,884
Likes
8,727
Location
Hampshire
Phantom power for microphones is an application that comes to mind where all three wires are essential.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
2,392
Likes
5,882
Location
Suffolk UK
That's semantics again.
Shield can be connected to ground and at one or both ends and be used as ground reference potential.

Good luck concocting equipment that's powered from 2 pin power plugs with XLR interconnect where shield is only connected to one side.
With a shielded twisted pair, both ends should be connected to pin 1 of the XLR. This is regardless of whether the equipment (at either end) is double-insulated or fitted with a safety earth. if equipment at both ends are earthed, then the balanced and differential circuit will reject any hum resulting from an earth loop. If only one end, or neither end is earthed, then the screen will either do nothing or be only partially effective, but the balanced and differential part will still be operating. Transformer centre-tapped balanced differential circuits are very rare these days, and those are referenced to earth as a leakage path, (as Don mentioned above) the signal is still merely differential between pins 2 and 3, without any reference to pin 1.

S.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
2,392
Likes
5,882
Location
Suffolk UK
Phantom power for microphones is an application that comes to mind where all three wires are essential.
Correct insofar as a screen is necessary for the powering of the microphone or whatever, but isn't necessary as far as signal goes. With microphone circuits, it's normal to have screened twisted pair or star-quad cables so one doesn't have think about a phantom power path. For line circuits, I've seen whole studio complexes wired with unscreened twisted pairs (for both digital and analogue) with no crosstalk or hum issues, even if I would be wary of doing that myself.

S.
 

Lambda

Addicted to Fun and Learning
Joined
Mar 22, 2020
Messages
841
Likes
656
"Can be", but normally is not. Fully differential circuits do not use ground as a reference; it is only a leakage path.

But this leakage path to a common voltage (sometimes ground) is need often a good thing to have.
Happens all the time and usually works fine, though without the safety ground/shield/chassis connection you'll lose some common-mode noise rejection.
If input our output is using a transformer its fine. they have near infinite common mode resistance and good common mode suppression.

A non ground referenced devices is Capacitive coupled to some ware between mains and neutral.
worst case it can be 240V and supply up to 0.5mA (depending on local regulations)
more typical would be 120V or 1/2 mains and ~0.1mA

But this would still put a huge common mode voltage on an typical line receiver.
Screenshot_2021-02-23_17-01-40.png


Lets say its a 100k, so 50k in common mode. so common mode voltage can be 5-25V? (EDIT: teste my speakers its it's 200k in my case)
Or 13 to 25dBV assuming the receiver can Handel 25dBV common mode
Whats a typical CMMR for a powered monitor? maybe 30dB?

So in this case -17 to -5dBV noise can be added by not using the shield as a ground reference

If you don't think that's a fair assumption or not typical feel free to bring your own numbers to the table and make your own calculations.

but normally is not.
Your normal seems to be different to my normal.
I tested all my equipment in reach and sure pin 1 is always connected to chassis ground by default (some has ground lift function)
Also ti app note calling it "Typical balanced audio-transmission system"
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
5,672
Likes
10,130
Location
Monument, CO
But this leakage path to a common voltage (sometimes ground) is need often a good thing to have.

If input our output is using a transformer its fine. they have near infinite common mode resistance and good common mode suppression.

A non ground referenced devices is Capacitive coupled to some ware between mains and neutral.
worst case it can be 240V and supply up to 0.5mA (depending on local regulations)
more typical would be 120V or 1/2 mains and ~0.1mA

But this would still put a huge common mode voltage on an typical line receiver.

Lets say its a 100k, so 50k in common mode. so common mode voltage can be 5-25V? (EDIT: teste my speakers its it's 200k in my case)
Or 13 to 25dBV assuming the receiver can Handel 25dBV common mode
Whats a typical CMMR for a powered monitor? maybe 30dB?

So in this case -17 to -5dBV noise can be added by not using the shield as a ground reference

If you don't think that's a fair assumption or not typical feel free to bring your own numbers to the table and make your own calculations.


Your normal seems to be different to my normal.
I tested all my equipment in reach and sure pin 1 is always connected to chassis ground by default (some has ground lift function)
Also ti app note calling it "Typical balanced audio-transmission system"

I think we have different terminology as well as different power grids. The differential circuits I've designed or played with have an internal reference to some sort of common "ground" so capacitively coupling the signals does not allow mains voltage or anything like it across the "floating" nodes. It's usually a high-value "bleeder" resistor as you said and that limits the potential in the line; I have never seen anything anywhere near line voltage at that point unless something was broken. It's usually in mV to a few volt range. But I have not measured that in years.

And yes pin 1 is usually connected to the chassis. But neutral is ~0 V unless the product is wired wrongly; having the chassis tied to the "hot" line would kill someone and probably blow a fuse (hopefully not in that order!)

I have no idea the CMRR of a powered monitor. For mixers and such, it is usually in the 60 dB range, 80 dB for a well-designed circuit, and 40 dB or less for a poorly designed down to maybe 6 dB for a simple quasi-differential receiver (basically single-ended but with a compensatory resistor in the (-) leg to balance the input impedance -- more or less).
 
OP
T

tifune

Senior Member
Joined
Jan 18, 2020
Messages
342
Likes
187
Is that speaker wire, aka zip cord? If so, analogue balanced signal with the highest voltage you can obtain is probably the best choice. What sort of inputs do your active speakers have?

Yes it is zip cord, so just +/-. Speakers all have XLR in, some like the Kali and LSR also have bal/unbal 1/4" or RCA. At the termination point it's 4-wire, but somewhere in the wall it breaks into 2x2 wire via butt connectors. So far i 've only found one of those break points out of 5, unfortunately.

I'll probably try SPDIF for giggles but I'm not expecting miracles - based on most replies here, because this is all 2 channel it seems as though tapping into my EVAL-1 with a passive attenuator switch box might be easiest. Maybe the switch box isn't even necessary depending on the tolerances if the input - I believe my Neumann KH120s will deal with +24dBu on the input, for example, and then i just turn them off when not in use. I was hoping to do it all via JRiver for inline room correction but one thing at a time I suppose
 
Top Bottom