How does noise affect an unbalanced signal on a balanced cable?

Question

With regard to audio and balanced signals with XLR connectors, how does noise affect an unbalanced signal on the same cable that normally carries a balanced signal?

To give some background, I have a five metre cable run from an unbalanced source to a balanced amplifier that isn't using very good cable and is picking up a lot of audible noise. I can either replace it with better shielded unbalanced headphone-style cable, or move the RCA to XLR adapter closer to the source and then run two long balanced cables to the amplifier.

The better shielded headphone cable will improve the noise, and of course a balanced signal the whole way would be even better, but what about an unbalanced signal over the balanced cable?

Will the noise be worse, because the balanced cable has less shielding? (I'm assuming it has less because the balanced nature of the signal allows noise to be better cancelled.) Or will the unbalanced signal still benefit from the noise reduction applied to the signal at the destination? What if the source ties the unbalanced audio GND signal to the same chassis GND used for the cable shield - does that reduce the noise in one of the balanced lines, resulting in less noise reduction when the two "balanced" signals are recombined?

The reason I am asking is that I'm not sure exactly how an unbalanced signal is presented on a balanced input. There doesn't seem to be any active conversion taking place in many of the adapters you find online, so I'm not sure whether the amplifier needs to be designed to accept an unbalanced signal, or whether any balanced input can accept an unbalanced signal without any problems.

Answer 1

How well this works depends on the output impedance of the source, and the common mode impedance of the input.

Basically a balanced circuit works perfectly and delivers the full interference rejection benefits even if only one line is driven, providing the other leg is connected to the reference plane at the source via an impedance equal to that of the driven leg.

Certainly moving the adaptor close to the source is worth a shot, but I have one further question:

Are you sure you don't have an earth loop going on?
Pin 1 problems are still out there, and weirdness in unbalanced gear is fairly common, you might find that moving the adaptor AND disconnecting the screen at one end works best (It makes the screen effective for low frequency electric fields only, but stops circulating current in the ground network).

A typical XLR cable is at least as well screened as a jack lead, plus the twisting of the pairs gives it MUCH better magnetic field rejection.