auxinput said:
That being said, if you are using a larger gauge solid-core Romex, such as an 8awg or 6awg, this may naturally help roll-off or remove any high frequency noise coming in from the power grid. Just make sure you are using solid-core. Stranded THHN (or similar) won’t do the same thing.
#10awg solid copper wire is the largest found in Romex. #8 and larger conductors are made from stranded wire.
The largest single solid insulated wire manufactured, per NEC (National Electrical Code) is #8. I might add a real pain in the a** to work with.
For those thinking of using #8awg or #6awg wire for a 20 amp branch not only must the wire size be reduced at the receptacle to a maximum size of #10awg the wire that connects to the 20 amp breaker in the electrical panel must also be reduced. (Of course where the wire is stranded a crimp wire to fork lug could be used.)
Something else to consider when using #8 or #6 Romex is the number of conductors in the Romex. IF you use 8-3 with ground or 6-3 with ground you take the chance of an induced voltage being transferred from the Hot and neutral current carrying conductors onto the equipment grounding conductor of the Romex cable because of the design, geometry, of the cable. Though I have not seen 8-2 with ground, let alone 6-2 with ground, Romex apparently it is manufactured. Two conductor with ground is what should be used to minimize induced voltage onto the equipment grounding conductor.
//
As for the MSB article, I don’t think the differences in SQ the listener hears, whether the wire is 12, 10, 8, or 6 gauge is necessary due to the resistance of the wire. Though for a long run VD, Voltage Drop, might come into play where the connected load is a power amp of a descent size, wattage. If that was the only reason VD would not be an issue for a dedicated branch circuit used to only power a preamp or a digital front end connected load.
So what is possibly making the differences with the different sizes of branch circuit conductors? Not sure, but I wonder if it has something to due more with impedance caused from inductive reactance of the power supply of the audio equipment.
.