When a power cord could obviously affect SQ

@ted_b 

Just recently while running 5Hz - 3500 MHz sweeps using a field strength instrument, I noted substantial differences in radiated EMI from various audio power cables. Some emitted surprisingly strong magnetic fields that radiated well beyond the cable jacket, whereas others had undetectable emissions. This implies some cables are clearly better suited for use in close proximity with sensitive digital components (e.g. DACs and streamers) and interconnects.

From an internal perspective, a power cable is not a one-way street, and the better way to view it is not as the last few feet of power feed, but rather as first few feet of what the component "sees" when propagating away problematic internal noise generated by the component itself. This noise exerts direct influence on audio performance, and cable characteristics determine how efficiently it is removed. When one views it from this perspective, it's not a stretch to understand why power cables can "sound" different.

I've quoted your entire post because it should be mandatory reading!  You have provided direct evidence for the theory that it is EMI coming from your component that could make a difference. 

@bruce19 many directly heated triode designs run AC to the cathode - straight from the transformer.

@retiredaudioguy that’s interesting, are there any that you know of that might have a schematic posted online. I have built 4 tube anmps nowv and I just don’t see how that works. Unless we are possibly miscommunicating. There is AC going to the cathode in the form of the signal super imposed on the DC current but that is coming from the signal path, not the power transformer.

Power cables make no objective difference in the sound of a stereo but, as pointed out, they provide a lot of profit for the companies who make them. Same with interconnects, digital cables, etc.

@bruce19 cannot find a a schematic but:

on a 300B pins 1 and 4 are connected to the 5V AC of the transformer.  A pair of 50 ohm resistors are in series from 1 to 4.  The center tap is then connected to the main cathode bias resistor and its bypass cap to ground.