I saw jneutron take a stab at it too over on another forum. His argument was that a power cord with significantly different impedance than your house wiring could reflect RFI back into said house wiring, instead of allowing it into your equipment.
Perhaps he will comment further?
Certainly.
What was claimed was that the parametrics of the cord helped keep "bad stuff" out. What they did was lower the characteristic impedance of the cable such that any high speed stuff coming down the romex was attenuated by reflection.
The claim itself was accurate from an E/M theory point of view. Whether or not there was a need for such action in a normal house was not considered by myself. High slew rate events, such as motor contactors, dimmers, etc, can locally produce such things. I believe it is better to design the equipment to filter out such stuff. The best option for a diy'er is to buy an IEC chassis widgit which has an integrated grounding bond, an integrated fuse, and an integrated fully rated filter section for common and non common mode hash.
There are two entities being affected by the power cords.
1. Allowing line to neutral hash entry into the equipment via the power cord itself. Filtering is the best option here. Either at the hash source, or at the equipment. A fancy line conditioner/UPS is also an option, albeit an expensive one. The assumption is that the equipment is not adequately designed to do this by itself.
2. Allowing the intrusion of ground loop currents by way of the IC shields in conjunction with the two power cords. This is much more difficult for the general user to understand, as the current will depend on the source location, the loop resistivity, the loop sectional area, the amplifier's susceptibility to ground loop current induction into the signal and/or ground path, and the generation of the currents due to amplifier power draw with haversine pulses. A general rule here is to keep the loops as small as possible.
Cheers, John
