To atmasphere: I’m puzzled by your comment that this Nelson Pass quoted 48 Amps is “how much current is present when the power supply of the amp is shorted. It’s available to the output section in the form of charged capacitance. This is used to help reduce IMD in the output section and is not something that the output section can pass to the loudspeaker.” Perhaps I failed to see what you did there; but I thought that the extended time that my tube amp can drive my speakers from the source when I turn the tube amp off (while the source is still playing) was, in fact, the capacitors discharging through the speakers. At least that was a comment I remember John Atkinson making in a Stereophile review of a tube amp that I read online. Other contributors to this forum also seem to assume that the capacitor banks are there to help the amplifier deliver transients to the speakers outside the range of their normal current demands. I’m not (yet) experienced in building amplifiers, but I am intensely interested in understanding how they work. If, as you say, the speakers are in series with the amp’s output terminals, and I never considered them that way before, but I guess they are, then they would definitely need current to drive them. Of course, your linked discussion clarifies their rather intimate relationship (you can’t have one without the other [current without voltage] if power is being produced or consumed). Further, the ‘transients’ at lower frequency, i. e., the ‘attack’ of a bass drum strike would require more power than a ‘transient’ of a high-hat cymbal strike because its relative duration would be longer. Just thinking out loud, here.
On second thought, if ‘what you were doing there,’ is saying that if the power supply is being shorted to produce 48 Amps; there isn’t going to be anything near 48 amps at the output terminals —at the same moment in time; then I totally get ‘what you did there’ —you made a joke, which this discussion sorely needs. 👍
Alternatively, if you are saying that 48 Amps cannot be used both for Intermodulation distortion AND output at the speaker terminals, then I accept that as well, as current can only be ‘consumed’ once in any circuit, I suppose.