I agree with Jafox that you should consult with Pass Labs. But FWIW let’s look at the specifics:
Gs5556 makes a good point in referring to the significance of the speaker’s impedance characteristics. And I see in one of your other recent posts that you are using Harbeth 40.2s. In terms of relevant specs the 40.2 appears to not be greatly different than the 40.1, for which Stereophile measured a sensitivity of 84.5 db/2.83 volts/1 meter, and stated that:
... the Harbeth is easy to drive in terms of its demand for current. The impedance drops below 6 ohms only in the middle of the midrange, and then only for a short while; and while the electrical phase angle occasionally reaches extreme values, this is only when the impedance magnitude is high, ameliorating the drive difficulty.
The relevant specs of the amp are:
Max power (8 ohms): 260 watts
Max power (4 ohms): 520 watts
Leaves class A at: 34 watts peak
Power consumption at idle: 3.1 amps, 375 watts
The 375 watt figure is low enough that I would have high confidence that turn-on surges will be reasonable and won’t trip the breaker, especially given that both amps will not be switched on at the same instant.
If we assume a 6 ohm speaker impedance the amp’s maximum power capability with these speakers can be calculated to be 347 watts. Given that the amps would be operating mainly in class AB at power levels approaching that amount I suspect that a safe estimate of the maximum amount of AC power each amp would ever draw with your speakers would be in the range of 800 to 1000 watts. So the total worst case draw of the two amps plus a preamp and say a couple of source components should fit within the 20 amps x 120 volts = 2400 watt capability of a single dedicated line, although without a great deal of margin if and when the amps are ever running close to full power.
From a sonic standpoint, though, the problem may be that since the amps will draw most of that AC power during just a small fraction of a given 60 Hz period, the total current drawn during those brief instants may be significantly more than 20 amps. Not enough to trip the breaker, given the short duration, but conceivably the resulting drop in line voltage during those instants could affect dynamics or have other sonic consequences, especially if the length of the run is long and/or if the gauge is not over-specified.
So the bottom line seems to be that you would probably be ok installing just a single dedicated line, which as has been mentioned would minimize the risk of ground loop issues, but it’s somewhat marginal and might not be sonically optimal, especially if you listen at high volume levels. Again, check with Pass.
Good luck. Regards,
-- Al