Scott,
I disagree with you on three points.
The 12 Amp load, in one part, was to figure if at 40 feet, there was any problem with Voltage Drop in the Branch Circuitry; one that would create additional heat, and alter the Ambient Temperature to something above 30°C. This is Design Issue #1.
I don't see how, for the purpose of ampacity derating, voltage drop creates any additional heat that will change the ambient temperature. If you have current flowing in a wire, you will get a certain amount of voltage drop per foot of wire and a certain amount of heat produced per foot of wire. If the wire were 100x longer, you would get 100x the voltage drop, and 100x the heat produced, but 100x the surface area in which to dissipate the heat, so the ampacity stays the same. The only way I could see additional heating with voltage drop is if the load were to draw more current as the supply voltage decreases.
The "THD" thing would become a real hard issue if it was higher. If THD = 20%, then this would be an additional 2.4 Amperes per Line, and an additional 7.2 Amperes in each Grounded Conductor.
Now we would have 14.4 Amperes per Line, and 19.2 Amperes per each Grounded Conductor.
Figuring this will be an LCL situation (Continuous Load), the Grounded Conductors - if sized to be #12, would be overloaded.
I believe that a #12 conductor in the situation described is _not_ considered overloaded, even if this is an LCL situation. Everything that I have seen that requires a differentiation between short term loads and continuous loads is related to the OCPD. The ampacity of a conductor (as calculated by 310.16 and related) is the _continuous_ ampacity, and in the situation described a #12 could carry 21A. However the OCPD can not be permitted to be loaded at more than 80% of trip rating for an LCL, unless the OCPD is rated for 100% operation, and in addition #12 conductors are required to be protected by a 20A or smaller OCPD. But the neutral doesn't usually have an OCPD, and so should be within ratings up to 21A. Note that if there were a breaker on the neutral, I would agree that it would be overloaded at 19.2A
FYI: as mentioned, Harmonic Distortion is not caused by the Electrical Distribution Equipment, but results from the Load Equipment Connected to the Electrical System.
Actually, harmonic distortion _requires_ interaction between the Electrical Distribution Equipment and the load. If you had a zero impedance distribution system that always provided exactly the proper voltage, then any load non-linearities would result in significant harmonic _current_ flows, but the system voltage would remain undistorted, and other loads on the system would not see harmonic distortion. The harmonic current flows would cause heating in conductors, of course. If the distribution system has high impedance where current flow is limited, then load non-linearities will result in less harmonic current flow, but only because supply voltage distortion will tend to reduce the voltage driving the harmonic current flows. Other loads on the system see distorted supply voltages in this case, but harmonic heating of conductors is reduced.
-Jon
