Bob,

I certainly agree that the NEC is _very_ clear on the position that it takes. I am curious as to _why_ the CMP takes this position.

'The rated input to the power conversion equipment' is a somewhat fuzzy number, kind of like the 'current rating' of a bit of wire. It is related to things such as how you define overload, how you define ambient conditions, and what you expect the life and reliability of the equipment to be. Take the same piece of hardware, but change the label to require operation in a 20C air-conditioned room rather than a 40C room, and the electronics will be able to tolerate higher operating currents.

The answer probably lies with UL requirements for how the power conversion equipment gets rated, and also with the fact that the manufacturer probably rates the drive under maximum load only, not under reduced load. In other words, the 'problem' created by MikeK's installs may not be a real safety issue, but instead a lack of data issue.

It may be the case that if the VFD drive manufacturer is willing to supply test data and a label that says:

with load XX the input current rating is YY
with load XXX the input current rating is YYY
with load XXXX the input current rating is YYYY
the maximum load served is XXXXX with an input current rating of YYYYY,

then using an oversized inverter with non-oversized wiring would be acceptable _under the current rules_. However I have never seen such a label, and the drive datasheets that I have looked at all specify maximum allowable currents.

Okay, here is a funny on the issue: If I have a VFD rated at 50A _at sea level_, and I install it in a mountain location at 9000 feet, and the manual specifies that the maximum allowable current is to be reduced by 10% for each 3000 feet, then what is the rated current number that I should use for the purpose of this code section? (Note: I cannot recall a drive manual that specifies this, however as I stated above the current handling capacity of the power electronics depends upon cooling, and I have seen manuals for systems that require derating with altitude because of reduced cooling. So this is a doubly hypothetical situation.)

Something that I've never done, but which really is the correct way to find the answer to this question: how do I trace back the history of a code section, including the CMP discussions?

-Jon