I take it you are referring to 250.122(B) for the EGC upsizing. I'm trying to understand the technical reasons for this.
First, 250.122(B) excludes ampacity adjustment factors. But it seems to me that when it is necessary to the EGC be sized for no less than the OCPD rating, a condition that would make that OCPD related sizing crucial (e.g. some fault condition is pulling full load current through the EGC), the thermal aspects in the raceway would still be effectively the same. I don't understand why this exception exists.
However, in the case of a long circuit being upsized only for a voltage drop condition, I don't see an issue like that. But maybe it could be the case that a fault to EGC could fail to open the OCPD where the circuit length has to much resistance?
Of course there is the concern that if upsized conductors are present at the circuit origination, some handyman might think he can upsize the OCPD. That risk is also present if the branch circuit changes to a smaller size at a later point.
What about cases where the neutral is oversized for reasons of harmonic mitigation? Does the EGC need to match the neutral in that case?
