Why does EGC size depends on OCPD (breaker) size?

[alej27]

In the TN-C-S grounding system used in electrical installations in the US, one of the main functions of the an EGC (equipment grounding conductor) is to serve as a back-up neutral wire, so that in the event of a line-to-case fault, a complete circuit is formed with low impedance to trip the breaker and clear the fault, thanks to the neutral-to-case bond at the main disconnect. Obviously this wire serves other purposes.

But in the previous purpose, a very high short-circuit fault current flows through the EGC. It's value depends on the impedance of the whole path and the voltage of the source. So we should make sure that the EGC doesn't melt or suffers insulation damage due to the high current during the very short duration of the fault. Hence intuitively, we should sice the EGC based on the available/prospective short-circuit current.

My question is why is the minimum EGC size required by the NEC to be based on the breaker size, and not on the short-circuit current?

Is it to simplify the process instead of performing a whole short-circuit analysis? But I don't think so, since the NEC already requires to consider the short-circuit current when selecting the interrupting rating of a breaker.

Is it for aesthetics, since big phase and neutral wires with a thin ground wire would look incorrect? I don't think so, since as we know the NEC is published by the NFPA, which is basically concerned with preventing fires. For that matter, the NEC doesn't even require voltage drop calculations, which are important but would not cause a fire if conductors and OCPD are properly sized.

So what's the reason?

 

[David Castor]

The important requirement is that the EGC have a low enough impedance to pass enough ground fault current to trip the breaker on its magnetic (or "instantaneous") trip. Larger breakers require more fault current to get into the instantaneous range, hence the larger size conductor required.

Below a certain size the physical strength of the wire becomes a consideration, so #18 or #16 ground wires are not allowed, even though they might be adequate in terms of circuit resistance for 15 or 20 A circuits. (I don't know). I'm sure there's a long history behind the EGC sizes in the NEC, but that's the primary concern.

Whenever I've looked at the issue of conductor damage during through faults due to conductor heating, it's almost never problem. The breaker or fuse will trip well before the cable damage curve is reached, especially for low voltage circuits where the wire impedance has a major impact on the fault current.

 

[tom baker]

The NEC is a prescriptive code, meaning follow the rules to comply with the performance rules for grounding. Section 250.122 has had many changes but the recent language is pretty good
Suggest you download GEMI from the steel tube institute, to see about sizing the EGC.
An excellent resource on the size of EGC is the Green Book by IEEE