Equipment grounding conductor question

[pv_n00b]

In my case in point, it's a ~1200 UG run in PVC conduit with (of course) a wired ECG and the CCCs upsized from #2/0 to 1000kcmil aluminum (equivalent) for Vd, and I will be the PE of record. I am not comfortable with keeping the EGC at #6 copper. Would you be, and if so, can you show me why?

I am satisfied that upsizing the wire EGC when increasing the CC conductor size in an AC system is not required for the same reasons that upsizing the wired EGC is not required for CC conductor size increases for conditions of use. Show me why you don't need to upsized the wire EGC if the CC conductors have to be upsized for conditions of use?

 

[ggunn]

I am satisfied that upsizing the wire EGC when increasing the CC conductor size in an AC system is not required for the same reasons that upsizing the wired EGC is not required for CC conductor size increases for conditions of use. Show me why you don't need to upsized the wire EGC if the CC conductors have to be upsized for conditions of use?

That isn't my question and it's a different discussion altogether. My question from the beginning of this thread has been if (or why) the size of a wired EGC in an inverter output circuit can or should be a different size from a wired EGC in a load circuit. Under fault conditions the inverter shuts down and conditions for the conductors from the OCPD to the fault in both cases are identical in every way as far as I can tell. If I am incorrect, please show me the difference.

 

[wwhitney]

Show me why you don't need to upsized the wire EGC if the CC conductors have to be upsized for conditions of use?

Why as in code-wise? 250.122(B) explicitly excludes upsizing due to 310.15(B) and (C).

Why as in physics-wise? Upsizing for conditions of use is to control conductor temperature to avoid insulation damage over an extended period of time. Neither of those are concerns for the EGC--it should not be called on to carry current for an extended period of time, and it doesn't need to be insulated, either.

It is obviously true that for some long circuits, the EGC needs to be upsized. The trigger of the circuit conductors being upsized for voltage drop is just a proxy for length. An alternative formulation of 250.122(B) could be based on computing the EGC resistance, dividing that into the L-G voltage, and dividing that current by the OCPD rating, with a requirement that the result be some minimum, say 10.

Cheers, Wayne

 

[ggunn]

It is obviously true that for some long circuits, the EGC needs to be upsized.

Interpretational loopholes in the NEC notwithstanding, why only some if/when current carrying conductor sizing, conduit, and OCPD are the same?

 

[wwhitney]

Interpretational loopholes in the NEC notwithstanding, why only some if/when current carrying conductor sizing, conduit, and OCPD are the same?

By "some long" I just meant "long enough" without specifying what "enough" is.

Cheers, Wayne

 

[ggunn]

By "some long" I just meant "long enough" without specifying what "enough" is.

Cheers, Wayne

OK, but IIRC like me you believe that PV inverter circuits and load circuits should be treated the same.

 

[wwhitney]

OK, but IIRC like me you believe that PV inverter circuits and load circuits should be treated the same.

Certainly for the case where both are connected to the utility, the physics says that the same considerations apply.

But the wording of the 2020 NEC allows a difference in treatment, because of a dispute between the Article 250 CMP and the Article 690 CMP.

Cheers, Wayne

 

[jaggedben]

To summarize...

... My question from the beginning of this thread has been if (or why) the size of a wired EGC in an inverter output circuit can or should be a different size from a wired EGC in a load circuit. ...

...[Because] the wording of the 2020 NEC allows a difference in treatment, because of a dispute between the Article 250 CMP and the Article 690 CMP.

In my opinion (FWIW) both code panels are wrong. Upsizing should be required only where the CCCs were increased in size to account for voltage drop due to their length, for both types of AC circuits. Let the AHJ be the final judge of why conductors were upsized.

 

[ggunn]

To summarize...



In my opinion (FWIW) both code panels are wrong. Upsizing should be required only where the CCCs were increased in size to account for voltage drop due to their length, for both types of AC circuits. Let the AHJ be the final judge of why conductors were upsized.

But there's the rub, innit? Right or wrong, load circuits and PV inverter output circuits should be treated the same with respect to wired EGC sizing because in a ground fault condition they ARE the same.

 

[wwhitney]

Upsizing should be required only where the CCCs were increased in size to account for voltage drop due to their length, for both types of AC circuits. Let the AHJ be the final judge of why conductors were upsized.

That was the wording in 1999 NEC 250.122(B): "Adjustment for Voltage Drop. Where conductors are adjusted in size to compensate for voltage drop, equipment grounding conductors, where installed, shall be adjusted proportionately according to circular mil area."

I didn't check the PI/PC for the 2002 NEC change to 250.122(B), but I expect the concern was the frequency of interactions that went like this:

Inspector: Looks like you upsized these conductors for voltage drop, you need to upsize the EGC wire.
Electrician: No, it's just for future expansion, it wasn't voltage drop.

So I don't think the CMP wants to put the inspector/AHJ in the position of determining why the wire was upsized.

Also, while voltage drop is a close proxy for the issue that would require upsizing a wire-type EGC, it's not exact. You could have a situation where the minimum size wire gives a voltage drop of say 2.5%, and you have some performance reason that you need to keep voltage drop below 2%. A normal size EGC will still be adequate.

Cheers, Wayne

 

[jaggedben]

That was the wording in 1999 NEC 250.122(B): "Adjustment for Voltage Drop. Where conductors are adjusted in size to compensate for voltage drop, equipment grounding conductors, where installed, shall be adjusted proportionately according to circular mil area."

I didn't check the PI/PC for the 2002 NEC change to 250.122(B), but I expect the concern was the frequency of interactions that went like this:

Inspector: Looks like you upsized these conductors for voltage drop, you need to upsize the EGC wire.
Electrician: No, it's just for future expansion, it wasn't voltage drop.

So I don't think the CMP wants to put the inspector/AHJ in the position of determining why the wire was upsized.

Yeah, I've heard that argument, and considered it, and I still think it's the wrong approach. The NEC gives AHJs lots of authority to interpret the facts of installations in all sorts of places and I don't see why this should be different. E.g. should the code spell out what is 'subject to physical damage' because installers and inspectors might have disagreements in the field?

I mean, conversely to your scenario, when it's obvious to everyone that voltage drop is not the reason, then the AHJ shouldn't be pushed to enforce this rule. For short runs the future expansion justification ought to be a slam dunk. I mean, if we had to upsize an EGC every time someone ran a 100A feeder to a subpanel that has less than 50A of calculated load... Thankfully, I don't see this in the field. I'd like to think it's selective enforcement because the inspectors also realize how stupid it is, although probably it's just unawareness.

Perhaps it calls for an IN with some real engineering info to guide AHJs on the length considerations. Or alternatively, I might settle for the entire upsizing rule not to kick in until a certain length of conductor is reached.

Also, while voltage drop is a close proxy for the issue that would require upsizing a wire-type EGC, it's not exact. You could have a situation where the minimum size wire gives a voltage drop of say 2.5%, and you have some performance reason that you need to keep voltage drop below 2%. A normal size EGC will still be adequate.

That's at least related to why I stipulate in my formulation that the *length* of the conductors must be part of the reason. Voltage performance reasons not really related to length shouldn't count.

I'm open to a better formulation, although doubtless changing either CMP's mind at this point will be difficult. My real point, again, is simply that while in reality a certain amount (a lot!) of voltage drop in a fault is the only valid reason to require EGC upsizing, both CMPs get that wrong; the 250 CMP by sweeping up everything that isn't voltage drop, and the 690 CMP by explicitly exempting voltage drop.

 

[wwhitney]

Yeah, I've heard that argument, and considered it, and I still think it's the wrong approach.

Well, one feature that the current wording (with exception) has over the 1999 wording is that the former makes a presumption that the EGC should be upsized, but permits the installer to rebut that under the exception. While the 1999 wording may be taken as putting the onus on the inspector to show that the reason for the upsizing was voltage drop, perhaps by computation.

I think that if there are any computations required, the onus to do them should be on the installer to provide, with the inspector verifying/accepting them.

Cheers, Wayne

 

[jaggedben]

Yeah having the current exception in 250 is definitely better than not having that exception. I disagree that anyone should ever have to provide calcs for a 10ft long circuit but at least as long as AHJs aren't asking for it then it's not a big deal to me.

 

[ggunn]

Just to play devil's advocate, what difference does it make why the CCCs are being upsized? Isn't Job One for the OCPD to be able to clear a fault irrespective of why the CCCs are larger? Is that what pv_n00b was saying?

 

[pv_n00b]

I have NECs back to 1984 and funny enough in 1984 it says the EGC only needs to be upsized when the conductor is upsized for Vd. No other increase in the CCC size gets an EGC upsize. It stayed this way until the 2002 version. In 2002 it was changed to say that any increase in the CCC size required an increase in the EGC size. It stayed that way until 2020 when the conditions up use exemption was added. So the exemption has not been in there long.
Some big things happened in the revision for 2020. Look in TerraView for some interesting reading. There as a plan to remove the increase in size requirements all together and change table 250.122 to be based on the CCC size and not the upstream OCPD rating. That idea was squashed between the first and second draft. In the second draft the increase in size was back with the exemption for conditions of use and table 250.122 was returned to being based on OCPD rating. I've attached a snip of one of the CMPs comments on the subject. Go over and read the comments on the second draft.
As for the addition of the exclusion for a size increase for conditions of use, the CMP comment just says that and increase in CCC size for conditions of use do not require an increase in the EGC size. No explanation why.

 

[wwhitney]

As for the addition of the exclusion for a size increase for conditions of use, the CMP comment just says that and increase in CCC size for conditions of use do not require an increase in the EGC size. No explanation why.

I gave you an explanation why, it's fairly straightforward.

Cheers, Wayne

 

[jaggedben]

Just to play devil's advocate, what difference does it make why the CCCs are being upsized? Isn't Job One for the OCPD to be able to clear a fault irrespective of why the CCCs are larger? Is that what pv_n00b was saying?

Well... Exactly.

Since upsizing for voltage drop is never code required, the reality of the situation is this:
1) Optionally upsizing the CCCs without upsizing the EGC will always reduce impedance in a fault and will never be worse than if the CCC and EGC were code minimum. So penalizing those who upsize CCCs optionally is illogical.
2) If the voltage drop under fault conditions is so bad on a long circuit that it affects clearing a fault, then something needs to be upsized to help clear the fault even if avoiding voltage drop isn't important for equipment functioning. And it doesn't necessarily matter if the CCC's or EGC is the thing upsized to help clear the fault. But the rule isn't written like that at all whatsoever.

I don't remember if I've said it this thread yet, but I've said it other threads: Dumbest rule in the book.

 

[jaggedben]

I'll just add....
If we can have more or less completely arbitrary 10ft and 25ft tap rules based on some notion of better fault clearing being suddenly being required at 10'-1/8", then we could just as well have a 250.122(B) rule that kicks in at some arbitary circuit length where we decide to draw the line.