EGC required in all non metallic conduits?

[Jesse7623]

I have a question I never thought I would have to ask here.

I have 3 solar arrays recently built by a reputable contractor that have no EGC in any of the string wire conduits. This is a ground mount array, fixed tilt with Chint String inverters back at the pad area.
Every row has a GEC conductor running back to the 4/0 ground ring at the pad, and the EC is claiming that this can substitute for the EGC in each conduit. There is an EGC in the AC feeder pipe to the inverters, from the Main disconnect at the pad.
I have wired thousands of structures, built over a GW of solar projects all over the US, hold a Master Electricians license in MA, NH, DE, UT and never have I ever seen an EC not run the EGC in the pipe with the branch circuits (unless the conduit can serve as the grounding means).

The EC is citing 690.43(C)

Isn't there a clear code requirement that every raceway that has CCC must have a grounding conductor in it? Or have we all just been wasting wire for decades now?

thanks in advance for all your advice.

 

[Jesse7623]

There’s been discussion between the construction team and the engineer of record (EOR) for the project. The EOR believes that the site is compliant with NEC 2017 as-is. I pushed back for additional information and this is what I’ve received so far. I’ve still got a few open questions with the EOR for clarification, so I’ll keep you looped in on responses.
The EOR is neither a licensed electrician nor the AHJ so their opinion doesn’t mean much here. And why did they run the EGC on 4 of the 7 sites and not on 3?


NEC 690.43(C) states “ ….otherwise run with the PV system conductors where those circuit conductors leave the vicinity of the PV array.
I interpret this as the egc does not need to run in the same conduit but can share the same trench.
"Raceway, Cable or otherwise run with the PV array conductors." That last part is talking about direct burial of PV conductors. There shall be an EGC in the same trench. Moreover, the only other grounding conductor we know of is the GEC so the excuse there is moot.
In addition, NEC 690.43 also references 250.134, See exception 2 for DC circuits. It allows the egc to be run separately from the circuit conductors. Extraneous but OK

 

[jaggedben]

With GES connections going to the pad it may not matter much practically, but it is definitely a code violation, not to mention pathetic IMO. Refer to 250 or 300 for more clarity. 'Otherwise run with' would only apply if there wasn't a raceway, because the raceway is explicitly mentioned, and there are raceways, and per your subject line the raceways are not the EGCs.

It is important to understand that DC ground fault detection is done at each inverter and that a connection to earth itself does not accomplish that, and that a roundabout ground connection that might get removed when working on something unrelated would therefore also compromise that function. Hence the code rule.

 

[jaggedben]

250.134 contains the basics for running EGCs with circuit conductors.. 300.3(B) reiterates. Also 250.121 prohibits using a GEC as an EGC, with an exception

I have no idea why dc circuits are treated differently.

 

[Jesse7623]

250.134 contains the basics for running EGCs with circuit conductors.. 300.3(B) reiterates. Also 250.121 prohibits using a GEC as an EGC, with an exception

I have no idea why dc circuits are treated differently.

I had forgotten about 300.3(B). Thank you for your replies.

 

[electrofelon]

Two comments:

1. Doesn't it come down to how one interprets what the "the vicinity if the array" is?

2. It is not clear to me that the conductor you are calling a GEC is indeed a GEC. It doesn't sound like one to me.

 

[Jesse7623]

Two comments:

1. Doesn't it come down to how one interprets what the "the vicinity if the array" is?

2. It is not clear to me that the conductor you are calling a GEC is indeed a GEC. It doesn't sound like one to me.

1.) Yes, but have you ever just skipped running the EGC in a grouping of conduits in a duct bank because, "well a pipe in a nearby trench has a ground rod with a 2/0 running to a sectionalizing cabinet nearby so that should cover it"? Probably not.

2.) The bare 4/0 copper ground ring and all bare copper wires that go to the foundations of the solar array (from the ground ring around the pad) are by nature considered part of the GEC system

 

[electrofelon]

1.) Yes, but have you ever just skipped running the EGC in a grouping of conduits in a duct bank because, "well a pipe in a nearby trench has a ground rod with a 2/0 running to a sectionalizing cabinet nearby so that should cover it"? Probably not.

2.) The bare 4/0 copper ground ring and all bare copper wires that go to the foundations of the solar array (from the ground ring around the pad) are by nature considered part of the GEC system

1. No, I don't recall that specific situation happening, but for DC I have no trouble not grouping the EGC with all it's circuit conductors if it's code compliant.

2. I'm not sure I agree. Sounds like at least some of the stuff you mention is not a NEC grounding electrode (and probably a case of some idiot designer wasting money and resources) and if that's the case, no it wouldn't be a GEC. But we would need more details to be sure.

 

[Jesse7623]

1. No, I don't recall that specific situation happening, but for DC I have no trouble not grouping the EGC with all it's circuit conductors if it's code compliant.

2. I'm not sure I agree. Sounds like at least some of the stuff you mention is not a NEC grounding electrode (and probably a case of some idiot designer wasting money and resources) and if that's the case, no it wouldn't be a GEC. But we would need more details to be sure.

1.) So absolutely, where we are doing direct burial, we will lay one EGC in the trench with 30 sets of feeders. and that is 100% NEC compliant. But where they go through their piping, up into the inverter, each conduit always has an individual EGC which gets crimped to the main trench EGC. I have built over a GW of solar all over the US and every single one is done like this to be code compliant. I am fairly certain that Mortenson and Bechtel and Invenergy aren't all getting it wrong and wasting millions of dollars on spare/ unneeded wire.

2.) Definitely not NEC required but where the EOR decides to run a bare wire from the ground ring, to the foundations of the array and bond them thus; we can in no way qualify that as a EGC. It is clearly a GEC.

 

[electrofelon]

2.) Definitely not NEC required but where the EOR decides to run a bare wire from the ground ring, to the foundations of the array and bond them thus; we can in no way qualify that as a EGC. It is clearly a GEC.

I just can't agree with that. If it's not an NEC required and/or compliant grounding electrode, you might as well call the conductor feeding it a pink flamingo. You may be able to use it as an NEC compliant EGC, but maybe not. Depends on the specifics.

 

[Jesse7623]

I just can't agree with that. If it's not an NEC required and/or compliant grounding electrode, you might as well call the conductor feeding it a pink flamingo. You may be able to use it as an NEC compliant EGC, but maybe not. Depends on the specifics.

I am not sure how familiar you are with utility scale solar. The foundations for many racking systems are UL 1547 listed grounding systems with the foundations being considered grounding electrodes. If I bond one Grounding electrode to another, the conductor connecting the 2 is inherently what?

 

[electrofelon]

I am not sure how familiar you are with utility scale solar.

Very. I did it for years.

I am not not aware of a UL 1547. Besides, I don't see how a standard other than the NEC can declare something a grounding electrode that the NEC doesn't consider one.

 

[pv_n00b]

I am not sure how familiar you are with utility scale solar. The foundations for many racking systems are UL 1547 listed grounding systems with the foundations being considered grounding electrodes. If I bond one Grounding electrode to another, the conductor connecting the 2 is inherently what?

There is no UL 1547. Grounding electrode systems are defined in the NEC and there is no need to list an electrode for it to be a grounding electrode. system. A foundation can be designed to be a UFER and the conductor connecting all the UFERs together and to other grounding electrodes is a grounding electrode bonding jumper and it is installed using the same requirements as a GEC, but is not a GEC. A GEC is a very specific conductor that connects the system grounded conductor or the equipment to the grounding electrode system.

 

[electrofelon]

Lots of confusion with this arises on utility scale PV systems as the engineers and designers typically go crazy with the grounding and bonding. Lots of it is not required by the NEC so there is really nothing that it has to comply with other than the design documents.

 

[pv_n00b]

Lots of confusion with this arises on utility scale PV systems as the engineers and designers typically go crazy with the grounding and bonding. Lots of it is not required by the NEC so there is really nothing that it has to comply with other than the design documents.

I remember a large ground-mounted array over a decade ago that the engineers had put a grounding mat design for a substation in for the whole array area. Must have cost a fortune.

 

[Jesse7623]

Lots of confusion with this arises on utility scale PV systems as the engineers and designers typically go crazy with the grounding and bonding. Lots of it is not required by the NEC so there is really nothing that it has to comply with other than the design documents.

Very very true. And while it may help reduce step voltage, it wreaks havoc with attracting lightning and destroying all the underdesigned comms/WS/DAS equipment.

 

[Jesse7623]

There is no UL 1547. Grounding electrode systems are defined in the NEC and there is no need to list an electrode for it to be a grounding electrode. system. A foundation can be designed to be a UFER and the conductor connecting all the UFERs together and to other grounding electrodes is a grounding electrode bonding jumper and it is installed using the same requirements as a GEC, but is not a GEC. A GEC is a very specific conductor that connects the system grounded conductor or the equipment to the grounding electrode system.

You're right. I was using the wrong numbers while thinking of IEEE 1547 and meant UL 1741 and UL 2703

 

[electrofelon]

You're right. I was using the wrong numbers while thinking of IEEE 1547 and meant UL 1741 and UL 2703

I am certainly no expert on UL 1703, but I think that is more about all the bonding between the racking structure components and modules. Even if it does cover earthing, they don't have any authority to designate grounding electrodes or when earthing is required. One final note, it depends on the code cycle whether a grounding electrode is even required at an array so that is another thing to throw in the mix.

 

[wwhitney]

One final note, it depends on the code cycle whether a grounding electrode is even required at an array so that is another thing to throw in the mix.

But for ground mount, surely the array is a structure, and the power conductors (DC or AC) going to/from it are feeders, so the array requires a GES per 250.32(A), yes?

Cheers, Wayne

 

[electrofelon]

But for ground mount, surely the array is a structure, and the power conductors (DC or AC) going to/from it are feeders, so the array requires a GES per 250.32(A), yes?

Cheers, Wayne

Well then we have the change of the definition of structure where the words "other than equipment" were added. The answer also would depend on whetjer or not AC or DC were run to the array.