PV Grounding - AHJ interpretation

[kanaka360]

I recently installed a 1kW array on my home, and had the rough-in inspection today. It is a microinverter based system, using listed grounding washers as equipment grounding conductors for connections between modules, racking, and microinverters. From the microinverters, #12 (20A breaker) conductors run through the attic in romex, and down the exterior of the house to the AC disconnect in conduit.

The AHJ has requested that the EGC be minimum #6, all the way from the inverter to the grounding electrode, based on 690.46. I'm not getting my hopes up, but I intend to try to argue this with the inspector, as he was not familiar with the requirements and had to look them up on-site during the inspection. 690.46 references 250.120(c), which I think makes it fairly clear that the EGC size is appropriate, and I think I can argue this point. However, he also wants the EGC to be connected directly to the ground rod and will not accept the connection to the main switchboard ground bus as it stands now. I have no idea where he's getting this, anyone run into this requirement before?

Unfortunately this caught me off guard and I did a fairly poor job of arguing this at the time of the inspection.

 

[Smart $]

... It is a microinverter based system...

What microinverters?

If they happen to be Enphase M250/215...

http://enphase.com/global/files/Enp...d-Ungrounded-Arrays_Whitepaper_English_NA.pdf

 

[GoldDigger]

He is probably, whether he knows it or not, assuming that the DC EGC is also serving as the DC GEC and is sizing it on that basis. Smart$'s link explains why this does not always apply.

 

[SolarPro]

We recently published a Grounding Compendium article for PV systems that might be of some assistance as well:

Ken Gardner, the principal engineer at Gardner Engineering Alternative Energy Services, explains his design approach for Enphase microinverter systems: ?We like to use the SnapNrack mounting system because then we can use WEEBs to bond the solar modules and microinverters. We then run 6 AWG solid copper wire between WEEB bonding lugs on the rails to a SolaDeck junction box, at which point we transition to 8 AWG stranded wire using an irreversible splice. The 8 AWG ground is required because this is the minimum size conductor allowed for use as a dc GEC for the microinverters per Section 250.166. The 8 AWG ground also serves as the EGC and is therefore bonded to the rooftop junction box. We install the 8 AWG ground with the current-carrying conductors?typically 8 or 10 AWG for residential systems?in metal flex, usually run to the main service panel where we land the ground on the main grounding lug per Section 690.47(C)(3). In theory, we could use Type MC cable, but the grounding conductor inside is usually a size smaller than the main current-carrying conductors and is too small to serve as a GEC.?

 

[kanaka360]

Yes, they're enphase 215 microinverters.

I spoke with him again, and he's interpreting 690.47(a) to mean that grounding electrode conductor (installed in accordance with 250.64) has to run from the array, to the grounding electrode. I don't think there's much arguing on this one. That's not what 690.47(a) says, but he seems to think it does.

 

[Smart $]

Yes, they're enphase 215 microinverters.

I spoke with him again, and he's interpreting 690.47(a) to mean that grounding electrode conductor (installed in accordance with 250.64) has to run from the array, to the grounding electrode. I don't think there's much arguing on this one. That's not what 690.47(a) says, but he seems to think it does.

Since you are using GTI's, the GES referred to in 690.47(A) should already be established.

Nevertheless, see 690.47(D) [2014].

I have yet to form any conclusion on what 690.47(D) Exception No. 2 is saying.

An additional array grounding electrode(s) shall not be required if located within 1.8 m (6 ft) of the premises wiring electrode.

If you have an electrode within 6' of any premises wiring electrode, isn't it an additional electrode?

 

[ggunn]

Never mind.

 

[Smart $]

...
I have yet to form any conclusion on what 690.47(D) Exception No. 2 is saying.

An additional array grounding electrode(s) shall not be required if located within 1.8 m (6 ft) of the premises wiring electrode.

If you have an electrode within 6' of any premises wiring electrode, isn't it an additional electrode?

Perhaps they mean if the array is located within 6'...

???

 

[jaggedben]

Since you are using GTI's, the GES referred to in 690.47(A) should already be established.

Nevertheless, see 690.47(D) [2014].

I have yet to form any conclusion on what 690.47(D) Exception No. 2 is saying.If you have an electrode within 6' of any premises wiring electrode, isn't it an additional electrode?

My best guess at their intention: if the premises electrode is within 6' of where the additional electrode is required to be (i.e. 'as close as practicable to the location of roof mounted PV arrays), then the additional electrode is not required.

BTW, I won't let this section be mentioned without reminding everyone that our host here, Mr. Holt, has called for a Temporary Interim Ammendment to remove this section from the code because he believes it's dangerous. Google 690.47(D)anger

 

[Smart $]

My best guess at their intention: if the premises electrode is within 6' of where the additional electrode is required to be (i.e. 'as close as practicable to the location of roof mounted PV arrays), then the additional electrode is not required.

No matter the intention, the wording could be better.

BTW, I won't let this section be mentioned without reminding everyone that our host here, Mr. Holt, has called for a Temporary Interim Ammendment to remove this section from the code because he believes it's dangerous. Google 690.47(D)anger

I don't have to look it up to agree that it could be dangerous.

 

[cowboyjwc]

Do you have approved plans for your install?

 

[petersonra]

I don't have to look it up to agree that it could be dangerous.

While I am inclined to agree that there is potential for this to be dangerous, I fail to see how it is anymore or less dangerous than running RMC underground and coming up inside a building. Or having any underground metallic object extend into the building. Like a water pipe. Or a piece of structure.

 

[Smart $]

While I am inclined to agree that there is potential for this to be dangerous, I fail to see how it is anymore or less dangerous than running RMC underground and coming up inside a building. Or having any underground metallic object extend into the building. Like a water pipe. Or a piece of structure.

The key word in my comment is "could"... and exactly why I stated it that way.

 

[jaggedben]

While I am inclined to agree that there is potential for this to be dangerous, I fail to see how it is anymore or less dangerous than running RMC underground and coming up inside a building. Or having any underground metallic object extend into the building. Like a water pipe. Or a piece of structure.

It seems to me that the danger is having multiple electrodes (whether intended to be electrodes or not) that are bonded above ground. Simply having a metallic object extending into a building doesn't create that, if say it just comes up to a box and doesn't bond to anything else. 690.47(D) essentially requires the danger in any situation where the array is not on the same side of a building as the existing electrode.

There's also a substantial difference between the code permitting something that might, in a fluke, be dangerous, and the code have a requirement to do much the same thing on 1000s of systems people are installing.

 

[Zee]

Yes, they're enphase 215 microinverters.

I spoke with him again, and he's interpreting 690.47(a) to mean that grounding electrode conductor (installed in accordance with 250.64) has to run from the array, to the grounding electrode. I don't think there's much arguing on this one. That's not what 690.47(a) says, but he seems to think it does.

The inspector is remembering the way it used to be, with the older Enphase inverters. In that regard he WAS correct. Prosolar's article pertains to that also.
However, if it is the newer M215 with INTEGRATED GROUND, then absolutely no GEC is required. (it should come w/o ground lugs mounted on inverter)

http://enphase.com/microinverters/#m215
Just click on datasheet link above and print and show to inspector. Enphase makes it very clear. Pretty much the whole point of that datasheet is to clarify this often misunderstood issue!

"......DC circuit is isolated and insulated from ground, so no Ground Electrode Conductor (GEC) is required for the microinverter. This further simplifies installation, enhances safety, and saves on labor and materials costs."