disconnecting one leg of inverter output as RSD initiation

wwhitney said:
Do you have a PV system disconnecting means at the array?

If so, then the 2020 and 2023 NEC make it clear 690.12 doesn't extend beyond the PV system disconnecting means, so your feeder from the building to the PV system disconnecting means would not be subject to RSD. While you've specified the 2017 NEC, sounds like there is some ambiguity about whether 2017 NEC 690.12 applies to AC circuits at all, so it seems reasonable to take advantage of that to reach the same conclusion as for the 2020 and 2023 NEC.

Cheers, Wayne
Click to expand...
No PV disconnect at the array.
 
tallgirl said:
I'm not convinced that's true because the issue has been conductors which are energized EVEN WHEN something like the main disconnect is opened. That's what firefighters have dealt with, since I believe it's common practice to de-energize a building when they start firefighting.
Click to expand...
What seems strange to me ( going a bit off topic here), is that the RSD doesnt need to be grouped with the service disconnect. Dont you want ALL switches that shut off power to the building grouped together? Also, 2017 doesnt require the service disconnect to be outside, but the RSD switch does? Why?
 
electrofelon said:
No PV disconnect at the array.
Click to expand...
And at the building, you don't want the PV system disconnecting means on the outside before the inverter output conductors enter the building?

So you are taking the position that the entire array is "equipment," so it is not a structure, so it does not require a 225.31 disconnecting means?

Cheers, Wayne
 
electrofelon said:
What seems strange to me ( going a bit off topic here), is that the RSD doesnt need to be grouped with the service disconnect. Dont you want ALL switches that shut off power to the building grouped together? Also, 2017 doesnt require the service disconnect to be outside, but the RSD switch does? Why?
Click to expand...
That's another bugger in all of this, but again, I think what prompted a lot of this was things like rooftop array fires where firefights got up near a roof, only to discover it was still energized "somehow", so perhaps not all that important the RSD be grouped with the other disconnects.

I did work for a company which acquired yet another company that was responsible for some fires. We discussed fire a lot there ...
 
wwhitney said:
And at the building, you don't want the PV system disconnecting means on the outside before the inverter output conductors enter the building?

So you are taking the position that the entire array is "equipment," so it is not a structure, so it does not require a 225.31 disconnecting means?

Cheers, Wayne
Click to expand...
I have two inverter circuits coming from the array, and I will be installing a 4 pole switch on the exterior of the building. My question is somewhat academic, but also thinking about a system with more circuits where you run out of poles and if you can just kill one leg.

225.31 is another wrinkle but in my experience, it seems almost everyone sees it has not required.
 
tallgirl said:
I'm not convinced that's true because the issue has been conductors which are energized EVEN WHEN something like the main disconnect is opened. That's what firefighters have dealt with, since I believe it's common practice to de-energize a building when they start firefighting.
Click to expand...
Sorry, what's the contradiction between what I said and what you're saying here?

electrofelon said:
I am not sure that is true:

(A) Controlled Conductors. Requirements for controlled conductors shall apply to PV circuits supplied by the PV system.

...logically, it makes sense to be concerned with the conductors that are NOT de-energized by turning off the service disconnect.
Click to expand...
To be clear, while this is a bit vague in the language you're quoting from 2017, it's clear in subsequent codes, which better cover what I understand to have always been the intent of 690.12.

Under the 2017 you can argue, if you like, about whether microinverter output circuits are subject to the requirements. But if you decide they are, then the test for compliance simply requires voltage between conductors and to ground be limited within 30secs etc.. At that point the requirement purely deals with voltage and not where that voltage comes from. In no case would a 1-pole disconnect for a circuit with 2 hots in a 120/240V ever create compliance for such a circuit. Either no additional disconnect was needed in the first place, or a 2-pole disconnect is required.

There is a double intent behind 690.12 and its companion labeling requirements
1) To reduce shock hazard for firefighters working on or around PV arrays or the wiring to/from them
2) To signal to firefighters that the PV arrays and their wiring can be made safe enough to work in, on, or around, so that they won't just let the building burn down.

An argument for applying 690.12 to micro-inverter output circuits under the 2017 is that it costs less than $10 to slap labels on the service disconnect to encourage the firefighters not to let the building burn down for fear of being shocked by PV circuits. (Okay, if the service disconnect is not outside on a one or two family dwelling it costs a little more.) The firefighter was going to kill the service disco anyway, but they see RS switch label (hopefully) and tell their comrades that the array is safe to operate around. But this is moot with your ground mount because if the ground mount is on fire already you are not still needing the firefighters to save a building. Bluntly, it's less of a problem if they let it burn.
 
You must log in or register to reply here.
Top