various options to interconnect remote array structure

[electrofelon]

I was just going over in my head the various options and pros and cons for getting juice to/from a remote array structure and thought I would run it by the group for input and to catch anything I might be missing. Say we have a single family dwelling and a ground mount array. Here is the way I see the options:

1. Inverter at house, Run strings or a combined string back to the house.

2. Inverter at array, run a feeder to a panelboard at the array structure, connect the inverter output circuit to this panelboard.

3. Inverter at array, run inverter output circuit back to house.

4. Inverter at array, Use 230.40 Exception #3 to run a set of SEC to a service panelboard at the array structure.

My thoughts:

----->#1 gets you the most bang for the buck for voltage drop, although The PV wire will likely be copper where as the other options could likely use AL, not sure what wins out cost wise without running some numbers.

----> #1 doesnt allow use of 1000V strings.

----> Regarding #2 vs #3, I dont see an advantage of #2 over #3 unless you have multiple inverters you want to combine or want power at the array for other uses.

----> I like #4 a lot and have done that several times.

----> IIRC, prior to 2017 #1 and #3 would not require a GES at the array right? (still havent gotten my 2017 code book)

----> #1 would invoke rapid shutdown issues if you go more than 5 feet inside the structure with strings. I am not up on the 2017 changes to RS.

anything else?

 

[ggunn]

----> #1 would invoke rapid shutdown issues if you go more than 5 feet inside the structure with strings.

Rapid shutdown does not apply to ground mounted systems under NEC 2014. 690.12 Rapid Shutdown of PV Systems on Buildings.

 

[electrofelon]

Rapid shutdown does not apply to ground mounted systems under NEC 2014. 690.12 Rapid Shutdown of PV Systems on Buildings.

Thanks for the correction. I forgot the wording.

 

[crispysonofa]

My thoughts:

----->#1 gets you the most bang for the buck for voltage drop, although The PV wire will likely be copper where as the other options could likely use AL, not sure what wins out cost wise without running some numbers.

If the run was long enough and cost was a concern, Could you not make a transition from copper to aluminum with listed fittings to take care of this?

 

[pv_n00b]

Rapid shutdown does not apply to ground mounted systems under NEC 2014. 690.12 Rapid Shutdown of PV Systems on Buildings.

Not correct. Both 2014 and 2017 NEC have the following wording: 690.12 Rapid Shutdown of PV Systems on Buildings. PV system circuits installed on or in buildings shall include a rapid shutdown function...

PV circuits are not limited to the DC circuits in the array but also include the AC circuits. So if the remotely mounted inverter's AC circuits terminate at a panel mounted on a building then RSS applies to the AC circuit. Unless the only purpose of the building is to house the PV system equipment.

 

[electrofelon]

If the run was long enough and cost was a concern, Could you not make a transition from copper to aluminum with listed fittings to take care of this?

If using 600V strings then certainly. If using 1KV strings, it would have to be PV wire. Aluminum PV wire is available, but not very common and likely special order. I do mostly commercial/utility scale PV, and the last few resi systems I did were 1KV strings, so I generally think in terms of 1 KV, so thanks for pointing that out.

 

[electrofelon]

Not correct. Both 2014 and 2017 NEC have the following wording: 690.12 Rapid Shutdown of PV Systems on Buildings. PV system circuits installed on or in buildings shall include a rapid shutdown function...

I admit this could be read both ways. I have generally seen it applied as Mr Gunn stated.

PV circuits are not limited to the DC circuits in the array but also include the AC circuits. So if the remotely mounted inverter's AC circuits terminate at a panel mounted on a building then RSS applies to the AC circuit

The Ac output circuit automatically meets Rapid shutdown.

Unless the only purpose of the building is to house the PV system equipment.

Not sure I agree with that. Could you elaborate?

 

[ggunn]

Not correct. Both 2014 and 2017 NEC have the following wording: 690.12 Rapid Shutdown of PV Systems on Buildings. PV system circuits installed on or in buildings shall include a rapid shutdown function...

PV circuits are not limited to the DC circuits in the array but also include the AC circuits. So if the remotely mounted inverter's AC circuits terminate at a panel mounted on a building then RSS applies to the AC circuit. Unless the only purpose of the building is to house the PV system equipment.

Well, OK, but in a grid tied PV system the AC circuits already conform to RSS requirements via UL1741. RSS requirements on the DC side do not apply to a ground mounted system except in the unlikely event that its DC circuits are routed in or on a building.

 

[electrofelon]

Ok what do you think of this setup. I am currently building this and hoping I wont get any flack for it:

Single family dwelling. Utility does not require a "utility disconnect". I use 230.40 exception #3 to tap the SEC after the meter (with no disconnect) and run another set of SEC (2/0 2/0 1) 100 feet to a ground mount solar array. There I hit a 100 amp Service rated MLO panelboard. The only thing in the panelboard is a 40 amp breaker feeding the inverter output circuit. Neutral bonded of course and 2 ground rods. What I like about it is I really trimmed costs doing it this way: Of course not having a utility disconnect helps, but I also get to use only three wires to the array. I used one of those homeline 6/12 3R little things for the panelboard at the array. IT took the 2/0, which was oversized for VD. Note this is a supply side connection (or is it if there are no other breakers?).

My only hesitation is if this array and structure meets the spirit of an "accessory structure" in 230.40 exception 3. I find myself wanting to add a convenience outlet and a light or something out there to make it "not just doing the PV", although there is no code justification for this.

 

[jaggedben]

Ok what do you think of this setup. I am currently building this and hoping I wont get any flack for it:

Single family dwelling. Utility does not require a "utility disconnect". I use 230.40 exception #3 to tap the SEC after the meter (with no disconnect) and run another set of SEC (2/0 2/0 1) 100 feet to a ground mount solar array. There I hit a 100 amp Service rated MLO panelboard. The only thing in the panelboard is a 40 amp breaker feeding the inverter output circuit. Neutral bonded of course and 2 ground rods. What I like about it is I really trimmed costs doing it this way: Of course not having a utility disconnect helps, but I also get to use only three wires to the array. I used one of those homeline 6/12 3R little things for the panelboard at the array. IT took the 2/0, which was oversized for VD. Note this is a supply side connection (or is it if there are no other breakers?).

My only hesitation is if this array and structure meets the spirit of an "accessory structure" in 230.40 exception 3. I find myself wanting to add a convenience outlet and a light or something out there to make it "not just doing the PV", although there is no code justification for this.

mmm, 705.31? Seems to me you need a fused disconnect, or cable limiters, within 10ft of your tap.

 

[electrofelon]

mmm, 705.31? Seems to me you need a fused disconnect, or cable limiters, within 10ft of your tap.

I do. Note the interconnection point is at the remote structure not at the house. The conductors between the house and the remote structure are Service entrance conductors permitted by 230.40 exception 3.

 

[shortcircuit2]

Ok what do you think of this setup. I am currently building this and hoping I wont get any flack for it:

Single family dwelling. Utility does not require a "utility disconnect". I use 230.40 exception #3 to tap the SEC after the meter (with no disconnect) and run another set of SEC (2/0 2/0 1) 100 feet to a ground mount solar array. There I hit a 100 amp Service rated MLO panelboard. The only thing in the panelboard is a 40 amp breaker feeding the inverter output circuit. Neutral bonded of course and 2 ground rods. What I like about it is I really trimmed costs doing it this way: Of course not having a utility disconnect helps, but I also get to use only three wires to the array. I used one of those homeline 6/12 3R little things for the panelboard at the array. IT took the 2/0, which was oversized for VD. Note this is a supply side connection (or is it if there are no other breakers?).

My only hesitation is if this array and structure meets the spirit of an "accessory structure" in 230.40 exception 3. I find myself wanting to add a convenience outlet and a light or something out there to make it "not just doing the PV", although there is no code justification for this.

Interesting and clever approach.

Are the conductors to the GM Array really "service conductors" if they don't "deliver electric energy" to the premises served??? CMP-4 members insist that they are NOT service conductors...because with PV, the intention is to deliver energy to the utility.

I disagree with this CMP-4 interpretation myself because inverter's do use electric energy, although very little.

Anyway, install a receptacle outlet as required by 210.52(G)(2) and a lighting outlet thats almost required by 110.26(D) and there will be no argument that the 3-wire lateral conductors are service entrance conductors.

I don't see anything that is unsafe with your design.

 

[electrofelon]

CMP-4 members insist that they are NOT service conductors..

Shortcircuit2,

What was the context and specifics of that statement by the CMP? I have always been curious about when a supply side connection is a supply side connection or just another set of service conductors under one of the 230.40 exceptions. Many times it seems like it could be either. It is certainly tempting to make the decision based on whether they are "PV only" conductors or "regular" conductors and adding other loads does seem to logically differentiate the two situations, but there isnt explicit language on this.

 

[jaggedben]

Same old, same old debate.

 

[electrofelon]

Same old, same old debate.

We are going to settle this once and for all, right here right now!!

 

[shortcircuit2]

Same old, same old debate.

:roll:...yup...:rant:

 

[shortcircuit2]

Shortcircuit2,What was the context and specifics of that statement by the CMP?

Well, i don't have any specific statement...just everytime I try to discuss the issue with certain colleagues on the issue, one who is a CMP-4 member...the statement gets thrown back at me..."they are not service conductors". I have broached this issue at section meetings also and get the same push back.

So in the eyes of the believer that they are NOT service conductors...none of the long laundry list of service and grounding requirements apply...which IMO creates hazards.

Plenty of threads on this forum bout this subject.

 

[electrofelon]

Well, i don't have any specific statement...just everytime I try to discuss the issue with certain colleagues on the issue, one who is a CMP-4 member...the statement gets thrown back at me..."they are not service conductors". I have broached this issue at section meetings also and get the same push back.

So in the eyes of the believer that they are NOT service conductors...none of the long laundry list of service and grounding requirements apply...which IMO creates hazards.

Plenty of threads on this forum bout this subject.

Ok, so not an official interpretation or statement embedded in a change refusal, thought thats what you meant.

 

[shortcircuit2]

Ok, so not an official interpretation or statement embedded in a change refusal, thought thats what you meant.

The thread below had 2 formal interpretations from NFPA stating that the conductors from a supply side connection are feeders and 1 statement that PV was not a service...

http://forums.mikeholt.com/showthread.php?t=177368&highlight=nfpa+line+side+connection

I will check PI later for CMP statements...

 

[ggunn]

The thread below had 2 formal interpretations from NFPA stating that the conductors from a supply side connection are feeders and 1 statement that PV was not a service...

http://forums.mikeholt.com/showthread.php?t=177368&highlight=nfpa+line+side+connection

I will check PI later for CMP statements...

San Antonio says line side connected PV is a service. Austin says it is not. When in Rome...