SMA 3000 TL Ungrounded Inverter

[Neal30]

I have a question about appropriate conductors for an Ungrounded inverter. We are looking at transitioning from PV wire on the roof, to either MC cable or NM cable. I was reading through the code and I ran upon this. If anyone has there interpretation of this, and some advice it would be greatly appreciated.

690.35 Ungrounded Photovoltaic Array

System complies with 690.35 (a) through (g)
(A) Disconnects
(B) OCP
(C)Ground-Fault Protection*
(1) Detects Ground Faults
(2) Indicates that a ground fault has occurred
(3) Automatically disconnects all conductors or causes inverter or charge controller connected to the faulted circuit to automatically cease supplying power to output circuits.

(D) The photovoltaic and source conductors shall consist of the following:

(1) Non-Metallic Jacketed multi-conductor cables? ( Does this mean that I can run NM wire or Romex from the J-box in the attic down to my inverters?) I am thinking that since the inverter is listed with Ground Fault Protection*, this is what is allowing me the option of running conductors not in a raceway, or in a metal jacket.

(2) Conductors installed in raceways, or
(3) Conductors listed and identified as Photovoltaic (PV) Wire installed as exposed, single conductors.


So what are your thoughts, from a Safety and durability standpoint for this type of conductor?

Thanks

 

[GoldDigger]

I have a question about appropriate conductors for an Ungrounded inverter. We are looking at transitioning from PV wire on the roof, to either MC cable or NM cable. I was reading through the code and I ran upon this. If anyone has there interpretation of this, and some advice it would be greatly appreciated.

690.35 Ungrounded Photovoltaic Array

System complies with 690.35 (a) through (g)
(A) Disconnects
(B) OCP
(C)Ground-Fault Protection*
(1) Detects Ground Faults
(2) Indicates that a ground fault has occurred
(3) Automatically disconnects all conductors or causes inverter or charge controller connected to the faulted circuit to automatically cease supplying power to output circuits.

(D) The photovoltaic and source conductors shall consist of the following:

(1) Non-Metallic Jacketed multi-conductor cables? ( Does this mean that I can run NM wire or Romex from the J-box in the attic down to my inverters?) I am thinking that since the inverter is listed with Ground Fault Protection*, this is what is allowing me the option of running conductors not in a raceway, or in a metal jacket.

(2) Conductors installed in raceways, or
(3) Conductors listed and identified as Photovoltaic (PV) Wire installed as exposed, single conductors.


So what are your thoughts, from a Safety and durability standpoint for this type of conductor?

Thanks

I think that section only allows you to use NM for that purpose in an environment where NM would otherwise be allowed. That would include properly secured and supported in a non-habitable attic, IMHO, but not exposed to damage in other spaces.

I do not quite get the point of not allowing MC outside of a raceway though.

 

[jaggedben]

...
(D) The photovoltaic and source conductors shall consist of the following:

(1) Non-Metallic Jacketed multi-conductor cables? ( Does this mean that I can run NM wire or Romex from the J-box in the attic down to my inverters?)

I don't think anything in 690.35 invalidates anything in 690.31(E). So you cannot use NM, period, until you've reached the first readily accessible disconnect. If the attic is not readily accessible then you cannot use it from a j-box there. If the attic is readily accessible, you need a disconnect there.

I am thinking that since the inverter is listed with Ground Fault Protection*, this is what is allowing me the option of running conductors not in a raceway, or in a metal jacket.

I am thinking that Ground Fault Protection has nothing to do with the question.

So what are your thoughts, from a Safety and durability standpoint for this type of conductor?

Since you asked, my personal opinion is that running DC source circuits inside a building in NM is a horrible idea, and positively the least safe way to do it. One drywall screw in the wrong place and you have a fire that no fuse or circuit breaker will stop. If you used NM on the PV side of a readily accessible disconnect then really nothing would stop the fire at all, which is why that is positively not allowed by the code. I think that's among the stupidest things one could ever do on a PV installation. If you use it on the inverter side of a readily accessible disconnect, then at least a disconnect could stop the fire source. But who would trust that there's going to be a person around who understands what is causing the fire and goes to shut off the disconnect?

Don't do it.

 

[Zee]

Valid Question, Neal30,
I know it is tempting.
I absolutely concur with previous opinion: don't do it. In over 12 years I have never seen DC conductors in Romex, and i have seen some stuff......
Is the Romex rated for 600 VDC?
Regardless, PV DC has to be in metal conduit. EMT (or flex conduit indoors) is fine and what we always use.

You also wrote:

"(2) Conductors installed in raceways, or
(3) Conductors listed and identified as Photovoltaic (PV) Wire installed as exposed, single conductors."

(3) above is not true as written! PV wire cannot be exposed, it must be covered. Basically pv wire under panels, or in metal conduit anywhere between arrays or leaving array.

 

[Zee]

I digress:
IF SMA SB3000Tl is outdoors.......make sure to use TR AND WR GFCI outlet for the Secure Power Supply outlet..... and bubble cover (aka in-use cover) per latest NEC requirements. Use a two gang bell box. :happyyes:

Inspector may fail you otherwise.

 

[GoldDigger]

[/B](3) above is not true as written! PV wire cannot be exposed, it must be covered. Basically pv wire under panels, or in metal conduit anywhere between arrays or leaving array.

There is a difference, which you point out, between being approved for exposure to sunlight and rain where neither accessible to casual contact nor subject to damage, and being approved for exposed routing in general.

 

[Neal30]

Total Agreement

Total Agreement

I don't think anything in 690.35 invalidates anything in 690.31(E). So you cannot use NM, period, until you've reached the first readily accessible disconnect. If the attic is not readily accessible then you cannot use it from a j-box there. If the attic is readily accessible, you need a disconnect there.



I am thinking that Ground Fault Protection has nothing to do with the question.



Since you asked, my personal opinion is that running DC source circuits inside a building in NM is a horrible idea, and positively the least safe way to do it. One drywall screw in the wrong place and you have a fire that no fuse or circuit breaker will stop. If you used NM on the PV side of a readily accessible disconnect then really nothing would stop the fire at all, which is why that is positively not allowed by the code. I think that's among the stupidest things one could ever do on a PV installation. If you use it on the inverter side of a readily accessible disconnect, then at least a disconnect could stop the fire source. But who would trust that there's going to be a person around who understands what is causing the fire and goes to shut off the disconnect?

Don't do it.

I am full agreement. It did seem a little sketchy to have the PV panels able to burn indefinitely if there was any damage to the wires. The subject of using MC/metal clad flex indoors was brought up. What is your opinion on using MC cable from the attic to the inverter?

Thanks

 

[acrwc10]

The phrase,"burn clear or burn down" comes to mind.

 

[GoldDigger]

...It did seem a little sketchy to have the PV panels able to burn indefinitely if there was any damage to the wires.

Fundamental misconception in play here:
A PV panel is a current-limited source. The current into a dead short will not be more than the normal operating current by more than about 10% and will not harm the panels at all.
The concern about panels on the building roof is not that they will burn, it is that the wiring will start a fire or the voltages present will injure a fire fighter. And wiring ignition, in turn, will only happen if there is an arc, since the wiring is rated for the full possible current that the panels can produce.

Under some circumstances (lack of (now mandatory) fusing allowing unlimited reverse current from the rest of the array into a damaged panel for example) the panels can "burn" or the wiring overheat, but those are not the primary hazards.

 

[acrwc10]

Fundamental misconception in play here:
A PV panel is a current-limited source. The current into a dead short will not be more than the normal operating current by more than about 10% and will not harm the panels at all.
The concern about panels on the building roof is not that they will burn, it is that the wiring will start a fire or the voltages present will injure a fire fighter. And wiring ignition, in turn, will only happen if there is an arc, since the wiring is rated for the full possible current that the panels can produce.

Under some circumstances (lack of (now mandatory) fusing allowing unlimited reverse current from the rest of the array into a damaged panel for example) the panels can "burn" or the wiring overheat, but those are not the primary hazards.

While the string circuit is "current limitting", it is common for multiple strings to have the output conductors grouped together before landing on a fuse. When several strings can exceed the max current of the module you have the potential for a fire.

 

[GoldDigger]

It may be common, but it is not legal. The maximum series fuse specification on the panels should allow at most two panels to be paralleled without separate fusing. And that current limit would still be violated if you then parallel multiple fused pairs after the fuses.

Tapatalk...

 

[acrwc10]

What I am saying is not that they are connected, just that the grouping of the wires from multiple strings, before the fuses, can be hazardous outside of conduit.

 

[ggunn]

While the string circuit is "current limitting", it is common for multiple strings to have the output conductors grouped together before landing on a fuse. When several strings can exceed the max current of the module you have the potential for a fire.

I misread this as well, possibly because you said "...grouped together before landing on a fuse." Each string (if there are more than 2) must have its own fuse.