EMT for DC PV inside the house. Why?

[John Schmidt]

Though code requires the DC PV wires to be in EMT (or some similar metal enclosure, I think), does anyone know the "why"? Other than, "because NEC code requires it", is there any logical reason why? How is EMT safer than PVC? If anything, it seems like PVC would reduce any shock hazard.

 

[Bluedog225]

Emt is harder to catch fire?

Edit to add. DC arcs in ways that AC doesn’t.

 

[John Schmidt]

Emt is harder to catch fire?

If that was the case, all house wiring would need to be in EMT. Granted, the PV wiring can hit 500 vDC. Still, PVC would seem just as good.

 

[hwy17]

Other circuit types have unlimited current supply so a short circuit can trip a breaker.

If PV wires short, nothing trips, the short just keeps sparking inside the conduit until the wires melt away from each other.

 

[Bluedog225]

No. DC arc is qualitatively different.

 

[hwy17]

I just did this part of my wiring and I used FMC. EMT is the cooler way to do it, but if you're going through an attic or crawlspace that makes it a pain, FMC is also legit. Both need weatherproof considerations if you're installing in a wet location. But up on the underside of the floor joists is not a wet location if you ask my inspector (that's me).

 

[Skypower]

AC has much better arc extinguishing since it crosses through zero volts each cycle. DC voltage can maintain a longer more damaging arc. Think more heat and melting

 

[hwy17]

As mentioned, DC does have more arc sustaining characteristics, but I don't think these are the operative concern here. It's the lack of feasible overcurrent protection. A constant current AC power source would have the same problematic risk of continuous shorting.

 

[Ampster]

My guess is because of the potential high voltage of DC PV that EMT provides an additional ground path to direct that potential to ground. EMT is also harder to cut through or pierce than PVC.

 

[Ampster]

I just did this part of my wiring and I used FMC.

FMC was not allowed in a former California jurisdiction I lived in, but I never asked the reason. I agree FMC is much easier to work with.

 

[hwy17]

FMC was not allowed in a former California jurisdiction I lived in, but I never asked the reason. I agree FMC is much easier to work with.

Good point. My building department would probably agree then, they like to read between the lines in the "spirit of the code". But that's why I do my own inspections

 

[wheisenburg]

I think there are at least three reasons.

1. It is higher voltage (up to 600 volts).
2. DC tends to sustain arcs better than AC.
3. A short in a PV circuit is less likely to trip a circuit breaker or blow a fuse.

So if these wires arc and burn up, the metal conduit is more likely to contain the fire within the conduit than say PVC which would melt and burn at a much lower temperature. I would note that 3 phase 480 volt AC power found in commercial building is also going to be run in metal conduit.

 

[robbob2112]

It is long so only read it if you want my real reasons.


I recently did a calculation in another thread. 4 x 200ah batteries hooked to a bus bar. Now assume there is a wrench that falls across the middle of one of the positive leads using 4/0 wire to connect them. My contention was the fuses should be at the battery posts. Others were irritating me by insisting the fuses should be at the bus bar end of things.

But, here is the deal - in that setup or any with more than 2 parallel connected batteries - the wire portion between the wrench/shelf/sharp object would take the full amount of current from the other 3 batteries from one end and the full amount of current from the battery being shorted at the other. Assuming short wasn't a dead short and you only drew 200amps from each of the other battery that means the wrench would take 800amps total. With fuses of 225amps none would blow and the BMS wouldn't shut down current at least not at first.

The reaction time on the fuses used (MRBF 225 or 250 amp or Class T 250amp) is between 0.01 and 0.1 seconds. In that 0.01 seconds the wire wouldn't even get warm to touch. In 0.1 seconds the wire would be hot enough to melt the insulation off it and the wrench would be melted as well.

Lucky us, when you have a short on that level you tend to have a plasma arc, gasses released and things generally are thrown apart. But the 1 in a million chance is that the wrench welds in place and keeps the short up for just a fraction of a second. Also lucky us a short anywhere near that will trip a good BMS into overcurrent in 0.000001 seconds and nothing melts or blows except the BMS has to be reset. But, what about a bad BMS that has a history of failures? A BMS is a microprocessor and a series of high current MOSFET devices - generally 50amps each, stack as many as you need to hit the current you want. One problem with MOSFET devices is that they tend to fail shorted as often as they fail open. So if your BMS is of marginal quality (like in a lot of no-name batteries) you can hit that 0.1 seconds pretty easy. And you have melted wires inside metal and the current will start a fire and it should be contained long enough for the BMS to blow open.

What does this have to do with EMT? Well, I rather have melted wires inside metal that plastic, wouldn't you?

Now your DC pv wires aren't pushing anywhere near the current, but they are much smaller wires.

 

[John Schmidt]

My guess is because of the potential high voltage of DC PV that EMT provides an additional ground path to direct that potential to ground. EMT is also harder to cut through or pierce than PVC.

If one strand of the DC PV shorted to the EMT, there would be not any current to direct to ground, because the panel voltage is not referenced to ground. The DC voltage of a panel is referenced to the panel + and - only. Even if both shorted to the EMT, the panels would continue to provide power while the insulation slowly melts (and perhaps starts burning). I think that hwy17 (above) nailed it!

 

[John Schmidt]

Other circuit types have unlimited current supply so a short circuit can trip a breaker.

If PV wires short, nothing trips, the short just keeps sparking inside the conduit until the wires melt away from each other.

I think that you nailed it. If the PV wires short together for some reason, the solar panel will continue to provide power and potentially start a fire inside the EMT. To a point, the fire would be contained. Even a fuse at the start of the string (out on the ground mount, etc) would have limited value, since the fuse would be unlikely to blow. For example, if the series solar panel string can produce 10A, a 15A or 20A fuse would do nothing if there is a short inside the EMT.

 

[John Schmidt]

Nice discussion. Thank-you everyone. Now, my dilemma, since I now see the importance of EMT, is to decide if I want to fix the errors that I made when I brought the PV wires into the house. The external PVC looks beautiful, to include slip couplings. From the inside, it also looks nice. The EMT came out great. However, the transition from PVC to EMT is in a nearly unreachable location, so I let the PV10 stay without any PVC-to-EMT transition box or adapter. The PV10 goes about 6 inches through the wall without any protection. Stupid, I know, now. As crazy as it sounds, I might be able reach in that small space and wrap the wires together with foil. Kludgy and shameful, and no inspector would ever pass it, but we don't do much for inspection around these parts. Thoughts, including shaming, is appreciated!

 

[robbob2112]

Make a decision and do it, then if you find out it was bad change it. Simple words to live by from my grandfather many years ago.

So, fix it. If you can't figure out how I bet there are folks on here that have faced the same issue. Wires you can't reach? How about a fish tape to get you past what you can't reach? Either push the fish tape to where you can reach and tape wires to it and pull back or do it in reverse. That won't do everthing but it will do a lot. Or describe with pictures and ask and someone will suggest an answer.

 

[John Schmidt]

Make a decision and do it, then if you find out it was bad change it. Simple words to live by from my grandfather many years ago.

So, fix it. If you can't figure out how I bet there are folks on here that have faced the same issue. Wires you can't reach? How about a fish tape to get you past what you can't reach? Either push the fish tape to where you can reach and tape wires to it and pull back or do it in reverse. That won't do everthing but it will do a lot. Or describe with pictures and ask and someone will suggest an answer.

Logical.

It's an absolutely terrible location. It's an intersection of where the main house power, propane and water all come within a 3/4 cubic foot space within the floor joists in a corner of the basement. Of course, I had to go and add another pipe with 6 PV10 for grid tie and then another pipe with 4 PV10 for off-grid. It's a crisscross mess of pipes and wires. No excuses, I know. What would Spock do? Or Kirk? Or McCoy? Scotty would know!

 

[robbob2112]

Pull the pipe with the pv10 wires and put it through someplace else. Or pull the pipe and then feed it through from the other end.

And, Scotty wouldn't have done it that way in the first place. He would have beamed the pipe into place.

 

[hwy17]

To a point, the fire would be contained. Even a fuse at the start of the string (out on the ground mount, etc) would have limited value, since the fuse would be unlikely to blow. For example, if the series solar panel string can produce 10A, a 15A or 20A fuse would do nothing if there is a short inside the EMT.

Exactly. It's a logical conundrum. The PV has to produce its max output so therefore it cannot be fused. Fuses can only protect 2 parallel strings if they backfeed into one, which is why we only fuse 3 strings or more.