Solar install turning out really bad - looking for some help/advice please.

[foo1bar]

I don't know if the inspector would allow that since the splices aren't shown on the permit, and the permit shows the 2AWG conductor gong all the way to the combiner inputs. If I were to do that, I'd probably be better off replacing the terminal blocks with ones that accept 2AWG.

Ask.

My guess is that the inspector would allow it.

"The connector on the equipment isn't sized to for #2 wire. Any problem with a 5" pigtail of 6 gauge to connect to a reducer splice?"

I did guess wrong about the permit revision being a reasonable price, so it's quite possible I'm wrong on this too - but ask and see.
(although $150 for a major revision might be reasonable - after all, you would be changing the permit from having 4 to 3 strings.)

If it were me, I wouldn't replace a portion of the inverter - even something that logically shouldn't be a problem like a terminal block.

[sensij]

I do believe that with a resubmitted permit, we could still use the existing wiring.

The way I figure it...

15A x 3 strings = 45A x 1.25 = 56.25A
6AWG is rated at 75A x temp correction of .82 (since it's in the attic, not on the rooftop) = 61.5A

61.25A > 56.25A so it would be ok, no? Did I leave out any figure there in the calculation?

Calculation looks right to me, but maybe not the conclusion. You could even tolerate the next higher temp range (0.76 correction, for 51-55 deg C) and still be OK. Does the temp in your attic exceed 131 deg F? Honestly, I bet it does, but am not sure there are hard and fast rules on what temperature to use. Here are a couple threads on MikeHolt forums, but I don't think there is a definite answer. Whatever the city approves, must be right.

Edit: Setting code aside, I would be a little concerned that the time of day when the attic is hottest is also when the most current will be going through those wires.

[HX_Guy]

So I think for your current design you need
3 * 15A * 1.25 == 56.25A
which means #6 might work.
(I think in the attic you might still need a temp correction factor. I'm not sure.)

But at $150 for a permit amendment, it's probably cheaper to replace it with #2 and do a pair of butt splice reducers at each end.
(At least that's what I'd do if I were you at this point. Splices are ~$10 each, 50 feet of wire will be <$100.)

I'd probably go with http://www.amazon.com/Gardner-Bender.../dp/B000FPDIHM for the splice. (UL listed, rated for copper, looks easy enough to tape or put a heat-shrink tube over.) But I don't have the experience so if others have better suggestions based on their experience, listen to them.

I don't know if the inspector would allow that since the splices aren't shown on the permit, and the permit shows the 2AWG conductor gong all the way to the combiner inputs. If I were to do that, I'd probably be better off replacing the terminal blocks with ones that accept 2AWG.

[foo1bar]

Probably better, seems to be pretty tricky when going over 2 strings as I'm finding out.

I emailed SE back explaining the original permit with 4 strings, which would require a 75A conductor after NEC calc, and with a 65% derate here in Phoenix, it would need 115A or a 2AWG conductor to which he replied back and said yes, that is correct.

So I think for your current design you need
3 * 15A * 1.25 == 56.25A
which means #6 might work.
(I think in the attic you might still need a temp correction factor. I'm not sure.)

But at $150 for a permit amendment, it's probably cheaper to replace it with #2 and do a pair of butt splice reducers at each end.
(At least that's what I'd do if I were you at this point. Splices are ~$10 each, 50 feet of wire will be <$100.)

I'd probably go with http://www.amazon.com/Gardner-Bender.../dp/B000FPDIHM for the splice. (UL listed, rated for copper, looks easy enough to tape or put a heat-shrink tube over.) But I don't have the experience so if others have better suggestions based on their experience, listen to them.

[HX_Guy]

As predicted by this post, although the temp correction I used was not quite enough.

Sorry about the misinformation on connecting multiple strings.

You've been very spot on Sensij.

I do believe that with a resubmitted permit, we could still use the existing wiring.

The way I figure it...

15A x 3 strings = 45A x 1.25 = 56.25A
6AWG is rated at 75A x temp correction of .82 (since it's in the attic, not on the rooftop) = 61.5A

61.25A > 56.25A so it would be ok, no? Did I leave out any figure there in the calculation?

[sensij]

Probably better, seems to be pretty tricky when going over 2 strings as I'm finding out.

I emailed SE back explaining the original permit with 4 strings, which would require a 75A conductor after NEC calc, and with a 65% derate here in Phoenix, it would need 115A or a 2AWG conductor to which he replied back and said yes, that is correct.

As predicted by this post, although the temp correction I used was not quite enough.

Sorry about the misinformation on connecting multiple strings.

[HX_Guy]

Definitely spots for 2 strings on all Solaredge dc disconnects. I don't combine on the roof unless absolutely necessary. Not sure on the conduit fill and temp corrections off hand though.

Probably better, seems to be pretty tricky when going over 2 strings as I'm finding out.

I emailed SE back explaining the original permit with 4 strings, which would require a 75A conductor after NEC calc, and with a 65% derate here in Phoenix, it would need 115A or a 2AWG conductor to which he replied back and said yes, that is correct.

[Ben25]

Definitely spots for 2 strings on all Solaredge dc disconnects. I don't combine on the roof unless absolutely necessary. Not sure on the conduit fill and temp corrections off hand though.

[HX_Guy]

Whoa, here's one out of left field!

I've been emailing back and forth with SolarEdge about this issue and trying to get some sort of documentation to show the City that the inverter will never have more than 34.5A running through the conductors but that their paperwork is unclear on this. I get this email back just a bit ago...

"The City is correct in taking the number of strings x 15 amps x 1.25 for sizing the conductors between a 3 string combiner and the inverter since this would be the available current in a fault condition. The inverter maximum input current vs calculated current would be used for calculating nominal operating current for the system labels."

I did not see that coming! So it seems nothing is wrong with the permit after all? Based on 4 strings, thats 60A x 1.25 = 75A. With rooftop conduit (as originally shown) a derate of 65% would mean the wire needs to be rated at least 115A @ 90º C. The smallest wire that meets that is in fact 2AWG.

[HX_Guy]

The SolarEdge inverter has terminals for only one string. The inverter's built-in overcurrent protection is sized for two strings, which is why they say they can handle two. If you run two strings, you would still need to parallel them in a combiner ahead of the inverter. Also, by running two strings to the inverter, now you've got 4 current carrying conductors in that FMC, and would need to apply yet another correction factor that has so far been avoided.

The DC terminals inside the disconnect actually have two sets of inputs, I assume you could come down with two sets of conductors for 2 strings? Like you said though, the conduit fill and temp corrections would need to be changed though.

[sensij]

But if you combined 2 of the 3 strings on the roof in the combiner and then ran the last string down separately, that would do it.

The SolarEdge inverter has terminals for only one string. The inverter's built-in overcurrent protection is sized for two strings, which is why they say they can handle two. If you run two strings, you would still need to parallel them in a combiner ahead of the inverter. Also, by running two strings to the inverter, now you've got 4 current carrying conductors in that FMC, and would need to apply yet another correction factor that has so far been avoided.

[sensij]

Yeah, but my city is using 2011 NEC and apparently, at least according to the designer, is using the 1.25 twice for DC. I guess they could be educated on the matter but still the numbers are pretty high.

If you have 4 strings, that would need to be calculated as 60A x 1.25 = 75A. A #2 wire on the roof would be rated for up to 84.5A after temp correction (based on 130A @ 90º C). Makes sense why the permit says 2AWG then I guess.

If I were to redo my permits, a #6 would still work apparently though, but I would have to show the new string configurations and everything, and hope they understand to use 1.25 only once, not twice. I don't have a lot of faith in that happening. Though with the conduit now in the attic, the derate is much lower (is it the standard .82 or something else?) and it may still work even with doing 1.25 twice.

Listen, I understand your design company is being professional about this and really trying to help, but based on what they've put in the permit so far, I'm not sure you can rely on them to be an accurate source of information. If you re-do the permit, make sure that each circuit is called out by name: PV Source, DC Output, Inverter Input, Inverter Output. Show the calculation specific to each circuit, don't just lump DC and AC calculations and corrections together. When referring to equipment on the line diagrams and in the data sheet attachments, make sure the optimizers are called "DC Utilization Equipment". That should help make it clearer for the city, even in the 2011 version of code.

[Living Large]

If I were to redo my permits, a #6 would still work apparently though, but I would have to show the new string configurations and everything, and hope they understand to use 1.25 only once, not twice. I don't have a lot of faith in that happening. Though with the conduit now in the attic, the derate is much lower (is it the standard .82 or something else?) and it may still work even with doing 1.25 twice.

As a newbie here, I didn't understand the distinction and implications for wire ratings between the setup I am considering (panels directly to a charge controller) and yours (optimizers). sensij appears to have gone to the crux of the matter of the difference between PV Source and DC output. I was very annoyed at having to multiply by 1.25 * 1.25 in sizing my PV Source wires - but for my situation it is required. My gut reaction to your concern, assuming sensij is correct and it sounds like they are, would be there should be a way of educating the city on this point. I'll keep watching the conversation out of interest...

[HX_Guy]

But if you combined 2 of the 3 strings on the roof in the combiner and then ran the last string down separately, that would do it.

That's a good thought Ben, it would possibly satisfy all parties...though who knows how confused the city would be with that setup. That doesn't sound "typical" of an install.

[Ben25]

But if you combined 2 of the 3 strings on the roof in the combiner and then ran the last string down separately, that would do it.