Cable type and size - Page 3

Mikesolar · 01-16-14, 06:26 PM

Quote:

Originally Posted by Exeric

I've been thinking about this and I just don't think this is close to right. I think you are leaving out something big - voltage drop/foot per gauge of wire.
Also you have to multiply 60' x 2 = 120' for up and back total length of homerun wire times that resistance. Also, though I didn't say it, I intend to have no more than a 1% V drop from the 240volts = 2.4 V. Finally, my plan is to have 14 panels not 12, which I think I stated. The .9A x 1.25 is correct.

Standard design for voltage drop is 5% max, 3% in a branch circuit and 2% in a service. I have been doing solar systems for 25 years and never had a problem following these rules.

17 inverters on a 12g wire is fine, use a 20A breaker but as I had stated, IF you were to use 12 inverters, a 14/3 wire would be just fine over that length. 14 inverters is definitely a 12/3 wire. Of course if you want a 1% drop, it certainly doesn't hurt.

Exeric · 01-16-14, 09:20 PM

Well, I can see it from your perspective and if I was installing them for other people I might not bother with getting it down to 1% V drop. It would be difficult to educate a customer well enough so that he/she would even appreciate what a lower V drop would do for them, much less be willing to pay for the upgrade. On the other hand, even with 1 % I'm still losing 24 watts that are just heating up the wire at peak solar with 14 microinverters. Multiply that times a few hours a day for most of the year here in California and you are talking quite a bit of wasted electricity.

I'm just guestimating but you'd probably be wasting about 36 and 60 watts if using respectively 12 or 14 awg. There is a law of diminshing returns when buying ever thicker wire but I think 10 awg and 24 watts is still right in the sweet spot, especially if it is for your own house and not a customer's. Even Enphase recommends not more that 1% V drop in the wire.

Mikesolar · 01-17-14, 05:10 AM

Yes, you are right about that. It is traditional by most electricians that I know, to use the rules I noted above for AC wiring but I don't apply the same rules to DC, which for some reason is stricter. 1% is closer to the norm.

BTW, I have not heard of counting both legs of a 240v wire when calculating voltage drop because it is 120v on each leg so your example would be for 60ft. The result is the same anyway.

Exeric · 02-02-14, 06:13 PM

I finally got around to installing the flashed roof solar electrical junction. Here it is:

It was fairly straight forward and easy to install. It isn't completely sealed at the bottom so weather resistant cable can come out there if you go that route instead of conduit. That means it isn't suitable for a flat or low incline roof as any pooled water will then get in. My porch roof is a low incline so I made sure not to install it there. Here's the tentative plan for future solar panels:

There is also provision for using conduit, which I plan to use between the junction and the panels. As you can see the cable length is much less than I originally thought so I was able to thread 12 awg Romex up through the roof and still get a calculated 1% voltage drop in the home run cable. I decided to run 2 each 50' 3 wire 12 awg cable so I can expand to include all 23 panels at a later date if I want to. Right now I'm planning on just the 14 panels on located on the porch roof. It's much better to add the provision now when I have access than later when it would really be a head ache. That's even if I don't do it.

BTW Mikesolar, the reason I calculated 240 volts across BOTH hot wires was because you yourself said only 10% or so of power goes through the neutral wire when it supplies power to the grid. So I was just using that paradigm instead of the usual one of ONE hot wire at 120 volts. It works out the same but it makes more sense to think of the round trip providing both hots making up 240 volts.

where2 · 02-02-14, 08:00 PM

Glad the OP seems to have figured out a solution, sorry I didn't chime in sooner.

In my case, I used sunlight and oil resistant TCER (12/4, same specs as Enphase trunk cable) in my short exposed run from my junction box on the roof down the weather head. Just inside the attic, at the base of the weather head, I transitioned to 10/3+G (UF) for the run between my second floor attic and first floor solar combiner panel. The UF wire gave me a 90°C insulation rating and a shade lower voltage drop on the rather long run I was making. Using 10/3+G also allows me the flexibility to upgrade panels or micro-inverters later without pulling new wire through two stories of walls and attic.

Having a metal tile roof (Decra tile), the self-flashing junction box/pass through boxes designed for asphalt shingles weren't going to work for me. So I chose to use a Hubbell NEMA4X fiberglass panel enclosure on the roof to keep my connections water tight, then jumped down a weather head into the attic. The only part of my PV installation visible from the street is the weather head.

Exeric · 02-04-14, 12:31 PM

It sounds like the weatherhead route worked for you. I wasn't really that familiar or comfortable going that route for my situation. I looked up weatherhead on Wiki and found that they can't be used in higher than 600 volts operation, which luckily I don't have. Then you need a pothead... There's a lot of those in my neck of the woods. The only advantage of going with the flashed panel is that it makes it easy to move the individual 20 amp C.B. for each microinverter circuit to near the main junction box instead of on the roof. I think microinverters are safer than high voltage PV because when power is cut to them from the grid they are supposed to all shut down automatically.

stevehull · 02-04-14, 03:24 PM

Why not just run some gray 1/1/4 inch conduit up there and leave in a "pull" (not pill (!) string (through ceiling and out to attic space). That is room for LOTS of wires - and cheap as well.

Steve

Exeric · 02-04-14, 04:11 PM

Are you addressing both me and where2 or just me? In my own case my house is already pretty torn up with sheet rock removed so it is very easy access to put in Romex. Besides that, I've been running plenty of it already as part of my renovation - no reason for me to change up for the microinverters if I don't have to. Romex can be used on any 120V system inside walls or in attics (240V is just two phase 120 V). Besides, it's already installed. And if I disconnect the power at the main panel, or a subjunction box next to it, it should shut off the individual microinverters, so there is no need for the extra mechanical protection inside the house structure that conduit would provide. I think my system is as cheap overall as using conduit and individual wires. The two cables I installed are sufficient to handle all the panels that could ever be installed on my particular roof and its south orientation.

where2 · 02-06-14, 10:43 AM

I didn't use 1-1/4" conduit because my second floor walls are only constructed from 2x4's, and punching a 1-1/4" conduit through the top plate would take away too much structural wood in an already minimal design. On my 5/12 pitch, 2x4 truss roof, turning the corner after punching through the top plate on the wall doesn't fit using conduit and working bend radii. That's where the UF cable really became the best option.

01-17-14, 05:10 AM	#23
Mikesolar Master EcoRenovator Join Date: Aug 2012 Location: Toronto Posts: 958 Thanks: 40 Thanked 158 Times in 150 Posts	Yes, you are right about that. It is traditional by most electricians that I know, to use the rules I noted above for AC wiring but I don't apply the same rules to DC, which for some reason is stricter. 1% is closer to the norm. BTW, I have not heard of counting both legs of a 240v wire when calculating voltage drop because it is 120v on each leg so your example would be for 60ft. The result is the same anyway. Last edited by Mikesolar; 01-17-14 at 05:13 AM..

02-02-14, 06:13 PM	#24
Exeric Apprentice EcoRenovator Join Date: Aug 2012 Location: California Posts: 274 Thanks: 19 Thanked 37 Times in 28 Posts	I finally got around to installing the flashed roof solar electrical junction. Here it is: It was fairly straight forward and easy to install. It isn't completely sealed at the bottom so weather resistant cable can come out there if you go that route instead of conduit. That means it isn't suitable for a flat or low incline roof as any pooled water will then get in. My porch roof is a low incline so I made sure not to install it there. Here's the tentative plan for future solar panels: There is also provision for using conduit, which I plan to use between the junction and the panels. As you can see the cable length is much less than I originally thought so I was able to thread 12 awg Romex up through the roof and still get a calculated 1% voltage drop in the home run cable. I decided to run 2 each 50' 3 wire 12 awg cable so I can expand to include all 23 panels at a later date if I want to. Right now I'm planning on just the 14 panels on located on the porch roof. It's much better to add the provision now when I have access than later when it would really be a head ache. That's even if I don't do it. BTW Mikesolar, the reason I calculated 240 volts across BOTH hot wires was because you yourself said only 10% or so of power goes through the neutral wire when it supplies power to the grid. So I was just using that paradigm instead of the usual one of ONE hot wire at 120 volts. It works out the same but it makes more sense to think of the round trip providing both hots making up 240 volts. Attached Thumbnails

02-04-14, 03:24 PM	#27
stevehull Steve Hull Join Date: Dec 2012 Location: hilly, tree covered Arcadia, OK USA Posts: 826 Thanks: 241 Thanked 165 Times in 123 Posts	Why not just run some gray 1/1/4 inch conduit up there and leave in a "pull" (not pill (!) string (through ceiling and out to attic space). That is room for LOTS of wires - and cheap as well. Steve __________________ consulting on geothermal heating/cooling & rational energy use since 1990 Last edited by stevehull; 02-04-14 at 08:22 PM..

01-16-14, 09:20 PM	#22
Exeric Apprentice EcoRenovator Join Date: Aug 2012 Location: California Posts: 274 Thanks: 19 Thanked 37 Times in 28 Posts	Well, I can see it from your perspective and if I was installing them for other people I might not bother with getting it down to 1% V drop. It would be difficult to educate a customer well enough so that he/she would even appreciate what a lower V drop would do for them, much less be willing to pay for the upgrade. On the other hand, even with 1 % I'm still losing 24 watts that are just heating up the wire at peak solar with 14 microinverters. Multiply that times a few hours a day for most of the year here in California and you are talking quite a bit of wasted electricity. I'm just guestimating but you'd probably be wasting about 36 and 60 watts if using respectively 12 or 14 awg. There is a law of diminshing returns when buying ever thicker wire but I think 10 awg and 24 watts is still right in the sweet spot, especially if it is for your own house and not a customer's. Even Enphase recommends not more that 1% V drop in the wire.

02-02-14, 08:00 PM	#25
where2 DIY Geek Join Date: Apr 2013 Location: Sunny Florida Posts: 401 Thanks: 74 Thanked 83 Times in 73 Posts	Glad the OP seems to have figured out a solution, sorry I didn't chime in sooner. In my case, I used sunlight and oil resistant TCER (12/4, same specs as Enphase trunk cable) in my short exposed run from my junction box on the roof down the weather head. Just inside the attic, at the base of the weather head, I transitioned to 10/3+G (UF) for the run between my second floor attic and first floor solar combiner panel. The UF wire gave me a 90°C insulation rating and a shade lower voltage drop on the rather long run I was making. Using 10/3+G also allows me the flexibility to upgrade panels or micro-inverters later without pulling new wire through two stories of walls and attic. Having a metal tile roof (Decra tile), the self-flashing junction box/pass through boxes designed for asphalt shingles weren't going to work for me. So I chose to use a Hubbell NEMA4X fiberglass panel enclosure on the roof to keep my connections water tight, then jumped down a weather head into the attic. The only part of my PV installation visible from the street is the weather head.

02-04-14, 12:31 PM	#26
Exeric Apprentice EcoRenovator Join Date: Aug 2012 Location: California Posts: 274 Thanks: 19 Thanked 37 Times in 28 Posts	It sounds like the weatherhead route worked for you. I wasn't really that familiar or comfortable going that route for my situation. I looked up weatherhead on Wiki and found that they can't be used in higher than 600 volts operation, which luckily I don't have. Then you need a pothead... There's a lot of those in my neck of the woods. The only advantage of going with the flashed panel is that it makes it easy to move the individual 20 amp C.B. for each microinverter circuit to near the main junction box instead of on the roof. I think microinverters are safer than high voltage PV because when power is cut to them from the grid they are supposed to all shut down automatically.

02-04-14, 04:11 PM	#28
Exeric Apprentice EcoRenovator Join Date: Aug 2012 Location: California Posts: 274 Thanks: 19 Thanked 37 Times in 28 Posts	Are you addressing both me and where2 or just me? In my own case my house is already pretty torn up with sheet rock removed so it is very easy access to put in Romex. Besides that, I've been running plenty of it already as part of my renovation - no reason for me to change up for the microinverters if I don't have to. Romex can be used on any 120V system inside walls or in attics (240V is just two phase 120 V). Besides, it's already installed. And if I disconnect the power at the main panel, or a subjunction box next to it, it should shut off the individual microinverters, so there is no need for the extra mechanical protection inside the house structure that conduit would provide. I think my system is as cheap overall as using conduit and individual wires. The two cables I installed are sufficient to handle all the panels that could ever be installed on my particular roof and its south orientation.

02-06-14, 10:43 AM	#29
where2 DIY Geek Join Date: Apr 2013 Location: Sunny Florida Posts: 401 Thanks: 74 Thanked 83 Times in 73 Posts	I didn't use 1-1/4" conduit because my second floor walls are only constructed from 2x4's, and punching a 1-1/4" conduit through the top plate would take away too much structural wood in an already minimal design. On my 5/12 pitch, 2x4 truss roof, turning the corner after punching through the top plate on the wall doesn't fit using conduit and working bend radii. That's where the UF cable really became the best option.