diy solar

diy solar

DIY 11.7kw ground based grid tied system questions?

All electrical components have to be UL listed, so as long as the wire is then it is fine.

Ground mounts wiring is usually in the ground so the wire is usually kept cool (heat is what causes resistance in wire). I have 10 awg on my 7kwh ground mount and it produces full power (I had a it just laying around and it barely was big enough for my almost 200 foot run). I really think some are missing the fact that the inverters are going to up the voltage more and more to push the power through the wire so do we really have to over size that much? Just an opinion....

I used one connection per panel.
Used the ones below with a self tapping ground type screw into the aluminum panel and just laid in like a 14 awg bare copper wire across them all. I then wire nutted them together in a small box and ran a regular multi strand ground wire.

 
For the ground wire to your ground mount located ground rod, how many attachment points did you use on the array, one, or many?
It depends on the mounting system you buy. Many systems now have ground continuity, so you only need one attachment point to the whole array assembly. They do that by incorporating WEEBs into the panel attachments and other fittings. That will be clearly called out on the array mounting system by the vendor because it's a selling feature. The alternative is running an individual ground wire to each panel and the mounting system itself, kind of a lot of wire and a pain.
 
It depends on the mounting system you buy. Many systems now have ground continuity, so you only need one attachment point to the whole array assembly. They do that by incorporating WEEBs into the panel attachments and other fittings. That will be clearly called out on the array mounting system by the vendor because it's a selling feature. The alternative is running an individual ground wire to each panel and the mounting system itself, kind of a lot of wire and a pain.
Good point... Mine was a wood pergola so there wasn't conductivity between panels...
 
It depends on the mounting system you buy. Many systems now have ground continuity, so you only need one attachment point to the whole array assembly. They do that by incorporating WEEBs into the panel attachments and other fittings. That will be clearly called out on the array mounting system by the vendor because it's a selling feature. The alternative is running an individual ground wire to each panel and the mounting system itself, kind of a lot of wire and a pain.
Hello
I am using Unirac for mounting the panels on a 2 inch galvanized well pipe support structure that is inserted into 14 ground screws that are in the ground 5.5 feet each. The grounding layout/needs for the newest Enphase inverters is a source of confusion to me. They have 2 wires and are bolted to the Unirac frame, but they are described as "ungrounded".

I fully understand the need for a ground wire in the 200 foot home run, my question is how much grounding is needed on the panels to the combiner box at the array. I don't understand why any added ground rod is needed if this giant metal structure is sitting on 12 ground screws in earth.

I have also been told that the Nema code has been updated to reduce the need for a long wire that winds around the bottom of the panels and is attached/grounded on each and every panel. I can do that but wow that is a lot of extra work .

Sounds like some here are doing that but I am going to investigate further and see what Unirac, and Enphase have to say...

Thanks for the education here.. I now understand that as long as my home run wires are THHN and meet code the gauge of the wire won't be a big sticking point with the inspector... as long as peak current voltage is supported .
 
Hello

At this link I see the following document.

Link to Enphase document

Q-cable no grounding needed.PNG

There are also install videos of installers who indicate that when you mount the enphase micro-inverters to the rail they are "bonded" and no ground is needed.

So.. if I can believe this information, I will run a ground lead from the array in the home run but don't need to add any additional copper wire to from the combiner box to the array with lugs as additional grounds.

Please feel free to correct me if I am interpreting this incorrectly..

Thanks in advance
 
Code will tell you what size ground wires need to be, the logic is the ground needs to be big enough to carry enough current for a short time to trip a breaker so the fault is isolated. You're carrying 48A at 240v, normally 10ga ground is adequate for 50A and you'd use 8ga wire. But since you have to upsize the load carriers for voltage drop,code says you have to upgrade ground proportionately. Not sure if that means by cross sectional area or what, but I think 6ga is more than adequate. If you want a definitive answer, I'd go on an electricians forum and ask the question.

For a battery or AC circuit, ground wire is allowed to be undersized and a short will result in very high current. At about 5x rating of OCP device (breaker or fuse) it will trip instantly.

For a PV array, current produced will be about Isc, which is maybe 10% or 20% above Imp. That isn't going to trip OCP any time soon, maybe never.
Exception is fusing of multiple parallel strings - if one string shorts and several others dump current into it, that current can trip OCP maybe after a few minutes.

If your PV system has GFCI, about 1A of leakage will shut it off so minimal ground wire would work. Whatever NEC requires for other circuits should be fine. If it does not have GFCI, ground wire could carry current continuously so I think it should be sized with ampacity sufficient for PV, which is 1.56x Isc of array.
 
SteinVT

I am doing a very similar ground mount install of 30 405W Trina panels with Enphase micro inverters. I plan on using a outdoor rated fuse box to combine the 310 panel strings, and the calculations say I need 3 20amp breakers one for each string. The homerun will be around 200 feet and I will be using 2 AWG copper and a 4 AWG copper ground that meet spec.

My question is have you confirmed that a single ground point on the array, with a ground wire into the combiner box meets your local code? I understand that this ground will then be continued into the home run 200 foot feed, and ultimately grounded in the home feed panel.

My local inspector is a PIA, and I want to avoid going round in circles with the guy over ground mount grounding attachments.

Thanks in advance.
Just an update, the wire sizing of 2AWG for the current, and 4AWG for the ground on the long run came from on online solar design tool that I used to layout my array. It takes some getting used to, and they have a free trial, and they even set up an appointment to review operation of the software online for free. I ended up paying them about $50 total for a very nice design document.

The conductor sizing option appears to tie the size of the current carrying wire to the minimum size ground wire allowed, as I increased the gauge on the current wires from 2 to 3 to 4 the maximum size ground wire also increased. So.. I am pretty confident that my 2AWG and 4AWG ground meet code.

No... I don't work or have any relationship with this company, just nice software for generating permit files needed for approvals.

Online software for Solar design
 
Hello
I am using Unirac for mounting the panels on a 2 inch galvanized well pipe support structure that is inserted into 14 ground screws that are in the ground 5.5 feet each. The grounding layout/needs for the newest Enphase inverters is a source of confusion to me. They have 2 wires and are bolted to the Unirac frame, but they are described as "ungrounded".

I fully understand the need for a ground wire in the 200 foot home run, my question is how much grounding is needed on the panels to the combiner box at the array. I don't understand why any added ground rod is needed if this giant metal structure is sitting on 12 ground screws in earth.

I have also been told that the Nema code has been updated to reduce the need for a long wire that winds around the bottom of the panels and is attached/grounded on each and every panel. I can do that but wow that is a lot of extra work .

Sounds like some here are doing that but I am going to investigate further and see what Unirac, and Enphase have to say...

Thanks for the education here.. I now understand that as long as my home run wires are THHN and meet code the gauge of the wire won't be a big sticking point with the inspector... as long as peak current voltage is supported .

Modern inverters and panels operate electrically "ungrounded", which is different than in the old days when the negative lead from the panels was also tied to ground. Don't confuse that operational requirement with the requirement for equipment grounding. Some mounting systems have electrical continuity, so you only need to connect them to the grounding conductor at one point. You need to ask Unirac if that's the case with their system. If not, you need separate wires to each panel and each electrically insulated part of the mounting system, then tie all those grounds together and run the ground wire back to the service panel. You are correct about the ground screws being a giant ground rod. However there is no NEC requirement for an auxiliary ground rod on an array, it's up to the whims of the local jurisdiction whether they require it. The best way to ground it is to carry the ground wire back to the main service panel and connect it to the neutral bus in the panel. I am with the school of thought that says you are better off without an auxiliary ground rod because they will conduct current due to the voltage potential built up along the surface of the earth during storms. If you have two ground rods separated by 200', they will likely be at different earth potentials of a few thousand volts, so they will conduct current through the ground wire from one ground rod to the other, and the wire may not survive. Ground rods are not intended to conduct the current from a lightning strike, and it won't matter how many rods you have or how big the wire is, they will be vaporized.
 
Thank you for the feedback.. I will check with Unirac and see what they say about the racking and grounding requirements.. My zoning permit said I needed 2 ground rods at the house 6 feet apart, and my plan is to meet that, as well as run a ground from the combiner box at the panels in the 200 foot long home run. My questions are what needs to be done between the combiner box at the array, and the array for grounding... no requirement for a ground rod, but I may need a copper wire to every single panel which is a PIA I am hoping to avoid.

Will let folks know what unirac says, and agree.. the whims of the local inspectors always rule.
:)
 
Traditional way to ground panel frames is a clip per panel that grabs a ground wire. This allows any panel to be removed without breaking ground to others.


"WEEB" is a system to ground rails together and frames to rails, rather than running a wire past all panels.

 
So progress continues. All the panels are up. The Enphase micro inverters are mounted and the outputs wired to an outdoor sub panel that I am using as a combiner. We buried 200' of 1 1/2" conduit to my garage and more conduit getting to the production meter and shut off switch.

Yet to do; pull the wires, run conduit to my main panel and install the Enphase IQ Enovy monitor. Last is to call the power company to take a look and put in a meter. As far as I can tell, that is the closest thing to an inspection that I will have to pass.

To you guys installing the Enphase micro inverters, a tip. When ordering the Q cable with drops, order a drop for each inverter and one for each sub-array. Then you can easily wire the arrays like the diagram in my first post by just building the trunk cables using the female Q connectors and Q trunk cable. I didn't so I had to splice in three more drops to my last array.

A couple of questions:
From the combiner, I will have two 2-gauge wires and a ground. These will connect to the production meter.

1. I should wire the incoming wires from the array to the contacts on the top of the production meter and the bottom contacts to the switch, right? (See picture)
production meter.jpg
2. Should I remove the screw in the production meter labeled "For Isolated Neutral Remove Screw" ?

3. In the switch, should the side with the moving arms be connected to meter, or to the main panel? I recall reading about someone failing an inspection because he had it wrong. I believe the micro-inverters shut down immediately with loss of 240v so I would guess the meter.

4. When wiring from the main panel, should I bring along a neutral? Even though it won't continue beyond the meter.

5. An last, does it mater if I connect into the main panel or a sub panel that is connected to the main?

You guys and this forum have been such a wealth of information. I sincerely thank you. - Mark
 
You mean the county/city doesn't require a permit and inspect these things?

1) Would think so.
2) Yes, one location bond at main meter or panel
3) I would say moving arms go to inverters, less prominent contacts to grid. So is switch between meter and main panel?
Here I thought separately metered went to utility grid. Then it would require mains rated breaker or fuse.
4) I think so, might be useful in the future? Considering where panels and meter are in the circuit, you might want to tap off this branch for a battery inverter and protected loads panel (or that could be done back at main or sub panel.)
5) Either could do, but every backfed breaker (going to PV) should be in last slot of a panel, far end away from main breaker (or lugs).
Sum of backfed breaker and main breaker (or branch breaker in main panel feeding sub panel) should not exceed 120% of busbar rating.
If sub panel, you'll have the option of making that a protected loads panel, with battery inverter between main and sub.
 
I question #1....The consumption and production meters you are wiring, usually from what I have seen, both get connected to feed from the road. Not that it can't be wired into your main breaker panel but most I have seen are wired into the main feed from the road.

Being that you aren't supposed to touch the feed side of the meter, your power company may place a jumper from the top of your existing meter to the new meter. OR they may force you to get a two meter box with a bus bar to feed both meter from the same feed wires. (I got caught in this scenario with a Geo meter and a main meter to our house). I would ask you power company where the feed for the new meter will be coming/supplied from. The standard for feeds is they come in the top of the meter.

The power company wants to keep the feed wires to the meters under lock so people can't move wires around and fake out the meters. In your proposed way of feeding your second meter from your main will give you access to both sides of the meter. So you could rewrought power flow through the meter and make it think you are producing more power then you are. People get creative at stealing power.
 
Hedges, thanks for your answers. The state requires a permit, I got that. Called the town and they require nothing. The power company is key.

Hedges and newbostonconst, here is what was provided from the power company.
PanelConnections.jpg

Could you explain a bit more about your answer to 5. Trying to understand why placement within the panel would matter. Is it because the grid power could be added to the PV power to supply a load below the PV input and overload the busbars? My sub-panel feed is not in that position.

When summing the ratings of the breakers, shouldn't I use the breaker where the PV energy enters the system? In my case that is 60a. The incoming breaker to the sub panel is 100a. So the busbar rating of the sub panel should be at least (100 + 60) * 1.2 = 192 ? And then the main panel busbar rating should be (150 + 60) * 1.2 = 252 ? Am I understanding this correctly? Are the busbar ratings listed on the panel somewhere? Thanks-Mark
 
This makes me question the info you got....I would ask for an wiring diagram and not a cartoon they have given you.
1633001183554.png
 
Newbostonconst, thanks for the link, it is very good. I can see now that I had the 120% rule all wrong. I have a 150a main panel, meaning my overhead is 180a. Subtract 60 amps for solar, I would have to downgrade my main breaker to 120a, probably not going to work.

I would love a new bigger main panel as we have outgrown the old one adding mini-splits and the like. I will check this out but it sounds pricey.

I also will investigate the Line Side Tap option with the my power company. That would require me to replace my disconnect with a fused version. Not really sure what one would use for a junction box with that many amps flowing through it. - Thanks, Mark
 
Some utilities have a newer style meter they plug in that has a line tap built in for solar, probably worth asking if they know about it. Based on the drawing they gave you I doubt if they're on the cutting edge of technology, but who knows.
 
Answer 5:
Hello

I just did some investigating of this as I have a 200A Cutler Hammer main panel in my home and a permit package for my install that says install a 60A feed breaker on the opposite end of the panel. On first look.. 200*1.2=240A max input so in my case I thought I was limited to a maximum 40A solar breaker.

Feedback from experts said look carefully at the panel specs as it is the BUSS BAR, capacity that is used in the calculations, NOT the main circuit breaker.
My Cutler-Hammer clearly says on the label 225A maximum ... so 1.2*225 = 270A max... so my 200A main breaker, AND a 60A solar feed is below that 1.2 Max...

I read the link, and it appears that it is a little misleading.. as it doesn't make this point clear.

Just for information, not trying to start any arguments.. and I believe my expert friend, and the online solar permit package (that I shared my exact model number of cutler hammer breaker box) are correct.

FYI... I am proceeding with my planned install of 60A solar feed circuit breaker from my 10Kw panel installation, in my 200A cutler hammer box with 225A max capacity.
 
Some utilities have a newer style meter they plug in that has a line tap built in for solar, probably worth asking if they know about it. Based on the drawing they gave you I doubt if they're on the cutting edge of technology, but who knows.

PG&E in my area offers line-side tap. I think it is an insert between socket and meter. But limited to 40A

Newbostonconst, thanks for the link, it is very good. I can see now that I had the 120% rule all wrong. I have a 150a main panel, meaning my overhead is 180a. Subtract 60 amps for solar, I would have to downgrade my main breaker to 120a, probably not going to work.

(As 12kW said) Double-check labels inside panel for rating of busbar, vs. rating of main breaker. Sometimes busbar rating is higher.
But if panel has main breaker in middle of busbar rather than one end, limit is 100% not 120%.

150A x 120% = 180A, 30A limit on PV breaker
30A x 80% = 24A max continuous on the circuit
240V x 24A = 5760W max inverter rating.

"120A breaker" may not be an option for your panel.
My QO 225A panels accepts 200A and 150A main breakers, 125A panel accepts 125A, 100A, 70A

Some brand panels (Zinsco) should be replaced regardless because they are a fire hazard.

Had expected "Line" side of meter to go toward utility line, but drawing clearly says PV goes to "Line" side. (question #1)

When summing the ratings of the breakers, shouldn't I use the breaker where the PV energy enters the system? In my case that is 60a. The incoming breaker to the sub panel is 100a. So the busbar rating of the sub panel should be at least (100 + 60) * 1.2 = 192 ? And then the main panel busbar rating should be (150 + 60) * 1.2 = 252 ? Am I understanding this correctly? Are the busbar ratings listed on the panel somewhere? Thanks-Mark

(100 + 60) / 1.2 = 133A minimum busbar rating.
(150 + 100) / 1.2 = 208A minimum busbar rating.
It's not the PV current, it's the breaker rating that's used to calculate required busbar.

If your PV needs 60A breaker, that backfed breaker plus main breaker of the panel it goes to are supposed to be no greater than 120% of panel busbar rating. I say each breaker could be 100% of busbar rating for a total of 200%, so long as located at opposite ends of busbar, and nothing bad would happen. The current from two sources subtracts, doesn't add. But NEC doesn't permit that, out of fear PV breaker will later get relocated next to main breaker so the current then adds.

But now you have a 100A breaker in main panel feeding sub-panel. It is backfed with PV, so it must be at far end of main panel. And guess what?? Sum of main panel main breaker and this 100A backfed breaker is limited to 120% of busbar rating.

A single 15A breaker for PV inverter going to one sub-panel (e.g. condo or industrial unit) could require a cascade of panel upgrades all the way through a 600A main panel connected to utility.

If you put in a Squared D QO 225A main panel with 150A main breaker (or other brand with similar values), 225A x 120% = 270A, 270A - 150A = 120A allowed backfeed. That will meet your needs, and perhaps upgrade of sub-panel (or smaller breaker feeding it) as well.

But what could be convenient is to have utility and main meter go to a 200A breaker or fused disconnect. From there, branch wires to a main panel with 150A or whatever main breaker. Also branch to fused disconnect for PV meter etc.

Once you have separate disconnect & OCP for the utility without a busbar, you can branch with wires to several panels (using Polaris or other suitable multi-tap connector). Current in each wire is limited by main breaker on each panel or by fused disconnects. Only the multi-tap connector might carry sum of currents from two sources (utility, PV) to multiple other loads, but could be 400A rated.

Nice thing about this configuration is you can disconnect power from utility to work on any branch or breaker box.

 
Back
Top