New at this trying to pick your brains. I am working on a shed to home that will be 100% off grid and only powered with 120v. My plan was to hook up my all in one inverter to a sub panel powering the loads. I know to separate the ground from the neutral. My question is do I worry about grounding the sub panel if my inverter is already grounded via it's own grounding pin and rod? I assume so just want to make sure. Thank you!
Grounding a stand alone sub panel
- Thread starter Havean
- Start date
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
Please clarify. This seems contradictory.
Absolutely
Every electrical panel, solar panel frame, metal raceways, metal enclosues, metal framing must be grounded to protect you from electrical shock hazards.
Basically, if it's conductive and you can touch it. You want it to be safe to touch.
I will be just running 120v appliances off my solar system. Will not be apart of any water service ect. My inverter is only a 5000 watt 120v system
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
No house at all. Smaller appliances like tv, mini split and a fridge and a few misc items battery and inverter will have it's own structure
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
Then it's the main panel, not a sub panel. (The first means of disconnect)
Does this AIO provide a N/G bond?
If not, you will have to create one in the panel.
Either way, I would connect the ground rod to the panel. It's better equipped to accept the #6 conductor.
solarHandyman
New Member
- Joined
- Jul 22, 2020
- Messages
- 55
It's not a sub panel, it's the only panel! What model inverter?
The bond between neutral and ground is likely in the inverter on the AC output, so just run you a 4 to 6 AWG copper ground wire from your inverter to the ground bar on your loads panel AND to rods in the ground, otherwise you'll be floating! You still need two 10 foot ground rods a minimum two foot outside the building and at least 6 feet apart, all bonded to the loads panel and the inverter. How many solar panels and how many KWH are your batteries? You'll probably push more current to the batteries than anywhere in the AC side. If your solar panels on the roof of the building they can share the ground rods for the panel and inverter. If your solar panels are free standing away from the building with the loads and solar power plant, you need a rod for each array bonded to the other grounding rods.
The bond between neutral and ground is likely in the inverter on the AC output, so just run you a 4 to 6 AWG copper ground wire from your inverter to the ground bar on your loads panel AND to rods in the ground, otherwise you'll be floating! You still need two 10 foot ground rods a minimum two foot outside the building and at least 6 feet apart, all bonded to the loads panel and the inverter. How many solar panels and how many KWH are your batteries? You'll probably push more current to the batteries than anywhere in the AC side. If your solar panels on the roof of the building they can share the ground rods for the panel and inverter. If your solar panels are free standing away from the building with the loads and solar power plant, you need a rod for each array bonded to the other grounding rods.
Don't go yell at me lol
. New and learning. It's a PowMr 5000 48v all in one been great for years now that's been a dolly setup my electrician friend helped out with. Main panel it will be! I will have 20 kwh in batteries and about 5.5kw in panels. Panels will be apart of a car porttimselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
Do not add auxiliary (extra) ground rods at the array.
While they are allowed. They are not required, and definitely not recommended.
While they are allowed. They are not required, and definitely not recommended.
Thanks for that! Was wondering
This is my free 2 cents. If your inverter outputs AC current, then for 120 volts, you should have a hot wire (black), a neutral (white) and a ground.
At the point of origin for an off grid system with no utility grid tie and no tie in to a main panel, then the inverter should have the neutral and ground bus bars connected together. And the inverter also needs to be earth grounded to a grounding rod in the ground. Then the output from the inverter to the "main panel" will contain a hot, neutral and a ground coming from the inverter's outputs.
AC current flowing can be placed in plastic or metal conduit if outdoors, metal if indoors. Metal conduit carries the ground to the connected metal boxes and panels. Any use of PVC conduit indoors or out will require the use of a ground conductor inside the conduit from the panel to any metal boxes and any outlets, lights, motors, etc.
At the point of origin for an off grid system with no utility grid tie and no tie in to a main panel, then the inverter should have the neutral and ground bus bars connected together. And the inverter also needs to be earth grounded to a grounding rod in the ground. Then the output from the inverter to the "main panel" will contain a hot, neutral and a ground coming from the inverter's outputs.
AC current flowing can be placed in plastic or metal conduit if outdoors, metal if indoors. Metal conduit carries the ground to the connected metal boxes and panels. Any use of PVC conduit indoors or out will require the use of a ground conductor inside the conduit from the panel to any metal boxes and any outlets, lights, motors, etc.
I have a couple of questions about ground mounted arrays:
1. If an array is built from steel or aluminum and in the ground with or without cement, and there is a metallic combiner box connecting the panels together, are you saying in this case there does not need to be a grounding rod because the metallic components making contact with the ground function as the earth ground in the event of a PV+ short to ground or a lightening strike?
2. If an array is constructed of wooden materials and no metal from the array is in contact with the ground, are you saying then in this case also, no grounding rod at the array is needed and if the PV+ shorts to the solar panel frame, that is okay and no DC will pass into an unknowing person or animal?
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
In that case.
The earth connection is unwanted but unavoidable.
The best option is no local connection to earth, at the array.
But you have to do the best you can with what you have.
Correct
A ground rod at the array doesn't protect anyone or anything from a short. That's what the required EGC is for.
It's never ok, it means that you have a problem.
But, yes it's completely safe. As long as everything is properly grounded. (With the required EGC)
To further my understanding, you are saying that if in an array, there is a short from a PV+ to the panel frame, that is a problem but, it is safe as long as everything is properly grounded. This raises the questions: Is it "safe" because there will still not be current flow from PV+ to ground if a person touches the frame or the voltage and amperage will be too low when caused by a say single frame issue?
Secondly, what exactly does "everything is properly grounded" mean if the array itself is not grounded?
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
Properly grounded means it will have a Properly sized EGC connected to the PV frames, all metal supports, raceways, and enclosues.
This EGC is ran along side of the PV conductors back to the grounding system where the equipment is located.
I think that you are confusing "grounding" with "connecting to the earth". The grounding system is connected to the earth (in one place). But connecting to the earth is not grounding (electrically).
The grounding system begins and is created by the N/G bond.
The purpose of the grounding system is to provide a low impedance (resistance) path back to the source for fault current.
So that a fuse or breaker will clear the fault, before someone can get hurt.
The grounding electrode serves two purposes.
1. Primary function
Makes the earth safe, in relation to the local electrical system.
2. secondary benefit
Dissipation of atmospheric (static) charge.
Last edited:
Thanks for your responses.
Next question then:
A.) The array has a bare copper loop connected at both ends of the loop to a grounding rod in the earth.
B.) This loop is attached to the frames of all panels.
C.) The loop is also connected to the ground buss bar in the combiner box.
D.) The combiner box then has the following measurements by meter between these points: PV+ and PV- = correct Voc. PV- and GND = 0 volts. PV+ and GND = Voc.
E.) All panels frames PV+ and frame = 0 volts, PV- and frame = 0 volts.
Now the ground buss bar in the combiner has a green insulated wire traveling in the conduit along with the PV+ and PV- wires. The PV+ and PV- terminate on the inverter MPPT inputs (Sol-Ark 15K). Where should the green ground wire from the arrays terminate? At the AC utility ground buss OR?
Thanks.
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
I wouldn't have used bare copper. But it's fine if you can make sure that it's not in contact with the aluminum frames anywhere. (Galvanic corrosion from dissimilar metal contact)
Disconnect it from the local auxiliary ground rod, to protect your equipment and appliances from the gradient pulse created by a nearby lighting strike. Unless you live somewhere that never has lighting.
But even then, I still wouldn't recommend it.
It has no purpose. And only provides a false sense of security.
This should also be 0v.
But maybe the Sol-Ark isn't completely isolating the PV. (That would surprise me)
This doesn't make sense with the above. (Contradictory)
The ground bar in the Sol-Ark connections areas.
Sorry, this was a typo:
"All panels frames PV+ and frame = 0 volts
This doesn't make sense with the above. (Contradictory)+
The Sol-Ark 15K is throwing an immediate F23 error : TZ_GFCI_OC Fault when the PV switch is turned from off to on.
Testing of arrays completed so far:
Each individual panel was visually inspected for damage. Each individual panels was tested for power output, Voc, Vmpp ,Imp, with a solar multimeter. Each panel had voltage measured by connection to panel frame and then to panel MC4+ and MC4 - PV outputs. all voltages read zero. All panels had newly installed MC4 connectors and conductivity verified.
The serial string terminates in a Midnite Solar MNPV6. The PV+ and PV- go into a Midnite Solar 15 amp 600 VDC breaker DIN mounted to the combiner box. The output from this breaker goes into a 20 amp little fuse, DIN mounted, both the PV + and - negative are fused.
The PV - goes to an isolated buss bar where there is the connection to the DC ON/OFF switch made by Schneider. The PV+ from the fuse goes directly to the DC ON/OFF switch. The grounding of the array is as follows: 10 foot grounding rod with bare copper wire is attached to the frame of each panel by an approved grounding lug. Each end of the bare grounding wire terminates on the buss bar of the MNPV6 combiner box. Then a wire from the ground buss travels to the ground buss of the inverter.
Observed voltages: Between PV + and PV- is the expected Voc. Between PV- and ground buss is zero volts. Between PV+ and ground is the Voc voltage. Between PV+ and the metal frame of the array or any part of the combiner box is the same Voc voltage. This the same for all three arrays and I cannot figure out why the PV+ to ground has a measured voltage or where this connection is coming from is there is a jump between the PV- and the ground.
Any ideas?
timselectric
If I can do it, you can do it.
- Joined
- Feb 5, 2022
- Messages
- 18,620
It would seem that PV- is connected to ground, somewhere.
I would separate PV- connections and test in both directions for continuity with ground to determine where.
offgrid-curious
New Member
I'm in a similar situation as OP and could also benefit from clarification (I'm shining the timselectric bat signal).
I have an off grid system with a Victron Multiplus II 48/3000 inverter with the AC out back feeding 120v into a sub panel (6 spaces, 12 circuits). I have a small 2600W floating neutral generator that I plan to connect to the inverter. The default behavior of the Multiplus is to disable its own N/G bond when the AC supply is connected (in my case, I start up the generator).
I'm thinking that it would be just as well to fully disable the N/G bond on the Multiplus and put the N/G bond in the sub panel (read: main panel), then connect the sub panel directly to my DC grounding bus bar (which is connected to the EGC).
This saves me from using a N/G plug on the generator, and I think should also allow me to avoid having to bond the generator frame to the ground system. Does this make sense?
Similar threads
