Should I earth ground this system?

[HLD]

Just got off the phone with two different electricians and they both had completely different opinions on what to do. Each thought the other was a donkey.

In planning my system I though that driving an 8' ground rod into the earth and attaching a bare copper wire to the main panel would close the grounding portion of the system plan - which is:

• 12 panels mounted with s5 brackets to a corrugated metal roof that is on a wood frame on top of a shipping container.
• The shipping container is on concrete footings.
• The main panel is on the side of the container in a shed.
• The circuits run inside and along the outside of the container.

*this system does not need to pass any inspections, but it does need to not kill me and preferably anyone else.

1 electrician said do not earth ground battery powered off grid systems. Earth ground the panels for shock prevention.

The other said earth ground the main panel and ground ac out on the inverter to that panel.

Who's right and why is it neither of them? ?

 

[Hedges]

The objective is to make sure that when you're standing on the ground, if you touch any metal electrical box, the frame of a PV panel, or a tool plugged into an outlet, that you don't get a shock.
For that reason I think all of those should be connected together with wire, forming a single net, and go to a ground rod (or other more through grounding system.)
I think codes require higher voltage DC (such as 48V battery, maybe not 24V) to be grounded as well.
All plumbing in the house, such as gas and water pipes, get electrically tied to the same ground net.

 

[Skypower]

My electrician friend said that the inspectors like to see a ground line from the inverters with the PV lines to the array. His personal belief that this is wrong. On his personal system there is a ground rod at the array attached to the panel frames and local PV surge arrestors. There are no ground wire between the array and inverters. He also has another PV surge arrestor at the inverter that’s attached to another ground rod directly below the inverters. The thinking is to attempt to guide an energy surge at the array into the nearest earth path and not provide an easy path back to the inverters.

 

[Skypower]

Dupe

 

[Goboatingnow]

In general localised ground rods are a bad idea , as variable resistances or wire faults can lead to unexpected fault paths.

If your house has a protective earth system , I’d simply extend that as a ground , irrespective of not being grid tied.

 

[chess-equality]

... Each thought the other was a donkey.

...

Who's right and why is it neither of them? ?

Second opinions could be a little messy if you tell them they're one

 

[HLD]

In general localised ground rods are a bad idea , as variable resistances or wire faults can lead to unexpected fault paths.

If your house has a protective earth system , I’d simply extend that as a ground , irrespective of not being grid tied.

I think this makes sense. I should have clarified that there is no AC input. It is a metal box on concrete pillars. Nothing inside or anywhere near it. Considering that, would a ground rod act as the most effective form of protective earth system?

 

[HLD]

Second opinions could be a little messy if you tell them they're one

It's even worse when they know each other and are friends. That's when the real colorful language comes out.

 

[HLD]

My electrician friend said that the inspectors like to see a ground line from the system with the PV lines to the array. His personal belief that this is wrong. On his personal system there is a ground rod at the array attached to the panel frames and PV surge arrestors. There are no ground wire between the array and inverters. He also has another PV surge arrestor at the inverter that’s attached to another ground rod directly below the inverters. The thinking is to attempt to guide an energy surge at the array into the nearest earth path and not provide an easy path back to the inverters.

Is this considered a ground loop with an arrestor on one end of the loop?

 

[Skypower]

Is this considered a ground loop with an arrestor on one end of the loop?

Both ends of the PV have surge protection. They are only a closed loop when there’s an event.

 

[Goboatingnow]

I think this makes sense. I should have clarified that there is no AC input. It is a metal box on concrete pillars. Nothing inside or anywhere near it. Considering that, would a ground rod act as the most effective form of protective earth system?

Well if there is nothing else, then it has to be ground rods. Of course yiu could develop a complete isolated floating pv mains supply and since it’s not grounded , there’s no ground fault paths and hence no utility in a protective wire at all !!

 

[Hedges]

My electrician friend said that the inspectors like to see a ground line from the inverters with the PV lines to the array. His personal belief that this is wrong. On his personal system there is a ground rod at the array attached to the panel frames and local PV surge arrestors. There are no ground wire between the array and inverters. He also has another PV surge arrestor at the inverter that’s attached to another ground rod directly below the inverters. The thinking is to attempt to guide an energy surge at the array into the nearest earth path and not provide an easy path back to the inverters.

There might be a case where high currents from lightning through the earth decide to take an easier path up one rod to array, down wire to inverter, back down another rod. Maybe your friend is trying to protect hardware.

My belief is that PV at inverter can have a path to ground, so fault at array could drive PV frame to Voc. That is 480VDC in my case. I don't want to touch a 480VDC PV frame while standing on 0V Earth. So I believe in a wire from PV frame back to inverter chassis and a ground rod. Another rod at PV array seems good to me from safety point of view (even if it could sacrifice hardware in the event of lightning) by making sure dirt and PV frame aren't at different voltages.

Something like a downed high tension line can produce "step potential", where as you walk there is voltage from one footstep to the next.
If you ran a wire 100' away from you to a ground rod and held the wire, that would be One Giant Leap and a lot more voltage. So I'd rather that wire was well grounded at both ends.

 

[Skypower]

If you have ground mounted array without a ground rod and you’ve ran a ground wire out to it that tied to house ground, you’ve still just created a ground loop. The array mounting posts even set in concrete act as excellent ground rods. The PV system is isolated both positive and negative, it’s only a loop unto itself. A PV surge arrestor has very little connection to ground when it’s working correctly and the should be electrically isolated from the AC system anyway. Ground loops are overrated unless it’s neutral/ground loop. Now that can be trouble.

 

[Skypower]

There might be a case where high currents from lightning through the earth decide to take an easier path up one rod to array, down wire to inverter, back down another rod. Maybe your friend is trying to protect hardware.

My belief is that PV at inverter can have a path to ground, so fault at array could drive PV frame to Voc. That is 480VDC in my case. I don't want to touch a 480VDC PV frame while standing on 0V Earth. So I believe in a wire from PV frame back to inverter chassis and a ground rod. Another rod at PV array seems good to me from safety point of view (even if it could sacrifice hardware in the event of lightning) by making sure dirt and PV frame aren't at different voltages.

Something like a downed high tension line can produce "step potential", where as you walk there is voltage from one footstep to the next.
If you ran a wire 100' away from you to a ground rod and held the wire, that would be One Giant Leap and a lot more voltage. So I'd rather that wire was well grounded at both ends.

I see what your getting at but you would be surprised at how well the earth conducts. Years ago a guy had pump house that had two wires run to supply the both hot legs to get 240 to the pump. The problem was he didn’t run a ground or neutral. I noticed he had a light on in there. He had ran a wire to the well pipe to get 120! (more like 95volts).The 300’pipe from the house to the pump was plastic. Freaking scary.

 

[JRH]

Honestly, I have read so many very impressive sounding opinions on grounding my system that’s mounted on a untility trailer ,till my head spins….it’s the only part of this craft I really cannot point to a science type answer that's agreed on …..
as soon as I do get what sounds good, 3 OTHER scientist come and say the other scientist was smokin somthing…this is totally frustrating.
All I wanna do is not electrocute myself out in the driveway By accident..
J.

 

[Goboatingnow]

Honestly, I have read so many very impressive sounding opinions on grounding my system that’s mounted on a untility trailer ,till my head spins….it’s the only part of this craft I really cannot point to a science type answer that's agreed on …..
as soon as I do get what sounds good, 3 OTHER scientist come and say the other scientist was smokin somthing…this is totally frustrating.
All I wanna do is not electrocute myself out in the driveway By accident..
J.

All safety grounding systems are compromises. There are several ways to achieve the end result. You pick one, sometimes forced by “ code” otherwise it’s just a choice

 

[Hedges]

GFCI is a good idea, especially for outlets used when you will be outside the trailer.
Doesn't hurt inside, either, so long as not nuisance tripping.
(but don't use for refrigerator)

 

[stok]

found this image on homepower

 

[JRH]

found this image on homepower

I like that little electric guy… but every time I see him he’s either getting shocked or almost getting shocked…I wish he would just stop touching stuff…he’s a bit clutzy too, ya know.

 

[Danke]