Grounding requirements for a ground based PV Array

Folks, I used an IronRidge design tool and I'm building a ground based PV array out of 2" galvanized steel schedule 40 pipe as the baseand universal strut as the panel mounting channels. Its understood that you have to 'ground' the solar panels... but I read somewhere that the metal framework holding up the panels also has to be grounded.

Does this mean that there has to be an electrical bond thru the whole structure (i.e. panels AND metal framework)? If so, are U bolts around 2" pipe to hold on the universal strut - i.e. u bolt touching 2" pipe and nuts touching the universal strut - considered 'bonded' for grounding purposes?

Advice on this welcome.

Here's a picture of the 2" galvanized steel framework to give you an idea but I haven't added the universal strut yet.

All bonding that I've seen requires a bolt or screw that goes "through" the support. So the U-Bolt wouldn't be enough alone, there'd need to be an anchor point drilled into the pipe.

Once I have things in place, I'll use self-tapping screws thru universal strut into the the 2" pipe so there is an electrical path thru all the universal struts to the frame. For the horizontal 2" pipe to the vertical 2" pipe, maybe I can use those IronRidge grounding straps.

Sigh... its always a bit more complicated (with DIY solar) than you think at first... that the entire frame + panels need electrical path for grounding. I'm not sure what risk there is but I do trust NEC electric codes as having a basis in reality/safety. Just hard to avoid 'denial' sometimes

OffGridInTheCity said:

Just hard to avoid 'denial' sometimes

Click to expand...

Just leave Egypt

It can be more complicated as DIY. But also to, it's usually a 1 (wo)man show as we usually do it completely ourselves. I suppose for the great price of a few cases of beer, you might wrangle in a neighbor or 2 to help out.

They just built a bunch of shade-type solar arrays at many of the parking lots at work. The small ones are 70kw arrays. Interestingly they are inverting to 3-phase and then transforming down to split-phase because many of the buildings don't have 3-phase service.

Anyway, if you have one of those around you at a mall or something I'd go check it out. They have grounding straps all over the place. I probably won't be at work for a few more weeks so I can't snap some pics.

Thank you all for helping me get over DIY worry and just doing it. Ultimately, I bonded the vertical/horizontal 2" steel pipe with these IronRidge bonding straps (very cheap)....




The cross universal strut to steel pipe with self-tapping screws like this....




And panels (to each other) and universal strut like this....




And it all came fine and I believe safe. Overall finished array looks like this





It turned out to be a reasonably low level skill set to build a 2" galvanized steel pipe array - the hard part was digging holes for the concrete but the rest was not actually that bad. Happy to share if anyone would like details.

Some of the reasoning behind the "bond everything with an earth strap system" rule is corrosion, eg rust, copper verdigris &/or aluminium anodizing.
Even if you have two bits of metal clamped together, they can build up corrosion & end up with significant resistance instead of a solid "short" circuit.
Seems local authorities prefer a bolt with star washer to penetrate corrosion,anodizing, paint (should have cleaned surfaces first but hey folks do stuff...)

On the house ground rods here, the copper wire attaches to a clamp that attaches to the (copper clad steel) rod. You then have to spray the done joint with zinc gal paint.
Apparently house earthing going bad is quite common.

Here in Australia, rules for solar are you have to ground panel frames & rails. They allow special "anodizing penetrating" shims between panel & rails so you only have to connect the grounding wire to the rail & the shim connects the panel to the rail. The shims have sharp raised points....
I'm not sure OffGrid's pics would be OK here unless he had star washers to penetrate the anodizing (maybe some other similar feature not shown in pics!)

Re bonding of panel frames & rails to house ground, rules here & my vote are yes, it's required. If lightning does hit nearby, you don't want differential voltage, eg the panels are different to the house = zap through some gear. If everything is bonded together, less chance of & lower differential volts.

Redpacket said:

They allow special "anodizing penetrating" shims between panel & rails so you only have to connect the grounding wire to the rail & the shim connects the panel to the rail. The shims have sharp raised points....
I'm not sure OffGrid's pics would be OK here unless he had star washers to penetrate the anodizing (maybe some other similar feature not shown in pics!)

Click to expand...

Yea - the grounding lugs that thread the 6 AWG wire / bond panels and universal strut have rough surfaces to scrape the surface of metal / make contact + the self tapping screws make a bond as they drill thru the metals. Those straps are regular IronRidge bonding straps (IronRidge is pretty mainstream system)and they are designed for self-tapping screws to bond so seems like that's OK as a bonding technique.

But I know what you mean - my electical enigneer friend commented that to him 'its all a bit ofhocus' because after 10or 20 or 30yrs the bonding will be suspect due to metal oxidation. Which I read as - one might consider re-bonding every x years.

Which makes me ask myself a question - is there a relatively simple way to test the bonding over time - maybe every other year? e.g. Hook up some current flow at one end and measure it at the other end - how much current would test for good bonding.. 20a or 50a or ??

OffGridInTheCity said:

But I know what you mean - my electical enigneer friend commented that to him 'its all a bit ofhocus' because after 10or 20 or 30yrs the bonding will be suspect due to metal oxidation. Which I read as - one might consider re-bonding every x years.

Which makes me ask myself a question - is there a relatively simple way to test the bonding over time - maybe every other year? e.g. Hook up some current flow at one end and measure it at the other end - how much current would test for good bonding.. 20a or 50a or ??

Click to expand...

Interesting.

Maybe testing would involve an Ohm meter. Probe before and after the joint to see what it reads. And then do that at each joint. This would narrow down any issues. Maybe(?)

Korishan said:

OffGridInTheCity said:

But I know what you mean - my electical enigneer friend commented that to him 'its all a bit ofhocus' because after 10or 20 or 30yrs the bonding will be suspect due to metal oxidation. Which I read as - one might consider re-bonding every x years.

Which makes me ask myself a question - is there a relatively simple way to test the bonding over time - maybe every other year? e.g. Hook up some current flow at one end and measure it at the other end - how much current would test for good bonding.. 20a or 50a or ??

Click to expand...

Interesting.

Maybe testing would involve an Ohm meter. Probe before and after the joint to see what it reads. And then do that at each joint. This would narrow down any issues. Maybe(?)

Click to expand...

* This link https://www.solarpowerworldonline.c...-testing-is-critical-for-solar-installations/ has this paragraph....
All good PV installers should be carrying out the following tests at the time of installation:
Ground Continuity Continuity testing is commonly used to verify grounding and bonding connections in electrical systems. Proper grounding of PV systems reduces the risk of electrical shock to personnel and the effects of lightning and surges on equipment.
Other ....

* Found a reference in UL 2703, chapter 13 titled "Bonding Path Resistance Test" - indicating there is such a test, but I can't access the document as it seems you have to buy it for a lot of $

* This has a resistance range - http://www.srbrowne.com/booklet/page04.html One of the paragraphs in the Testing section says this:

One lead of the ohmmeter is connected to a clean spot on the container, the other lead is connected to the paint grounding bus, metallic piping, or other fixed equipment. The measured resistance should be less than 25 ohms, and will usually be about one ohm.

Notice it gives "25 ohms" as max and "1 ohm" as normal - if I read it right.