Need Advice on Mounting Solar Panels to Wooden Frame and Grounding Off-Grid System

[alexzan]

I am seeking advice and suggestions regarding an issue I am experiencing with my off-grid solar system. I have a non-isolated inverter (LV6548), and I recently noticed a voltage leak between the inverter ground and the earth (between 2 and 95v observed today). Upon investigation, I found that my PV panels, which are currently just laid on the ground, might be the source of this leakage. When I disconnect them the voltage goes away.

Concerns:​

1. Mounting Panels to Wooden Frame:
   • I am considering mounting my solar panels to a wooden frame to isolate them from the ground and potentially resolve the leakage issue. Would mounting the panels to a non-conductive, wooden frame help in preventing any grounding conflicts and leakage currents? What about when it rains, wont there be some conducting? I am not sure of the solution here.
2. Grounding the System:
   • My system is entirely off-grid, with no AC input (although I might consider adding a generator later). I am contemplating whether I should install a grounding rod or keeping it a floating system. Would grounding be beneficial in my case?
3. Grounding Rod Installation:
   • If grounding is recommended, could anyone provide insights or resources on proper grounding rod installation for an off-grid system housed in a metal shipping container?

Additional Information:​

• The inverter manual specifies that only certain types of PV modules are acceptable and warns against connecting any PV modules with possible current leakage, such as grounded PV modules.
• The system will remain off-grid, and I might consider adding input from a generator in the future.

Questions:​

1. Would mounting the PV panels to a wooden frame effectively prevent grounding conflicts and resolve the leakage issue?
2. Is installing a grounding rod recommended for my off-grid system, and if so, how should it be done considering the system is in a metal shipping container?
3. Are there any specific considerations or precautions I should take while grounding the system and mounting the panels to ensure safety and compliance with electrical standards?

I appreciate any advice, suggestions, or experiences you can share to help me address these concerns and ensure the safe and efficient operation of my solar power system. Thank you in advance!

 

[zanydroid]

Curious, how are you measuring the voltage?

Stationary system is required to have grounding rods compliant with NEC (and therefore you can follow standard DIY advice on how to do this for an outbuilding/house/whatever), and then this grounding system connected to a single N-G bond in the system (likely inside your inverter). The ground rod will pull the local soil/concrete/etc to the same potential as your neutral and ground which should get rid of that voltage reading.

Note that if you do this without grounding the PV frames the PV frames will likely then have similar VAC as what you read now relative to the ground.

Did you ground your PV frames back to the grounding terminal on the inverter?

There are plenty of threads here/YouTube videos talking about how the AC induced on the +/- conductors (from the fact that they are ungrounded/non-isolated) will be electromagnetically coupled onto ungrounded PV frames, with low enough impedance to be a shock risk.

Grounded PV module can mean several thing. One is when either of + or - is supposed to be bonded with ground. The vast majority of new PV today is ungrounded non-isolated PV. The inverter manual is NOT saying "don't ground the PV frames", not doing so is a code violation and none of the people I trust on the forum leave it ungrounded.

The right way to ground your modules is to either attach to metal racking with mounting clamps that are listed to also provide ground bond, and then bond the panel back to EGC using a listed lug (like a Weeb lug). Or put weeb lugs directly on the solar panel in the pre-drilled holes intended to accept them. #6 copper is the right size to use for exposed run (code requires oversize for durability) at the panels, and then you can transition down to #10 copper for most system sizes (good up to 60A per circuit).

Hopefully if you bond the solar panel frames (sitting on the ground) to your inverter then there is no longer a voltage between the earth and the inverter G/N.

 

[timselectric]

I am seeking advice and suggestions regarding an issue I am experiencing with my off-grid solar system. I have a non-isolated inverter (LV6548), and I recently noticed a voltage leak between the inverter ground and the earth (between 2 and 95v observed today). Upon investigation, I found that my PV panels, which are currently just laid on the ground, might be the source of this leakage. When I disconnect them the voltage goes away.

The Lv6548 bonds neutral and ground.
So, your system is not floating. But anything not connected to your grounding system is.
This is why you are seeing stray voltage from the panels.
All exposed conductive parts should be connected to the ground provided by the LV6548.
Including, but not limited to:
Panel frames, metal racking, metal conduit, metal enclosures, the shipping container, and the earth.

 

[alexzan]

I am measuring between sticking an electrode in the soil, and another on ground.

Thank you everyone for your input. Based on your advice and further consultation, here’s my refined plan to address the grounding issue:

1. Grounding Solar Panels: I will ground all the solar panels together. The grounding conductors from each panel will be securely interconnected and then routed to the common grounding bar within the breaker box.
2. Grounding Bar Connection: The common grounding bar within the breaker box is already connected to the grounding terminal of the LV6548 inverter/charger. All equipment grounding conductors will be securely connected here.
3. Ground Rod Installation: I will drive a grounding rod near the installation and connect it to the common grounding point within the breaker box, ensuring a safe path for any fault current to dissipate into the earth.

This should address the stray voltage issue and make the system safe and compliant. Does everyone concur that this plan seems sound, and did I overlook anything?

Also, I should be able to leave the panels on the ground at that point without the voltage leak (until I build a better frame for them)?

 

[timselectric]

I am measuring between sticking an electrode in the soil, and another on ground.

Thank you everyone for your input. Based on your advice and further consultation, here’s my refined plan to address the grounding issue:

1. Grounding Solar Panels: I will ground all the solar panels together. The grounding conductors from each panel will be securely interconnected and then routed to the common grounding bar within the breaker box.
2. Grounding Bar Connection: The common grounding bar within the breaker box is already connected to the grounding terminal of the LV6548 inverter/charger. All equipment grounding conductors will be securely connected here.
3. Ground Rod Installation: I will drive a grounding rod near the installation and connect it to the common grounding point within the breaker box, ensuring a safe path for any fault current to dissipate into the earth.

This should address the stray voltage issue and make the system safe and compliant. Does everyone concur that this plan seems sound, and did I overlook anything?

Also, I should be able to leave the panels on the ground at that point without the voltage leak (until I build a better frame for them)?

Yes to everything.
Except that fault current doesn't dissipate into the earth. It returns to the source, and trips the breaker to make it safe.

 

[zanydroid]

Grounding Solar Panels: I will ground all the solar panels together. The grounding conductors from each panel will be securely interconnected and then routed to the common grounding bar within the breaker box.

This is probably fine, but the EGC should follow the same path as the DC circuit, and landing it on the breaker box may not achieve that. I think my first choice would be landing it in the inverter (and splicing all grounds traversing that box I believe is required by code since there are current carrying conductors spliced inside the inverter).

 

[timselectric]

That is true. Code does require the PV EGC to be ran along with the PV conductors. All of the way to where they terminate at the PV receiving device.
But since this is off grid and probably not getting inspected. As long as everything is grounded, you'll be fine.

 

[zanydroid]

Electrically it is fine, but I assume the "bond all EGC" rule is to guard against some kind of maintenance brain fart. ?‍

 

[garetwo]

The inverter manual is NOT saying "don't ground the PV frames", not doing so is a code violation and none of the people I trust on the forum leave it ungrounded.

I am glad you said this, because I do not understand this part in the manual
COPIED
"WARNING: Because this inverter is non-isolated, only three types of PV modules are acceptable: single crystalline and poly crystalline with class A-rated and CIGS modules.
To avoid any malfunction, do not connect any PV modules with possible current leakage to the inverter. For example, grounded PV modules will cause current leakage to the inverter. When using CIGS modules, please be sure NO grounding"

I am clueless as to how one determines a PV module may have current leakage. My panels are all bonded together and run to the EGC grounding electrode, my friend's are on a wooden frame and NOT bonded together and are isolated from ground. It is in his manual the above warning statement is made. I am about to go look at what a CIGS module is as well.

Does this mean they should not be used at all? "do not connect any PV modules with possible current leakage to the inverter"

 

[zanydroid]

I am glad you said this, because I do not understand this part in the manual
COPIED
"WARNING: Because this inverter is non-isolated, only three types of PV modules are acceptable: single crystalline and poly crystalline with class A-rated and CIGS modules.
To avoid any malfunction, do not connect any PV modules with possible current leakage to the inverter. For example, grounded PV modules will cause current leakage to the inverter. When using CIGS modules, please be sure NO grounding"

I am clueless as to how one determines a PV module may have current leakage. My panels are all bonded together and run to the EGC grounding electrode, my friend's are on a wooden frame and NOT bonded together and are isolated from ground. It is in his manual the above warning statement is made. I am about to go look at what a CIGS module is as well.

Does this mean they should not be used at all? "do not connect any PV modules with possible current leakage to the inverter"

Keep in mind non-isolated inverters are the standard nowadays (EXCEPTION: probably a lot of microinverters are isolated because there are safety/code compliance implications of non-isolated that might opt them out of the exemptions they enjoy thanks to lobbying, but they may still do leakage checks through the panel). So by transitivity most modules should be compatible.

Read the specs I guess.

My algorithm would be:

• New solar panels that aren't suspiciously cheap because they use something exotic? No need to worry.
• Used stuff of unknown lineage? ? I think about 15ish years ago the negative and positive grounded panels started going away.
• If something feels like an outlier, or if you want help to validate how a panel can be used, post pictures of the panels to the forum.

If you're living in California, it's unlikely for a solar panel "deal" to be the last one, so better to do your due diligence than to pull the trigger before you know anything.

You can read about CIGS for personal education and edification but you're not going to be installing those in scale on a house. $$$

 

[garetwo]

If you're living in California, it's unlikely for a solar panel "deal" to be the last one, so better to do your due diligence than to pull the trigger before you know anything.

thanks, the trigger for installing was pulled long ago with no permit, so it is now in the "how do we make what is in place as safe as we can". I live 45 minutes from his home but I will be out there next week and take a picture - a LOT of things gleaned from this forum are being implemented.

This bonding/grounding issue is close to the final item to address. I am pretty sure the frames should be bonded together and tied to the system ground, I am just worried about bringing that back to the inverters - I guess this too is a trigger unto itself.

 

[garetwo]

You can read about CIGS for personal education and edification but you're not going to be installing those in scale on a house. $$$

I see this as true via this thread:

Why not CIGS?

Hi, just wondering about why CIGS panels haven't really taken off for RV use in particular. I've seen that with many discussions of CIGS flexible solar panels that folks seem to get them conflated with some of the drawbacks of flexible crystalline panels, focusing on the 'flexible solar panel'...

diysolarforum.com

I am off and running...

 

[garetwo]

The panel is a Mono-Crystalline CanodianSolar CS3N-390MS. How does one test for " possible current leakage" on a panel? Put a DVM from each output to the frame? If there is a potential, does that imply current leakage?

 

[zanydroid]

If the specs / panel type say it won't, then you don't need to do anything.

Installers won't bother with that, they just buy the right panels and throw them on.

Those are very new panels and have contemporary residential black on black styling, which all says vanilla modern electrical behavior. I probably won't even do due diligence check for grounding type if I bought them.

 

[zanydroid]

You would probably isolate one panel and measure resistance from DC+ to frame and DC- to frame. With appropriate penetration of any oxidation on the frame service to avoid false positive/negative

 

[garetwo]

You would probably isolate one panel and measure resistance from DC+ to frame and DC- to frame. With appropriate penetration of any oxidation on the frame service to avoid false positive/negative

That seems a good way to do it.

If there were a measured potential, would that not mean the panel is bad? How could that "hurt" an inverter? I can see that this would not create a short circuit but provide a measured voltage between all bonded components to the non-leaking leg. If those bonded frames were in turn bonded to ground (common to the grounded system in place), it would then become a potential at the inverter as well.

It's almost like "why bond/ground the panels at all?" Or, is really a lighting issue where this may be necessary?

 

[zanydroid]

That seems a good way to do it.

If there were a measured potential, would that not mean the panel is bad? How could that "hurt" an inverter? I can see that this would not create a short circuit but provide a measured voltage between all bonded components to the non-leaking leg. If those bonded frames were in turn bonded to ground (common to the grounded system in place), it would then become a potential at the inverter as well.

It's almost like "why bond/ground the panels at all?" Or, is really a lighting issue where this may be necessary?

Why do you care though unless you suspect a fault? This is not a regular maintenance thing unless a problem is suspected.

I'm not super clear on how it would hurt an inverter, haven't thought through that.

If there is a leakage then inverters are supposed to detect a ground fault leak and ideally shutdown the system. There is also a shock risk. If there is no leakage then the main shock risk is induced AC on some non-isolated inverters. Bonding will overpower this and make the frames more touch safe.

Wrt bonding to ground. This is a risk-benefit analysis, and the recommended installation by sane people represents what the industry has accepted as standard. The standard has changed over the years, and rebels have acquiesced or are unhealthily angry about it still.

If there is minor amounts of leakage then code/UL mandated leakage detection stuff at the inverter SHOULD kick in and raise an alarm. Either by notifying you directly and/or by shutting down the system so you pay attention. If there is major leakage then in some cases overcurrent protection will kick in. Not all cases though, because there is no meaningful overcurrent protection possible on the solar panel side in most cases, and the solar panels will always be producing. So to some extent safety is dependent on monitoring. I'm not an expert, but I think this is similar to non-solidly grounded electrical systems, which is allowed under NEC in some facilities but requires staff to monitor the alarms for ground faults, since ground faults do not automatically trip breakers when things are not solidly grounded (impedance too high to do this). This is a good tradeoff in some places where tripping the breakers will lose millions of dollars in product or damage $$$ of equipment as the power shuts off.

100% forget the heck about lightning with bonding & grounding unless you want to get a degree in lightning mitigation. Just parrot what Mike Holt wants you to do in basic situations and know when you need to hire someone to design it /s

 

[timselectric]

It's almost like "why bond/ground the panels at all?

So that they always remain safe to touch.

Or, is really a lighting issue where this may be necessary?

Grounding is for personal protection against electrical shock.
Lightning protection is a completely different and separate system.

 

[Onehand]

This thread has kicked me deep into the weeds. Am I to understand I need to pull ground wires along with the PV feed lines, back to the inverters? Or do I drive a ground rod at the PV panels and ground frames locally?

System design:

2 - 6000XP's grid tied (AC input with pass through output) neutral/ground bonding(s) in mains and loads panel.
8 - racks of panels, 4 panels each rack, 4 racks per inverter consisting of 16 panels per SCC (8 per MPPT) Solarever 455 watt panels.

 

[zanydroid]

This thread has kicked me deep into the weeds. Am I to understand I need to pull ground wires along with the PV feed lines, back to the inverters? Or do I drive a ground rod at the PV panels and ground frames locally?

There should be an article in resources and plenty of threads on this.

Do not ground frames locally, doubly so if there’s no path back to MPPT.

Pull ground along with PV to inverter.

Do not watch YouTube videos for how to do this. It is already a challenging issue to understand. Combine that with low IQ of many channels and it’s a recipe for disaster