• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

Risks of using cracked/shattered solar panel in series-parallel with other panels?

I'm still trying to understand, however, how that ground bond could possibly be effective in conducting AC current and opening an AC circuit when it is prepared strictly for the DC side of the system
The PV lines from the solar charge controller in the AIO have superimposed AC voltages on them due to design deficiencies.
This AC is capacitively coupled through the glass to the metal frames of the panels.
The frames have a measurable AC voltage on them if not grounded.
Sometimes it even hurts. See Adam’s video I posted above.

If the AIO were to develop a short and send full AC voltage to the panels, someone could die.
Thus, we ground the frames of the panels to carry the nuisance and AC fault current and keep everyone safe.
All metal in an electrical system should be grounded to avoid hazards.
 
I can't imagine how.
Because it's definitely obvious that I am not saying that.
We have been very clear in what we are saying. Yet, you still not getting it.
I'm beginning to wonder if you are genuinely misunderstanding the information provided. Or if you are just trolling the forum..
I think one of the issues here is that many people still work from the old understanding of what "grounding" means. I'm guilty of using the term loosely, as well, in part because I don't know of a better term. If you are talking about actual earth ground, then I think I will disagree with you and leave it at that. If you are talking about a so-called "equipment ground," which is what I have been assuming you are addressing, then I simply question the reasoning behind it, and particularly, in this case, I am questioning how it will help to open an AC circuit when there are no AC breakers on the DC side to begin with.

This is not trolling. It's a genuine question. But getting an answer for it seems rather elusive here, as people seem to want to focus on differences of opinion rather than on the electrical theory and rationale for doing things the standard way.

Take a look at Holtz's definition of "grounding" (starting at time 18:49) and notice the points of electrical theory which were formerly held by many that he is now dismissing as invalid. By his definition of "ground," the Earth is involved.


For my system, it was not designed to be grounded, per Holtz's definition, at all. The bond I have installed between the inverter rack and the solar array is not connected to ground. The rack is on insulative wheels, and the array was originally screwed into wood roof joists. That has now changed somewhat, as the setup was transferred to a new, concrete building, and is now anchored to metal trusses. So the array itself is grounded. But the bond back to the inverter is not required. It could be disconnected. It is that bond which I am addressing. And Holtz establishes plainly that if the inverter is not earthed, there should be no path via earth ground to it, i.e. no option of a hazardous parallel path (shock) via contact with earth ground and some electrified part of the system.
 
The PV lines from the solar charge controller in the AIO have superimposed AC voltages on them due to design deficiencies.
This AC is capacitively coupled through the glass to the metal frames of the panels.
The frames have a measurable AC voltage on them if not grounded.
Sometimes it even hurts. See Adam’s video I posted above.

If the AIO were to develop a short and send full AC voltage to the panels, someone could die.
Thus, we ground the frames of the panels to carry the nuisance and AC fault current and keep everyone safe.
All metal in an electrical system should be grounded to avoid hazards.
All of the above is quite possibly true if and only if the inverter is earth-grounded. If it is isolated from ground, then one could not possibly get shocked by touching those frames, as the earth would provide no parallel path for the current to return to its source. Electrons do not naturally seek nor "leak" to ground. They will only travel through ground to return to their source if their source was connected to ground.
 
All of the above is quite possibly true if and only if the inverter is earth-grounded. If it is isolated from ground, then one could not possibly get shocked by touching those frames, as the earth would provide no parallel path for the current to return to its source. Electrons do not naturally seek nor "leak" to ground. They will only travel through ground to return to their source if their source was connected to ground.
You are counting on perfect isolation of the inverter and everything it is connected to from earth ground to avoid a return path for the AC voltage that may be present on the panels? That’s very courageous.

How do you achieve such isolation in a house where every circuit is grounded? Does the house hover above the ground? :)
 
You are counting on perfect isolation of the inverter and everything it is connected to from earth ground to avoid a return path for the AC voltage that may be present on the panels? That’s very courageous.

How do you achieve such isolation in a house where every circuit is grounded? Does the house hover above the ground? :)
I'm in Thailand. There is not a single grounded circuit in the house. And it is entirely off-grid. In fact, the system is installed in a national park area where the government forbids any electrical utility. So there is no grid electricity for kilometers, and no one in that village has anything but generator or solar power to provide electricity in their homes.
 
For those following along, I found this to be helpful.

From another article:

It is a fact of nature that all power systems are grounded in some way or another. In terms of an electrical system, isolated or ungrounded refers to the notion that there is no intention of grounding. However, the conductors of all electrical equipment have a distributed natural capacitance between them and the ground. An ungrounded system is a system grounded through capacitance.

So I would take a high impedance AC voltmeter, dig it into the panel, and measure the voltage to a ground rod.
If no AC voltage is capacitively coupled, then it is truly isolated (at the time the measurement is made).
 
For those following along, I found this to be helpful.

From another article:

It is a fact of nature that all power systems are grounded in some way or another. In terms of an electrical system, isolated or ungrounded refers to the notion that there is no intention of grounding. However, the conductors of all electrical equipment have a distributed natural capacitance between them and the ground. An ungrounded system is a system grounded through capacitance.

So I would take a high impedance AC voltmeter, dig it into the panel, and measure the voltage to a ground rod.
If no AC voltage is capacitively coupled, then it is truly isolated (at the time the measurement is made).
Do you think Holtz would agree with this "fact of nature"? Holtz evidently held considerable influence over the revising of the NEC codes as one of their advisors/consultants. I don't see anything of this nature in his materials, nor does this "capacitance" theory make sense to me.
 
Do you think Holtz would agree with this "fact of nature"? Holtz evidently held considerable influence over the revising of the NEC codes as one of their advisors/consultants. I don't see anything of this nature in his materials, nor does this "capacitance" theory make sense to me.
I’m here to learn, and Mike Holt clearly has many good discussions and I read his site regularly.

Here is an article on transmission line systems discussing capacitance in isolated systems:
Scroll down to the isolated system section.
 
Last edited:
For everyone reading this thread, please understand that Will's word is not gospel. I'll make myself unpopular rather than having people get hurt.
The safest way is to follow NEC code and as you can see from this thread, they are still trying to wrap their heads around certain parts of electrical safety.
Don't use broken panels, they are not safe.
 
For everyone reading this thread, please understand that Will's word is not gospel. I'll make myself unpopular rather than having people get hurt.
The safest way is to follow NEC code and as you can see from this thread, they are still trying to wrap their heads around certain parts of electrical safety.
Don't use broken panels, they are not safe.
Most of what I have recently learned about grounding is from Holtz, not Prowse. Holtz influenced the current NEC codes to which you adhere. Will Prowse was learning some of the same things from the same source, which is why he recommended Holtz's video. Holtz himself implies that most electricians do not understand the basic principles of grounding and are basing their practice on outdated concepts that have been shown to be untrue. So I would go a step further here and say that even an electrician's word is not gospel. Be aware that the NEC codes on grounding have changed, and that there are electricians who seem unaware of the updates.

There are two sides to every coin, and grounding is no exception: Birds sit on high-voltage power lines every day without harm--and they are not harmed precisely because they are not grounded. :)
 
Let's forget about inverters for a moment and whether or not they can impose a hazardous voltage on the DC pv circuit. Let's look at just the PV circuit itself.

If there is an insulation failure in this circuit a low impedance connection will be established between the PV source circuit and the frames/structure. lnsulation failures in PV arrays are quite common. They can be from a broken module (as in your case), from a module edge delamination, backsheet pinholes, damaged module wiring, etc.

If this low impedance connection occurs you now have placed the structure/framing at the same potential as some point in the PV circuit. If one of the PV circuit conductors is grounded somewhere you have current flow. If you don't have a grounded circuit but instead get a second fault somewhere else in the circuit, you have current flow. Your body could be the ground path, either the first one in a grounded system or the second one in an ungrounded system.

The US electrical regulatory system protects against these scenarios by requiring all conductive enclosures and structures in electrical systems be bonded and grounded. That way you can't become the currrent path simply by touching the framing/sttucture. This code requirement has not changed, ever, AFAIK.

You can still become the current path if you touch an exposed PV circuit and a grounded structure at the same time. This is I think your argument against grounding the support structure. But consider that in this scenario a second fault somewhere could lead to you being the current path without having to touch an exposed circuit. You can become the path just by touching the framing.

Anyhow what you're suggesting is not allowed by the NEC, so it's a moot argument here. You can't legally do it in the US.

Perhaps you can in Thailand. If Thailand follows EU practice possibly. In the EU some appliances are not required to be connected to earth if they have a high degree of insulation, so called "double insulation". However I don't think that applies to module frames and structures, which are required to be grounded in the EU too AFAIK. I'm not an expert on EU electrical practice tho.
 
Let's forget about inverters for a moment and whether or not they can impose a hazardous voltage on the DC pv circuit. Let's look at just the PV circuit itself.

If there is an insulation failure in this circuit a low impedance connection will be established between the PV source circuit and the frames/structure. lnsulation failures in PV arrays are quite common. They can be from a broken module (as in your case), from a module edge delamination, backsheet pinholes, damaged module wiring, etc.

If this low impedance connection occurs you now have placed the structure/framing at the same potential as some point in the PV circuit. If one of the PV circuit conductors is grounded somewhere you have current flow. If you don't have a grounded circuit but instead get a second fault somewhere else in the circuit, you have current flow. Your body could be the ground path, either the first one in a grounded system or the second one in an ungrounded system.

The US electrical regulatory system protects against these scenarios by requiring all conductive enclosures and structures in electrical systems be bonded and grounded. That way you can't become the currrent path simply by touching the framing/sttucture. This code requirement has not changed, ever, AFAIK.

You can still become the current path if you touch an exposed PV circuit and a grounded structure at the same time. This is I think your argument against grounding the support structure. But consider that in this scenario a second fault somewhere could lead to you being the current path without having to touch an exposed circuit. You can become the path just by touching the framing.

Anyhow what you're suggesting is not allowed by the NEC, so it's a moot argument here. You can't legally do it in the US.

Perhaps you can in Thailand. If Thailand follows EU practice possibly. In the EU some appliances are not required to be connected to earth if they have a high degree of insulation, so called "double insulation". However I don't think that applies to module frames and structures, which are required to be grounded in the EU too AFAIK. I'm not an expert on EU electrical practice tho.
Perhaps your username gives some indication of the electrical-theory era behind your philosophy. Here is a tidbit related to the NEC codes coming from 2023.

This section reiterates the rule stated in sections 250.4(A)(5) and (B)(4) regarding the prohibition of using the earth as an effective ground-fault current path. Accordingly, it is not permitted to use grounding electrodes connected to equipment – such as ground rods – in the place of the equipment grounding conductor.

That doesn't sound very positive toward the use of an earth-ground to me. An "equipment ground" is not an "earth ground."
 
Perhaps your username gives some indication of the electrical-theory era behind your philosophy. Here is a tidbit related to the NEC codes coming from 2023.

That doesn't sound very positive toward the use of an earth-ground to me. An "equipment ground" is not an "earth ground."

Of course an equipment ground is an earth ground. The EGC is connected to earth is it not? The point of that code commentary is to emphasize that you cannot rely on a separate earth ground system to carry fault currents high enough to clear OCPDs when faults occur in grounded circuits.

Nowhere do any of the NEC code revisions or any of the folks you've referenced here say that it's ok to do what you claim to have done, which is to leave the PV structural system ungrounded. It is not allowed in the US.

Anyone with a broken module who was actually concerned about the safety hazards it could present would have by now simply removed it from service. Problem solved.

Instead you would appear to prefer to engage in insulting comments and esoteric discussion of US electrical code, which doesn't even apply in your location. Don't try to fit US electrical practice into practices used other countries, which have different methods to assure electrical safety.

If you want to leave your structure ungrounded in Thailand then you need to find justification for it in the regulatory system in force there, or just do whatever you can justify to yourself if there are no electrical code requirements. It seems to me that this latter case is where you are.

I'm done with this. By all means feel free to do whatever you want with your system. I wish you good luck with it.
 
skipping all of the BS, if it has not rained yet... get a windshield repair person there to do a UV epoxy repair. its good enough for windshield on moving cars... it will work for a solar panel... if it has rained... YMMV.

then beat the ever loving shit out of the snot nosed brat that did this.

pain is a great motivator.
 
Of course an equipment ground is an earth ground. The EGC is connected to earth is it not? The point of that code commentary is to emphasize that you cannot rely on a separate earth ground system to carry fault currents high enough to clear OCPDs when faults occur in grounded circuits.

Nowhere do any of the NEC code revisions or any of the folks you've referenced here say that it's ok to do what you claim to have done, which is to leave the PV structural system ungrounded. It is not allowed in the US.

Anyone with a broken module who was actually concerned about the safety hazards it could present would have by now simply removed it from service. Problem solved.

Instead you would appear to prefer to engage in insulting comments and esoteric discussion of US electrical code, which doesn't even apply in your location. Don't try to fit US electrical practice into practices used other countries, which have different methods to assure electrical safety.

If you want to leave your structure ungrounded in Thailand then you need to find justification for it in the regulatory system in force there, or just do whatever you can justify to yourself if there are no electrical code requirements. It seems to me that this latter case is where you are.

I'm done with this. By all means feel free to do whatever you want with your system. I wish you good luck with it.
An equipment ground is not an earth ground. An earth ground is just that--a connection to the earth, i.e. soil/clay/dirt/etc. An equipment ground is a connection between pieces of equipment. It is quite possible to have an earth ground without an equipment ground, and one can have an equipment ground without an earth ground. They are separate things.

For those who are unaware of this, an earth ground is really not a very good means of electrical fault protection. The entire point of a grounding system is to ensure enough current returns to the breaker to trip it and open the circuit so that the current stops. Earth is hardly as good a conductor as copper or aluminum. An equipment ground, which connects pieces of equipment with an actual wire, is far superior to an earth ground.

I am not seeking advice on US electrical practice. I am seeking information about actual electrical safety. Electrical practice and electrical safety are not directly related.

Many who think themselves patriotic may look down on the practices of other countries. I sometimes complain about Thai practices myself. However, you imply in your post here that "different methods" exist "to assure electrical safety." That said, perhaps people would keep an open mind regarding those methods. I want to hear actual facts, not merely codes. Codes are just policies--they are not laws of physics. It seems apparent that those policies have a large fan club here. I'm not here to pour cold water on that--I just am not interested in playing politics with the electrical principles. I simply want to know actual, real-life facts about how the electrical devices in my system may or may not function given a shattered panel.

I am telling you that my system is not grounded, and therefore poses less risk of one getting shocked via a parallel connection with ground. Instead of helping me understand the principles of my system, people instead want to force me to ground it, so that it then conforms to their ideas of how things should work (which then introduces said shock hazard that must be mitigated according to their SOP).

So be it. Perhaps I should have known I would not get much help here among people who are more interested in codes than in actual electrical principles.

There is a very good reason why I have chosen NOT to ground the system. I've not brought that reason into this thread, but have previously discussed it elsewhere here. That reason is lightning. It is a known fact that grounding does not prevent lightning damage, nor damage to equipment in the event of a strike. The NEC codes appear to acknowledge this. People ground their equipment to reduce the hazard of themselves being harmed by the lightning--but let the equipment suffer its fate. I have found, through experience, that having the only ground for miles around is an excellent lightning rod, and will increase the odds of a lightning strike substantially. For this reason, I prefer isolation.

Many refuse to accept that there are indeed two major systems to protect oneself from electrical harm: grounding and isolation. The two do not mix. One cannot go partway with either of these, it's all or none. I believe that the grounding method is easier to implement successfully. The isolation method can be more technically challenging. But both have their pros and cons. None should be shamed for intelligently choosing one over the other to fit one's particular circumstances--yet the shame game is quite active here among those who think their chosen method superior in every case. It is not superior in every case. There are exceptions to every rule, and the electrical situation in Thailand and in other developing countries certainly provides for some exceptions.
 
still waiting to get shocked... alarm-ism 3rd world countries learn all of this the hard way and guess what... no huge dirge of dead Malaysians with solar panels
I have yet to be shocked as well, and my system isn't even grounded! Imagine! I wonder how the squirrels on the power line manage to run along it without harm....shouldn't they be better off if they were grounded? Or maybe they would just be better off if the line they were running on was grounded....hmmm.
 
You seem to have some kind of chip on your shoulder or are otherwise deliberately misunderstanding me.

I am the first to say that the US electrical code is not gospel. That is why I said you need to follow the standards in your jurisdiction, if any. I specifically said there is more than one way to assure electrical safety, and suggested the EU "double insulation" method as an alternative example. What you don't want to do is attempt to mix safety standards that have evolved over many years of experience in different ways. That's dangerous.

But you failed to state until very late in this discussion that you were not in the US, did not have a grounded system, and were not required to have one. Instead you made multiple references to the NEC, implying that the NEC applies to your situation. It does not. Then you complain that folks are being US centric when they respond that what you are doing does not meet US code.

The NEC is a prescriptive code, if it applies to you you must follow it. In the NEC the EGC must be connected to ground. So in the US an equipment ground is an earth ground.

If you don't want to ground your structural system, feel free. You have your reasons. I wouldn't do it even if I wasn't subject to the NEC because I've seen too many faults between PV circuits and structures and want to be sure that when I, one of my family, or animals touch my structure or a module frame we're not going to see 500V potential relative to earth ground. That's lethal. Besides, my stucture has about 200 feet of steel embedded in soil so it is grounded whether I want it to be or not.
 

diy solar

diy solar
Back
Top