Lightning Grounding

[AgroVenturesPeru]

I feel this video deserves more discussion. Especially the part starting around the 50 minute mark. @Will Prowse recommended watching it in one of his recent videos.

I also feel this video deserves more explanation. I've always been told to use copper grounding rods, and even bought a kit with different conductor gels, sodium bentonite, concrete boxes, and 5/8", 8ft long copper rods with their wire clamps.

After watching this video, it sounds like I shouldn't have made the purchase.

If anyone else is knowledgeable on this topic, please elaborate. After watching the video, I'm left with the conclusion that "bonding" is the correct alternative to grounding.

 

[FilterGuy]

Bonding and grounding serve two distinctly different purposes. They are not alternatives to each other.

Bonding provides a low impedance path for a short to clear a fault.
Grounding keeps the circuits at the same voltages relative to the earth.

Grounding can be helpful in nearby lightning strikes (if done correctly). However, when done incorrectly, it can be damaging to equipment.
Grounding can not help in a direct lightning strike.

Sadly, the internet and even some professionals provide misleading and even dangerous information about grounding and since so many people do not understand the fundamentals.... they take the misinformation as gospel truth. Part of the problem is that the term 'ground' is used for things that have nothing to do with earth ground.... this creates confusion that supports the misinformation.

If anyone else is knowledgeable on this topic, please elaborate.

Grounding Made simpler - Part 1: AC/houshold grounding

To get the paper, click on the orange button at the top of the screen. The subject of grounding is a complex, multifaceted subject, that is often treated as an after-thought but needs to be considered from the beginning of the design and build...

diysolarforum.com

Grounding Made simpler - Part 3: Solar Panels

To get the paper, click on the orange button at the top of the screen. The subject of grounding is a complex, multifaceted subject, that is often treated as an after-thought but needs to be considered from the beginning of the design and build...

diysolarforum.com

 

[MrM1]

Looking at your diagram, I am wondering 2 things.
1. why wouldn't you need to bond the two grounding electrodes together with a copper wire? Wouldn't this create a 2nd grounding location for the home?
2. by bringing the grounding conductor into the house (as well as to the grounding electrode) does this "invite" the lightening strike into the home if the solar panels are hit? I believe your illustration would be for a near hit, not a direct hit?




Not a challenge, I am truly trying to learn. I have no idea. Is this acceptable?

 

[KaFun]

Hey, this is a great video! Thanks for bringing this in!

I am not an expert as Tim is, but I hope, I can shed some light on it:
Ground WOULD solve any fault problem, if dirt would be as highly conductive as e.g. copper. The reason grounding does NOT solve the problem is explained in the video around 38:26 (the guy kneeling next to the pole). If dirt would be conductive as copper, there would be no voltage building up in circles around that pole because it would be directly guiding the fault current back to its origin and the fuse would trip.

BUT AS DIRT IS NOT HIGHLY CONDUCTIVE, GROUNDING DOES NOT PREVENT FATAL ACCIDENTS! Just as explained in the video.

Here in Germany, each installation has to be secured with an earth leakage circuit breaker (ELCB). For that to work, you run three wires to each piece of equipment that has metal parts one could touch. Two wires are the "work" wires, one is "protective earth" (PE - yellow/green in Germany). The "work" wires are the Live wire (L) and the Null wire (N). They provide power and return and carry the load current during normal operation . The yellow/green wire (PE) is connected to all of the metal parts in that piece of equipment.
In case of a fault, one of the "work" wires touches any metal part of the equipment. Now the voltage splits and a part (or all) of the current runs to the metal part and thus to through the PE wire back (and one could get an electric shock, if he touches the metal part). In other words, the PE wire carrys parts of the load as well. Now comes the important part: The "work" wires run through the ELCB. The ELCB checks if all the current that goes out the live wire comes back on the return wire. If so, there is no leakage, i.e. no fault that could cause electric shock.
In case of a fault, parts or all of the current are carried by the PE wire and hence is missing in the comparison "Live wire vs. Null wire" in the ELCB. If there is a difference (in residential housing of 30mA), this ELCB then trips disconnecting BOTH wires and effectively disconnecting the faulty equipment. In modern German installations there are usually several circuits each with an ELCB. This has the advantage that if your wife's hairdryer falls in the bathtub, only the bathroom circuit ELCB trips disconnecting the bathroom, but the lights in the basement running on a different circuit with its own ELCB will still be working. So you can take the hairdryer out of the bathtub (i.e. clear the fault) and switch back on the ELCB even in the middle of the night with no flashlights.

Hope, this helps.

 

[FilterGuy]

1. why wouldn't you need to bond the two grounding electrodes together with a copper wire? Wouldn't this create a 2nd grounding location for the home?

That would certainly be better, but it would still invite a large strike to flow through the home. (The energy from voltage differential created the near strike would split between the paths.)

I believe your illustration would be for a near hit, not a direct hit?

That is correct, A direct hit is going to destroy things and the grounding will do little to change that.

2. by bringing the grounding conductor into the house (as well as to the grounding electrode) does this "invite" the lightening strike into the home if the solar panels are hit?

Short answer: No.

Another misunderstanding is that conductors will 'attract' lightning better than non conductors. It turns out that the overwhelming factor in what gets hit by lightning is proximity. Conductive poles are only marginally more likely to get hit than non-conductive poles. If you have a wood pole and metal pole in a field the one that will get hit is the one that is closer to the lightning bolt.


The second diagram in your post is misleading. By code the panels, inverter, and combiner box must all be tied to the primary grounding electrode at the service entrance. I modified the diagram to show this:


Except for the bonding between the electrodes, this makes the diagram essentially the same as mine.

As an aside: The auxiliary electrode is not required by the NEC and therefor has no requirements on what it has to be. You could stick a 10 penny nail in the ground and call it a auxiliary Grounding Electrode if you want. Mike Holt (The guy in the video) is trying to get the NEC to remove the allowance for the auxiliary electrode and I hope he succeeds. They are a bad idea

Lets look at what happens in a nearby strike that creates a potential voltage difference across the ground.


Even if the auxiliary electrode is bonded to the primary electrode, the surge will go through the house as well.

 

[MrM1]

A house is supposed to have only one grounding electrode bonded to the main panel. And so, at least in theory, (or maybe in Code) all grounding electrodes are supposed be bonded? That is, there is not to be a secondary grounding electrode say on the other side of the house?

So is the secondary electrode just for lightening and not for faults and not necessarily a good idea?

And So you're saying it is best to NOT have the 2nd electrode and to bond the panels on the roof with a grounding conductor thru the combiner and to the main panel grounding electrode bond?

 

[time2roll]

My entire house has one ground rod. All items including solar have grounds bonded to the single ground rod.

Not sure why a second ground would be added or needed except for convenience of installation. Looks like bad juju to me.

 

[DerpsyDoodler]

A house is supposed to have only one grounding electrode bonded to the main panel.

correct. A second grounding rod is supposed to be bonded to the first and acts as a parallel path to earth.

And so, at least in theory, (or maybe in Code) all grounding electrodes are supposed be bonded? That is, there is not to be a secondary grounding electrode say on the other side of the house?

So is the secondary electrode just for lightening and not for faults and not necessarily a good idea?

And So you're saying it is best to NOT have the 2nd electrode and to bond the panels on the roof with a grounding conductor thru the combiner and to the main panel grounding electrode

its not a bad idea, you just need to make sure you don’t introduce ground loops.

 

[FilterGuy]

A house is supposed to have only one grounding electrode bonded to the main panel.

The ideal situation is that all Equipment grounding conductors go back to one point at the service entrance. From that single point a single Grounding Electrode Conductor goes to all grounding electrodes. If you can't have a single grounding electrode conductor going to all of the grounding electrodes, a parallel grounding electrode from the single grounding point to the other grounging electrodes is acceptable. This is not perfect but not horrible. Unfortunately, the NEC allows for a 2nd 'auxiliary' grounding electrode that ties to the equipment grounding conductors someplace other than the one common point.... this is a bad allowance in the code.



Note, in order to get down to the 25ohm grounding requirement, multiple electrodes are often required at the service entrance. In fact, some inspectors are known to just require the 2nd electrode regardless. When there are multiple grounding electrodes to meet the 25ohm requirement, they should be wired in daisy chain as shown in figure 3 above.

In the diagrams of the houses in above posts, the auxiliary grounding electrode is optional but should not be installed.

Where the NEC requires a second Grounding electrode tied to a second point on the system is for a branch circuit that goes out to a different building. If it is a single circuit, the 2nd grounding electrode is not required. However if the circuit goes to a breaker box in the second building and then splits into multiple circuits, a second grounding electrode system is required at that point. This gets tied to the same Equipment Grounding Conductor system as the grounding electrodes at the service entrance. Since it is a code requirement, it should be done..... but I am not convinced it is a good idea.

 

[Notes]

This is my proposed setup. No grid power available. The array is 225’ from power shed that will have inverter, scc, batteries, and genny. 120 volts from shed to cabin is 75’. Main grounding rod is at cabin. I was going to add ground rod at array and power shed and connect all three. Is this not recommended? I have all equipment and wire mounted and trenched. I just have to connect the dots. I forgot to draw shunt in diagram.

 

[DerpsyDoodler]

The ideal situation is that all Equipment grounding conductors go back to one point at the service entrance. From that single point a single Grounding Electrode Conductor goes to all grounding electrodes. If you can't have a single grounding electrode conductor going to all of the grounding electrodes, a parallel grounding electrode from the single grounding point to the other grounging electrodes is acceptable. This is not perfect but not horrible. Unfortunately, the NEC allows for a 2nd 'auxiliary' grounding electrode that ties to the equipment grounding conductors someplace other than the one common point.... this is a bad allowance in the code.

View attachment 44685

Note, in order to get down to the 25ohm grounding requirement, multiple electrodes are often required at the service entrance. In fact, some inspectors are known to just require the 2nd electrode regardless. When there are multiple grounding electrodes to meet the 25ohm requirement, they should be wired in daisy chain as shown in figure 3 above.

In the diagrams of the houses in above posts, the auxiliary grounding electrode is optional but should not be installed.

Where the NEC requires a second Grounding electrode tied to a second point on the system is for a branch circuit that goes out to a different building. If it is a single circuit, the 2nd grounding electrode is not required. However if the circuit goes to a breaker box in the second building and then splits into multiple circuits, a second grounding electrode system is required at that point. This gets tied to the same Equipment Grounding Conductor system as the grounding electrodes at the service entrance. Since it is a code requirement, it should be done..... but I am not convinced it is a good idea.

Maybe it’s important to restate that regardless of how many rods you have, your connection to those rods should be from a single point and through a single conductor (known as the equipment grounding conductor or EGC). Tie ground circuits back to the panel as they should be, where-in they will be connected to the rods through the EGC.

 

[FilterGuy]

This is my proposed setup. No grid power available. The array is 225’ from power shed that will have inverter, scc, batteries, and genny. 120 volts from shed to cabin is 75’. Main grounding rod is at cabin. I was going to add ground rod at array and power shed and connect all three. Is this not recommended? I have all equipment and wire mounted and trenched. I just have to connect the dots. I forgot to draw shunt in diagram.

• By code, the panel frames, all ground wires in the power shed and all ground wires at the house should be tied together. I would definitely do this.
• There will be a breaker box at the house, so code will require a grounding electrode of 25 ohm or less there. I would definitely do this.
• I would have to know more about the power shed and look at the code, but I am pretty sure code would require a grounding electrode there....but to tell the truth I am not sure I would install it if I am not going to have the place inspected.
• Your ground mount array may be naturally grounded through the mounting rack and I can not recall off hand what the NEC requirement is.
  If the place is not going to be inspected, I would be tempted to not add grounding electrodes at the array.

 

[FilterGuy]

your connection to those rods should be from a single point and through a single conductor (known as the equipment grounding conductor or EGC).

A small nit: In the NEC, the wire from the common point to the grounding electrode is called a 'Grounding Electrode Conductor' (Not EGC). The wires that daisy chain between the electrodes are called Grounding Electrode Jumpers.

 

[DerpsyDoodler]

A small nit: In the NEC, the wire from the common point to the grounding electrode is called a 'Grounding Electrode Conductor' (Not EGC). The wires that daisy chain between the electrodes are called Grounding Electrode Jumpers.

Happy to defer to the superior knowledge you offer.

 

[FilterGuy]

Maybe it’s important to restate that regardless of how many rods you have, your connection to those rods should be from a single point and through a single conductor (known as the equipment grounding conductor or EGC). Tie ground circuits back to the panel as they should be, where-in they will be connected to the rods through the EGC.

Yes.... they should be connected to the common point through one conductor.... but the NEC does allow multiple conductors from the common point to the electrodes... Having said that, multiple Grounding Electrode Conductors should be avoided whenever possible.

 

[DerpsyDoodler]

Yes.... they should be connected to the common point through one conductor.... but the NEC does allow multiple conductors from the common point to the electrodes... Having said that, multiple Grounding Electrode Conductors should be avoided whenever possible.

I agree. Multiple can lead to confusion and thus the accidental creation of ground loops.

 

[FilterGuy]

but to tell the truth I am not sure I would install it if I am not going to have the place inspected.

To be complete, I should probably give the arguments *for* adding the grounding electrodes at the array and the power shed.

When you have a long run of wire, the natural inductance of the wire makes it difficult for high frequency current to pass. A pulse from a lightning strike is very high frequency (but not AC). In the event of a nearby lightning strike a very high frequency electromagnetic pulse is generated and this can create a high frequency pulse on the wiring. The theory is that by having grounding electrodes at the shed and at the array, the pulse can drain off to ground and is not impeded by the inductance of the long run back to the house.

So.... the choice is between two evils: The pulse generated by the electro magnetic pulse through the air or the pulse created by the voltage potential across the ground. Both are bad but I can't prove one is worse than the other.

 

[AgroVenturesPeru]

I feel this video deserves more discussion. Especially the part starting around the 50 minute mark. @Will Prowse recommended watching it in one of his recent videos.

I also feel this video deserves more explanation. I've always been told to use copper grounding rods, and even bought a kit with different conductor gels, sodium bentonite, concrete boxes, and 5/8", 8ft long copper rods with their wire clamps.

After watching this video, it sounds like I shouldn't have made the purchase.

If anyone else is knowledgeable on this topic, please elaborate. After watching the video, I'm left with the conclusion that "bonding" is the correct alternative to grounding.

Original Poster here. Still as confused as ever. Can anyone explain this in non-technical terms? For an Off-grid situation, please.

What I got from watching the video: Lightning grounds don't do anything to help, and can actually be harmful. You should bond all components together instead so that voltage passes evenly through everything.

In my situation I have metal sheets of roofing material. The metal PV panels are connected to aluminum structures, which are sitting on the metallic roof sheeting, and drilled metal screws through the aluminum structures into the steel roof supports, which are touching the steel support columns of the house, which are attached to a steel base plate, with 4 legs of half inch rebar reaching 50cm down into a concrete footing that sits inside of the earth....Furthermore, I'm planning to weld some thin steel panels to one interior wall of the house, and plan to mount most of the components onto these metal structures, which are themselves attached to the main steel support columns of the house.

So...how much good would it do to break up the cement patio, and run some thick gauge wire to a couple copper rods buried in the ground?

A) I originally was told that I need to connect the frames of all the PV panels, and run them to a common wire that connects to one copper rod.
B) Also was told I needed another cable touching all the components (MPPT, inverter, batteries, etc.) to run to a common wire to copper rod.

I bought these copper rods thinking of them as an insurance policy, to save my solar kit. I'd rather keep the copper rods as a precious metals investment if they're really not helpful.

 

[EJansen]

Great content on this thread, and great writeups @FilterGuy !

I wanted to throw a bit of a twist to the above with what I did for my shed. I never got a permit / electrical inspection, but I think I'm going to as sort of practice since I plan to do a 10KW array for my home next. Anyhow, please see attached. I have some questions based on the diagram that I was hoping some folks could help me out with:

As you can see in the diagram, I haven't grounded my panels. I have one 200w 24v panel that's run in parallel with two 100w 12v panels. The single panel is at the cusp of 50v, if going to 125% of PV Imax, but the two 12v panels are just over. Regardless, I'm going to ground the panels, but where in the diagram would you tie the ground in?

Another Monkey wrench to throw in: The house power comes from an extension cord that's plugged into the back of my garage a few feet away from my shed - so if that were to become unplugged, there would be no connectivity back to the main house ground.

Also, and the bigger question, and I know it differs for every state, but would this configuration pass inspection, or at least NEC (assuming that I had my panels grounded)? The questionable part is when the transfer switch is on solar. Though both solar and home connect to the same grounding bar in the transfer switch, which ties to the homes single grounding point, most of the time the shed would be on solar. So if I had a load that had a fault within the shed, and current then travelled over the ground trying to get back to the source, if that ground even made it back to the inverter (since I have a GFCI between the breaker box and the other outlets), I would think that inverter would kill power and go into an alarm state. I don't think the breaker would pop though. I kinda want to test it to see, but I figured maybe it would be better to just ask the pros Would the GFCI go first? I would think so. Anyhow, just curious if my setup is ok or not and was looking for feedback (even if it's something else that you see missing in the diagram)?


Thanks,
Eric

 

[time2roll]

Original Poster here. Still as confused as ever. Can anyone explain this in non-technical terms? For an Off-grid situation, please.

At some convenient point you should have one ground rod. Connect all to this one rod to be as safe as possible.

Nothing is going to make you and your equipment impervious to a direct lightning strike. Ground it and move on.