NPFA 780 / LPI 175 Lightning Protection System for ground mounts

[tyab]

Sunking, I'm open to any recommendations you can give me to correct the misinterpretations I have made or make necessary repairs to above ground components. I'm new to this - and I know that you have spent a lifetime in these fields so I really would appreciate any knowledge transfer.. Below ground, I simply can't see any way I can get to that deep ring without massive work. I tried to follow 780 and LPI 175 as best as I understood what was in the specs - so any knowledge you can give me would be most appreciated. I have the printed 2014 versions.

For my entire solar project I used the printed version of NEC 2014 (and online 2011 since CA was still on 2011), and printed versions of NFPA 780 and LPI 175 and LPI 177 (I found LPI 177 to be a repeat of 175 with no new information). I tried at every step to follow and meet or exceed what I found in those specs. Any mistakes are mine. I also found very little on-line information about real world application of NFPA 780 in relationship to solar other than the references in the 780 spec. As one of the co-authors I am hoping that you have references that can really help everyone out understanding this subject more.

I can't find anything about class B conductors - I don't know what they are. Table 4.1.1.1.1 is class 1 ( <= 75' high structures) and table 4.1.1.12 is class 2 (> 75' high structures).

4.13.1.4 says that bolting, brazing, welding, or high compression connectors listed for purpose.... The ones I used are bolting and they are listed for class 2 (most of Robbins components are class 2). I could not find any elaboration on this section in Appendix A nor in LPI 175 on a best practice so I used the listed ones that Robbinslightning had.

LPI 175.135 says that concrete encased electrodes cannot have more than 2" of concrete around them and must be at least 20 ft long. With the Iron Ridge system, each pipe is in the middle of a 2' hole filled with concrete and for my engineering at least 3'-4" of pipe is in the hole (I have about 3'-8" in each hole). 780.4.13.3.1 contradicts this saying encased in at least 2" of concrete but both say at least 20' long. My guess is the LPI 175.135 is a typo given that 780 matches what NEC 250.52(A)(3) says. In either case I could not see how my 3-1/2' of pipe in concrete could be a considered a valid concrete encased electrode being so short. Even if one added up the four I have per mount that is still under 20' but I was under the impression it was a continuous 20' - not sections.

inetdog you are correct I used the wrong term - its has been a lifetime ago I took AC classes when I was first going for EE but then switched to computer science and never looked back. I will correct the document. Thanks!

[tyab]

DanS26 - you are correct. I would not expect this to be standard for most folks - odds are a system will never have a strike over its lifetime. I seem to remember reading somewhere that Germany requires LPS for anything over 10kW but I can't for the life of me find that now so take that as just hear-say and not true unless verified. One of the reasons I did it was I recognize that 10 years from now I may have zero interest (or ability) in trying to replace panels from the middle of a system - so I thought of it as an insurance policy that *may* prevent me from having to make repairs as I get into my 60's. After all of the reading I have done, I am seriously considering an LPS for our entire house but that is something I will pay to have done - I have no desire to crawl around on the top of a steep 2nd story roof.

The stainless was not that expensive - at least not as much as I thought it would be. I bought all my stainless online from metalsdepot.com - and I did two big orders so shipping was not bad considering the weight - it was shipped by truck. The key with that vendor is to order the maximum lengths they offer - the prices drop dramatically. I also used 1" stainless box tubing for all of my junction box supports on the system so that was part of the order (these came as 20' lengths). Stainless fittings mostly from mcmaster.com - the ones at the local True Value seem to be low quality and expensive. I did not have a band saw so I picked up a low cost one from Harbor Freight and it did the trick. The drill press I borrowed from a neighbor - he basically told me to keep it until he needs it - which he hasn't yet. Misc stuff mostly from Home Depot and other online vendors. I tried hard to find the best prices I could.

Butchdeal - I don't know what will happen with the PV panel edge about 10" from the lightning conductor. The closest microinverter is another 18" from that, so maybe 28" from LPS conductor to microinverter. My knowledge for this comes from 780.4.16.2.5 that has you compute if something needs to be bonded to the LPS due to distance, and the panel edge is easily in that distance. Everything for the panels is bonded via the IronRidge hardware, but I'm not sure if that meets 780.4.16.4. One of the reasons I did a NEC 690.47(D) even though CA was under the 2011 code is along the top of the rails with #6 copper is also a compliant bond according to 780.4.1.1.1.1. I also used #6 for all EGC's - the only EGC that s not #6 is the Enphase cable which has #12 in it, but the moment it gets to the first AC combiner box I take it to #6. All aluminum brackets that hold the AC combiner boxes have another #6 bond tying them to the rail and both horizontal pipes.

Maybe sunking has some real stories of what happens to systems that have a LPS system and if they work or not. I tried to follow 780 to the best of my understanding so I hope it does what it is suppose to do - prevent damage - but the truth is - I have no real idea if what I have will prevent damage or not.

I view the LPS costs as two things. The ground ring I was planning on doing even if I had no above ground LPS - to give all parts of the ground ring and the frame the same potential vs time ramp to prevent side flashing from a nearby event. I hope it will do that - hopefully sunking can elaborate if I am dreaming or not. The total cost of that for 620' of cable, rods, listed splices and bonds consumed about 2% of my total costs so it was worth it for me.

The above ground components are what I don't know if it was worth it. Again I tried to follow the spec and build stuff that met the sq in bounding requirements and used listed components and follow the examples in the spec but I don't know if it really will help on a close/streamer/direct hit. The cost there was cable, listed bonds, and all that stainless and was about 3% of the total system costs. However all of that stuff took (and I think this is close) about 20% of the total time - just insane hours drilling and cutting. If I had better metalworking equipment - that time would have been dramatically reduced but it was all manual for me - one hole at a time. Was it worth it? I learned a lot but will it really protect the system - don't know. I'm really hoping that sunking can give me specific things I have done wrong so I can correct them and learn more.

[Sunking]

Sunking, I'm open to any recommendations you can give me to correct the misinterpretations I have made or make necessary repairs to above ground components. I'm new to this - and I know that you have spent a lifetime in these fields so I really would appreciate any knowledge transfer.. Below ground, I simply can't see any way I can get to that deep ring without massive work.!

You do not have to. Gotta go right now, more later.

[Sunking]

OK fist thing to understand a LPS is not installed to save your equipment. It is there to help prevent your house from burning down. The best way to protect your house is not to have anything metallic bonded to earth on the roof. So if you place panels on your roof, then you are required to do something by code to give Lightning a planned path to dirt. If you follow NEC 2014 code you are asking to bring Lightning into you house and have a look around because bone headed Solar Advocates got a rule sneaked in allowing the HO to use a Equipment Ground Conductor to be used to bond the Solar Panel Frames, or use an Isolated Ground Rod. Pure stupidity and is gone in 2017 code cycle. Most jurisdictions amended it out. People Have lost their homes so contractors can cut corners and cost.

So by using a Ground Mount like you have done pretty much eliminates your chances of being struck by lightning if you take a few simple steps. It does not take anything elaborate.

1. Use concrete piers to support the Galvanised Poles. Concrete is a very good conductor and has massive surface area in contact with the dirt. It is called a Concrete Encased Ground Electrode in NEC and in industry a UFER GROUND. All the Bomb Bunkers in the USA only use a Ufer Ground. A Ufer Ground was invented during WWII by Hubert Ufer. Bomb Bunkers in the AZ desert had problems blowing up from static discharges because they used traditional methods to establish a connection to dirt. UFER came up with a very extremely effective method that did not require much cost. He had Contractors tie the reinforcement steel bars with Hog Wire, bonded a Copper to the rebar exothermically and tagged it before concrete was poured. Problem solved, there is no better electrode period.

You really did not need to do any of that work. At most if it makes you feel better drive a rod at the 4-corner of the Ground Mount array and bond each corner to the support post with a exothermic weld.. or even a irreversible crimp. The whole structure is already bonded together bu the construction method you used.

2. This s a big deal. Do not bond the Ground Mount electrodes to your home ground electrode. Treat it like a utility connection.

3. Where the positive and negative circuit conductors leave the Ground Mount Station use a good high quality 3-Mode SPD device (aka Surge Protector. The SPD's 3 modes are Positive to Frame, Negative to Frame, and Positive to Negative. Sized correctly will Clamp voltages to safe levels which might save the panels, and most importantly shunt Lightning current to DIRT.

4. Route PV +/- conductors together and take them to the entrance point of your home. Not only should this be Electric Power, but TV, Phone, SATV, Water line or anything that enters your home. You want it all right there together at a SINGLE POINT From that point you either bond it to a ground electrode like a metal water pipe, and install SPD's all bonded to the same point. That means another SPD on the two incoming PV circuit conductors. You do not need an elaborate Ground Electrode system at your home. It just needs to be code compliant and use best practices.

So what is Best Practice. Well it goes back to Single Point Entry Point and more importantly Single Point Ground of all electrical services and water pipes. For current to flow requires an Entry Point and an Exit Point. So if you bring in say your TV antenna on one side of the house, and water on another side, and power on yet another side, you have just invited Lightning inside to makes itself comfortable. Bond everything together at a single point, and even if lightning does strike, everything voltage rises and falls together at the same time including you. That means no potential difference in voltage and NO CURRENT. So lightning goes where you want it to go, stays outside in the dirt.

[bcroe]

A cheap bandsaw and drill press will handle the SS, but you need premium blades and drill bits. Bruce Roe

[tyab]

Thanks for the info sunking. I have some questions for you.

1. UFer. Why is there a 20' requirement? I'm referring to 250.52(A)(3) the continuous run must be 20' (but for rebar they allow a number of standard combination techniques like wire tying to get the 20'). 780.4.13.3 says the same thing, it must be at least 20' long. I don't see how my 3-1/2' of pipe in concrete meets that continuous 20' requirement.

780.12.5.1.1 says this for ground mount systems: "Systems that include a metallic structure shall be grounded in accordance with 4.13.4, utilizing a ground ring electrode encompassing the perimeter of each array". Am I reading the spec wrong? If I am following 780 - how can I not use a ground ring for a ground mount solar? What is really interesting is 780.8.4.1. (not that it applies to solar at all).

2. This is not an issue for me since my solar system is 400' from the main service panel and my ground ring does not leave the solar area. I'm running a 400' oversized EGC from the solar subpanel to the service panel per 250.66 and thus that subpanel has to have its own 250.52(A)(5) and 250.53(A)(2), and those must be bonded to the ground ring per 780.4.14 but I don't think you are referring to that.

3. I don't have any DC circuits except for the couple of feet from the PV panel to the microinverter.
4. same, no dc circuits.

For my above ground LPS stuff - I believe I am following everything in 780.12 and the referenced material in 780.4 including allowed location of strike termination devices based on my array tilt, recommended runs of down conductors, bonding of everything metallic using listed bonds and bond sizing, The custom brackets I made all met the 8 sq in of contact, etc.

I completely agree with you that there is no risk of fire on my system. Everything is metal and no PVC to melt/burn. Everything is bonded - excessively so in my case . There is no risk of loss of life - no sane person would be near that solar system during a storm. Virtually no risk to our home 400' away. Annex L talks about risk to equipment as part of the cost/benefit and I would agree that many would not find it justifiable.

But back to the main subject - what errors do you see in my implementation of 780?

[Sunking]

1. UFer. Why is there a 20' requirement? I'm referring to 250.52(A)(3) the continuous run must be 20' (but for rebar they allow a number of standard combination techniques like wire tying to get the 20'). 780.4.13.3 says the same thing, it must be at least 20' long. I don't see how my 3-1/2' of pipe in concrete meets that continuous 20' requirement.

Times how many pipes? Any contractor would have driven two grounds rods, or is not possible a Radial. The point here is it needs a planned path to dirt capable of safely handling the current. When Lightning discharges in dirt it spreads out in all directions along the surface.

2. This is not an issue for me since my solar system is 400' from the main service panel and my ground ring does not leave the solar area. I'm running a 400' oversized EGC from the solar subpanel to the service panel per 250.66 and thus that subpanel has to have its own 250.52(A)(5) and 250.53(A)(2), and those must be bonded to the ground ring per 780.4.14 but I don't think you are referring to that.

Do not run any EGC or any ground conductor between your house and solar array. If you do you just invited lightning into your breaker Panel. Does not matter if the circuit from the panels is AC or DC. they get treated the same. You treat it like any standard service from any utility. No utility code or NEC code would permit you to connect to the utility ground for a dang good reason.

The only time you SHALL use a EGC between structures is if there a common metallic conductive structure connecting the two like Conduit Raceway or water pipes. You want to use a good SPD at the Ground Mount, and another where it enters the home on the outside bonded directly below to GES. Refer 250.32(B)

AC or DC makes no difference. At the Ground Mount you bond N-G in the Inverter, run 3-wire to the house exactly like a utility does. Then at the house treat it like a Utility 3-wire service where you bond N-G again.

But back to the main subject - what errors do you see in my implementation of 780?

Other than the items I addressed and know about, you over killed it. You treated the ground mount like a building with occupants built with flammable materials. In other words a single shot to the head of a horse enough. You kind of used a machine gun with a 1000 round magazine and emptied into a dead horse. Having said that I am impressed as you did your homework and did good work.

[Sunking]

Any questions about what you see?

Both pictures are on golf course greens and the flagsticks were struck