Direct lightning strike, what will the damages be?

mcgivor · May 2017

Well it had to happen sooner or later, had a direct strike about 5 minutes ago, can't say for sure what was hit but 4 days ago I installed a mast 6 meters high (20 feet) about 10 meters (35 feet) away from the array, consisting of a 2 meter ground rod driven into clay soil and an air terminals mounted on top. Flash bang all over in an instant, it's raining and not going to investigate until morning, so far inverter is dead, that's all I know, was sitting about 10 meters away, just a loud crack and boom with white flash, charge controller still functions on display but who knows what other damage occurred, thankfully have spare components to replace everything should the need be, quite an experience, got me thinking am I attracting lighting or would it just have hit something else, like the wooden structure I was sitting in, about 5 meters high

BB. · May 2017

McGivor, I am very sorry to hear of your strike. Sounds like everyone is OK--That is a good start.

A (multiple) lightning rod and good lightning grounding system is best.

Grounding your negative battery bus to earth ground is also helpful.

Adding lightning/surge suppressors on the DC bus (at panel combiner and at panel into home) and AC output bus to electrical ground (and electrical ground to earth ground) can be helpful.

I like metal conduit for electrical wiring (ground the circuit breaker panel to earth ground, and all conduit should have good electrical connections throughout the installation). Each time you run a cable from the AC panel to your loads--it is an "antenna" that can get struck by (or pickup a nearby strike) and bring the energy to your loads and AC inverter output.

Years ago, I asked our forum founder (now retired) what failed most often from lightning strikes, he said that it was the output stage of the AC inverter.

When you have direct strikes--There is not much you can do to prevent damage (other than the first three steps above to reduce fallout) short of going full Tempest (military level protection against nuclear EMP).

And, anything electrical/electronic--Once stressed by lightning, it is possible that some of your electrical items can fail days/weeks/months down the road. You might get into the issue of (for example) a failed AC inverter output is repaired--Then months later, an internal control board (IC or Transistor, etc.) fails and needing to be sent out for a second repair. Vs just replacing the failed device with new.

A couple threads about Lightning:

Off Grid Grounding Technique?
Another Question, this time about Lightning

Note, the above are discussions, not a do A, B, and C--and you will be "safe". There probably is no such thing with lightning. Several different techniques are discussed--and a few of those posters even have experience with lightning. :cool:

And our host's FAQ:

Lightning Protection for PV Systems

From other past posts here, Windsun (admin/owner of NAWS), he said that most of lighting induced failures he saw were in the Inverters' AC output section.

Towards the end of this thread is a very nice discussion of proper generator grounding.

-Bill

jonr · May 2017

From what I know, a lightning rod X meters away should be at least X meters above the object being protected. Ie, locate your shorter one a couple of meters north of the panels.

There is something to be said for physically disconnecting things in storms. And surge protectors.

Dave Angelini · May 2017

Sorry to hear but a direct strike is hard to defend. On the inverter based power system, it is usually better to shut it down once you hear the strikes on an AM radio changing and getting louder. There usually is a warning and I always tell my clients to turn on the radio when the skies get dark. We design to leave it on but if you are home, by all means shut it down.

mike95490 · May 2017

Dave Angelini said:

Sorry to hear but a direct strike is hard to defend. On the inverter based power system, it is usually better to shut it down once you hear the strikes on an AM radio changing and getting louder. There usually is a warning and I always tell my clients to turn on the radio when the skies get dark. We design to leave it on but if you are home, by all means shut it down.

I ran out about midnight a couple years ago, to shut down the whole system, when a storm came through, Missed me, and everything powered up OK the next day

mcgivor · May 2017

The initial damage is to the inverter, everything else works, for now at least, it appears the newly installed rod did what it was supposed do do, but a nearby overhead feed to an outbuilding is too close and probably had an EMP induced spike, been removed.

Thanks @BB for the links, thoughts and suggestions, had read most of the posts previously, some very informative reading especially from Solar Guppy and Icarus, amongst others. Took the time to visit a couple of commercial solar farms to speak with their engineers on their thoughts for protection, both said the most important component was lighting rods, which I promptly installed, just didn't expect it to be put to test so soon.

@jonr The way it was explained to me by the aforementioned engineers, is along the lines of what you said, they stated the lighting rod should be at least 10 meters away and high enough to cover the arrays highest point within a 45° angle from the top if the air terminals, had read some similar information which was in agreement, but there is so much contradictory information on Web sites, chose to pay attention to real life experience, they also confirmed it isn't a gaurentee, after all lighting dose what it wants to. Surge suppression is ordered from Schneider along with a new SW 2524 inverter.

@Dave Angelini both you and jonr suggest disconnect which is always a good idea strangely there was rumbling in the distance about 20 seconds after the lighting, then everything was quite for 15 minutes then without warning the sound like a capacitor discharge arc, on a grand scale, there was thunder but not as loud as would be expected, perhaps because, as I suspect, it was directed to earth, wish I had physically seen it hit, there is no evidence anywhere. Dave do you use surge protection on the DC side from Schneider, would be nice to keep all the same manufacturer and I do have an excellent supplier I deal with locally in Thailand, thanks.

jonr · May 2017

Yes, 45 degrees - which isn't consistent with 6 meters high and 10 meters away.

Even better if there is a non-shadowing tree near your panels.

Dave Angelini · May 2017

I always specify at the array, at the DC buss, at the AC buss. Schneider does make them but a combo of the Midnite SPD and the Delta
at the array is easier for me to get and has worked on over 100 systems now! Knocking on wood

At my home I always have an Oak tree on the North side of the arrays and I have found tree damage in the past, so I think that is a good thing. With all of your different systems you do have alot of exposure. Better buy a case of SPD's or simplify

mcgivor · May 2017

jonr said:

Yes, 45 degrees - which isn't consistent with 6 meters high and 10 meters away.

It's actually mounted about 3 meters higher on a berm with a concrete base, 1 meter with a copper extension the air terminal screws onto, about 1 meter, so in total about 11 meters, but I'm definitely going to add more perhaps even higher to cover the 2 houses, should have done this during the winter, but better late than never.

mcgivor · May 2017

Dave Angelini said:

I always specify at the array, at the DC buss, at the AC buss. Schneider does make them but a combo of the Midnite SPD and the Delta
at the array is easier for me to get and has worked on over 100 systems now! Knocking on wood

At my home I always have an Oak tree on the North side of the arrays and I have found tree damage in the past, so I think that is a good thing. With all of your different systems you do have alot of exposure. Better buy a case of SPD's or simplify

You're right about simplifying, the plan is to consolidate with one system based on a Contex SW 2524 now ordered and protect it the best way possible, the systems grew over time but nature has taken care of one. Unfortunately it's difficult to obtain all the protective devices commonly available in the U.S. mostly due to language, the Schneider dealer has been exceptionally helpful but not evertything is available, probably due to certification, even the SW 2524 230 ordered has different ratings to the same model, same part number available in the U.S., here its rated at 3000W, as apposed to 2500W, see specifications in attached files. The model ordered is Contex 2524-230 E, Europe?

dennis461 · May 2017

mcgivor said:

....., got me thinking am I attracting lighting or would it just have hit something else, like the wooden structure I was sitting in, about 5 meters high

Sorry to hear your system got hit, and more concerned that you were nearby when it happened!
(I tend to stay far indoors when I hear rumbling.)

Yes you are attracting lightning (with a single tall mast).

Modern applications use devices to dissipate the static charge on a large front so that lightning will not occur.
This technique, as I recall, was developed when NASA had multiple strikes at the launch pads in the 60's 70's..
It is used in Colorado at ski resorts which are lightning prone areas.

There are variations by other vendors, but the approach is to use multiple, distributed grounded electrodes rather than a single point.
http://www.lightningprotection.com/lightning-protection/

NASA upgraded there systems around 2007 to include more towers and wires.
https://space.stackexchange.com/questions/2958/what-are-the-towers-around-the-spacex-launch-pad-used-for

The costs of the system needs to be balanced with the cost of repairs.
The NASA launch pads, tall structures on flat and are a prime target for lightning.
The ski lodges and lifts in Colorado are easy targets since clouds have to move up over mountains, increasing the static charge buildup.

mcgivor · May 2017

Thanks @dennis461 there is so much information on the subject from installing multiple pointed lightning eliminating devices which provide an umbrella effect, which others claim to be B.S., and just sales tactics, to multiple masts creating a Faraday cage. The single mast worked, I think, this time and my intention is to erect more, much higher masts around the structures. Some say/claim you cannot attract lightning, you can only offer it an easier path to where it wants to go, the ground, personally I think this is correct, the more paths offered to ground.the more likely it will choose said paths. I'm no expert on the subject, but it seems even the supposed experts can't agree, lightning is unpredictable and will do what it pleases.

Dave Angelini · May 2017

The different models world wide reflect that split phase is mostly in the Americas.

mcgivor · May 2017

Dave Angelini said:

The different models world wide reflect that split phase is mostly in the Americas.

Understood, but the single phase SW 2524-230 is available in North America but has different specifications, same part number according to the pdf files.

BB. · May 2017

Here is an oldy but goody article about an FM radio station and huge grounding issues--That were fixed:

https://www.copper.org/applications/electrical/pq/casestudy/nebraska.html

My understanding... Trying to bleed off energy to prevent a strike--more or less trying to prevent a flood by opening a 3/4" spigot at the base of Hover Dam.

From an old post of mine:
http://forum.solar-electric.com/discussion/comment/74865#Comment_74865
================================================================
What the heck--try a SWAG and see what it looks like. From some basic Lightning facts:

Say a stroke lasts about 1 milliseconds at 20,000 Amps (very large strike--but there can be larger)... And you want to bleed off the current at less than 5mAmps (below that current is human safe--guessing from "stories" in earlier post--that person did not die/cook/etc... So the current from dissipation is relatively low.

20,000 Amps * 0.001 seconds = 20 Amp*Hours

To dissipate the energy down below 100 amps (small strike)--basically:

20 AH / 0.005 amps = 4,000 hours

So, with low amperage dissipation (air is a "poor conductor" until it is ionized--then we get the 100-20,000 amp current)... We would need the dissipation to "discharge" the strike within 5 minutes. The time frame would seem to be "too long" for that to have any meaningful advantage.

Take 20,000 amps in 1 msecond and dissipate over 5 min:

20,000 amps * 0.001 sec / 300 seconds = 0.067 amps of current

That sounds plausible... a 1/10 amp of current over five minutes to dissipate the strike energy... However, how do I get 1/10 (or even ) 1/100 of an amp) without ionization of the air. And if I ionize the air, I improve the conditions for attracting a strike.

If you look at the total energy of a single large storm (from link at top):

Williams says a moderate thunderstorm generates several hundred megawatts of electrical power.
One storm can discharge enough energy to supply the entire U.S. with electricity for 20 minutes

I don't see how any "one" small location can dissipate enough energy to protect from a good sized electrical storm.

At best, you attract the strike to something that won't suffer economic (tree, lamp post) or to something designed to absorb the bolt (lightning rod + wiring to earth).
================================================================
I could not find a quick number on the current discharge from s static charge--But it is very small (on the order of micro amperes or less?).

-Bill

Dave Angelini · May 2017

mcgivor said:

Dave Angelini said:

The different models world wide reflect that split phase is mostly in the Americas.

Understood, but the single phase SW 2524-230 is available in North America but has different specifications, same part number according to the pdf files.

I am missing your point! I can buy any of the models they make, no matter where I live for my clients. There are applications in split phase land where the equipment is all 230VAC loads.

Lumisol · May 2017

I would think about replacing all the components that were in service and connected at the time of the strike.

Lumisol · May 2017

A lightning strike will be a LOT more than 20,000 amps. lol 200, 000 has been recorded, and you are also looking at a billion volts.

Lumisol · May 2017

Ground wires will not contain the current in a strike, but it can direct it, the actual current will still be conducted in air.

mcgivor · May 2017

Lumisol said:

A lightning strike will be a LOT more than 20,000 amps. lol 200, 000 has been recorded, and you are also looking at a billion volts.

Lumisol said:

I would think about replacing all the components that were in service and connected at the time of the strike.

How the hell do they measure that current? and voltage for that matter??
I will definitely replace all the components, when they stop working.

BB. · May 2017

Set up a laboratory in Florida (lightning capitol of the US). Shoot rockets into storms trailing wires, etc.:

http://www.lightning.ece.ufl.edu/
http://www.lightning.ece.ufl.edu/ECEnewsletterFinal.pdf (about the laboratory--2013 newletter)

The numbers for lightning strikes are quite variable regarding current and voltage. Ionization of the air by high voltage makes the air more conductive and drops the "arc over" voltage, etc.

-Bill

Lumisol · May 2017

mcgivor said:

Lumisol said:

A lightning strike will be a LOT more than 20,000 amps. lol 200, 000 has been recorded, and you are also looking at a billion volts.

Lumisol said:

I would think about replacing all the components that were in service and connected at the time of the strike.

How the hell do they measure that current? and voltage for that matter??
I will definitely replace all the components, when they stop working.

Rockets can be used as well as assessing and measuring fulgurites.

mike95490 · May 2017

re http://www.lightningprotection.com/lightning-protection/
I've got to say that that scheme with the "fuzzy / spikey" "Dissipation Systems" ESE is not a recognized or certified system, by either UL or IEEE
The standard "Franklin Rod", properly installed in the correct locations, has proven reliable..
http://www.thesundaily.my/node/156236

mcgivor · May 2017

mike95490 said:

re http://www.lightningprotection.com/lightning-protection/
I've got to say that that scheme with the "fuzzy / spikey" "Dissipation Systems" ESE is not a recognized or certified system, by either UL or IEEE
The standard "Franklin Rod", properly installed in the correct locations, has proven reliable..
http://www.thesundaily.my/node/156236

Thanks for the links, been reading extensively on the subject, came to the same conclusion about the ESE method,
so using the standard copper 4 point ones available at any hardware supplier, mounted on a galvanized steel tower, bonded to a ground rod. The house has a steel roof connected via 500 screws to steel rafters, welded to the rebar, which has 16 concrete posts and footings packed with rebar. Going to add some more rods around the house and array about 20 meters high. It was a wake up call to do something, over 10 years with nothing and 4 days after installing the rod, it gets hit, to the best of my knowledge. Kind of reluctant to install on the house itself, although inherently safe, thought it better to discharge a litte further away.

mvas · May 2017

mcgivor said:

Well it had to happen sooner or later, had a direct strike about 5 minutes ago...

Where & how was your (damaged) Inverter grounded?
Do you have a single or multiple grounding connection(s) for all of your equipment:
a) Inverter
b) Controller
c) Battery Bank
d) Panels
c) Array
Are the outside wires inside metal conduit or plastic conduit?

mcgivor · May 2017

mvas said:

mcgivor said:

Well it had to happen sooner or later, had a direct strike about 5 minutes ago...
Where & how was your (damaged) Inverter grounded?
Do you have a single or multiple grounding connection(s) for all of your equipment:
a) Inverter
b) Controller
c) Battery Bank
d) Panels
c) Array
Are the outside wires inside metal conduit or plastic conduit?

Common ground for a,b,c, no grounding for panels or array, outside wires in PVC underground ~10 meters from rod, one overhead feed to an outbuilding too close, 3 meters from the rod, since removed,, EMP probably did the inverter in as it's the only casually so far..could be more down the road.

mvas · May 2017

1% of strokes exceed 200,000 Amps
10% of strokes exceed 80,000 Amps
50% of strokes exceed 28,000 Amps
90% of strokes exceed 8,000 Amps
99% of strokes exceed 3,000 Amps

Maximum Amps are achieved within 10 microseconds.
Each pulse, usually more than one, lasts around 350 microseconds.
It is DC, with a very significant High Frequency component.

Now that that you have attracted the lightning to your tower, "highest point with lowest resistance",
what does it take to properly divert 28,000+ amps into the earth?

mcgivor · May 2017

Attract lightning? From all research done, by no means am I disputing your claim, in fact it seems that is exactly what happened, experts in the field claim, lightning is either going to happen or not, offering a path to ground therefore is not attracting, it is diverting what would have hit something else close by. It's a controversial subject and there is no clear consensus and by no means do I claim to have the answers. Thanks for the input.
Edit, did you mean diverting the strike to the tower?

jonr · May 2017

> no grounding for panels or array,

So there is no way for current to flow from the panel frames to earth ground? Isn't this unusual?

mcgivor · May 2017

The panels are bolted to a welded steel frame, supported by six legs encased in concrete footings 50 cm. In diameter burried 50 cm., the grounding would be whatever the conact and resistance values are, no intentional grounding or bonding. The rational being, why make the array a target by intentionally offering a good ground path to something you don't want to take a strike in the first place and was the prime reason for adding a mast with a good ground, to offer a better target. There is much controversy regarding what is the best method including some who claim the NEC is wrong in their recommendations when it comes to PV systems, the internet is flooded with information, by "experts " in the field whose recommendations are so polarized it's impossible to come to a conclusion, not to mention non experts who base their ideas on what the believe to be correct. What I have observed is lightning hitting the two cellular and one satalite communication towers 800 meters away on numerous occasions, their lightning protection must work as the network stays up and running, each one has a single air terminal the same I use only larger and more than likely a more elaborate grounding system, to protect there more valuable equipment. Wish I knew all the answers, but based on the general consensus it would appear there is no, one answer, every strike is different and unpredictable.

Lumisol · May 2017

The panels should be grounded at the frames to discharge the charges that build up from the charged air and not to protect against a strike.

Direct lightning strike, what will the damages be?

Comments

Categories