Faraday Cage For Lightning Protection

[Jeff Rosenbury]

Exactly what do you mean by "this?"? I have designed, and built professional antenna systems that model much like yacht rigging and which never fail while surviving lightning strikes often.

How is a differential potential possible? The tip of the mast and the sea surface can only have one potential difference.

The strike does not travel in the conductor but on or above the surface, It travels as a wavefront over the conductive structure. That surface breakdown current is guided by the conductor potential but travels outside or away from insulation on the conductor. Breakdown paths through gasses are curved = arc. Conducted currents avoid sharp bends and dielectric insulation.

Now that you have resorted to credibility assassination by FUD, (Fear, Uncertainty and Doubt), you have lost credibility.

By "this" I meant trying to use the lightning protection system of a sailboat as a Faraday cage to protect the electronics inside from the EMP effects of the lightning strike.

It's my understanding the sea surface, being a conductor, will act as a mirror of the charge distribution in the sky above it. The (let's say) positive charge in the clouds will draw a negative charge under them. So unless the boat is centered under the moment (center) of the clouds charge distribution, the potential on one side of the boat will exceed that of the other. Perhaps I'm wrong about this.

My understanding is that such a lightning strike is not one thing. There will be a series of EM waves of different frequencies traveling at (presumably) slightly different speeds (due to frequency dispersion). There will also be moving charges moving much more slowly, but generally still faster than the speed of sound (I could be wrong about this, charge carriers typically move slower, but with millions of volts?). The EM waves moving as a surface wave (if they do, I've not done the math) will themselves induce moving charge carriers in the conductors. How much of this goes on will depend on the electrical characteristics of the conductor set (i.e. a waveguide). These characteristics include the dielectric which is wildly variable if a bare wire gets wet (and still somewhat variable if an insulated wire gets wet). I have no idea what the dielectric constant of ionized water is, or if it matters. (One of those thousand unknowns, BTW.) I do know ionized gas streams are subject to pseudo-random movement as their magnetic fields interact with themselves, the explosive nature of hot gasses, and the tendency of those gasses to move with the wind.

All of these concepts cause me to doubt lightning current can be evenly split in 4 wires, cancelling the EMP. While I admit the theory is intriguing, I could not recommend such a solution in a situation where lives might be at stake without extensive testing.

 

[Baluncore]

There will be a series of EM waves of different frequencies traveling at (presumably) slightly different speeds (due to frequency dispersion).

A lightning strike is a current impulse with a broad spectrum that covers from about 1Hz to 100MHz. The phase of the spectral components are synchronous at the instant of the impulse. Frequency dispersion is quite a different issue, it becomes important only when EM waves are traveling for miles through plasma or for light years in space.

There will also be moving charges moving much more slowly, but generally still faster than the speed of sound (I could be wrong about this, charge carriers typically move slower, but with millions of volts?). The EM waves moving as a surface wave (if they do, I've not done the math) will themselves induce moving charge carriers in the conductors.

EM transients travel on or above the surface of conductors at close to the speed of light. The electrons and ions do move very slowly in the conductive medium. The movement of charge carriers is induced by and guides the near EM waves. Energy actually propagates directionally in the EM field between and guided by the conductors. The “current carriers” represent an energy loss for the propagating EM fields. That energy loss appears as heat and causes steam explosions in resistive living trees, but if you watch lightning strikes hitting seawater you will not see explosions, because sea water is more conductive than air so less energy is lost in water than air.
The millions of volts are in series through the air prior to a strike. Many kV of potential difference appear at the tip of a lightning strike as it searches for a path through the air. The effect is similar to a crack propagating through glass, the tension is focussed and released at the tip of the crack. The voltage drop along a lightning strike is very much less than at the tip, likewise after glass has cracked, the stored energy has been released.

These characteristics include the dielectric which is wildly variable if a bare wire gets wet (and still somewhat variable if an insulated wire gets wet). I have no idea what the dielectric constant of ionized water is, or if it matters. (One of those thousand unknowns, BTW.) I do know ionized gas streams are subject to pseudo-random movement as their magnetic fields interact with themselves, the explosive nature of hot gasses, and the tendency of those gasses to move with the wind.

To you this is still a black art, full of fears and unknowns. To me it is a highly predictable science that can be modeled and if necessary, calculated. Many years from now you may see this field differently.

I could not recommend such a solution in a situation where lives might be at stake without extensive testing.

It is not for you to judge my professional competence in fields in which you have little experience or understanding. When you suspect you are out of your depth, you should not criticise, but watch, ask questions, and discover. PF has many experts on many subjects.
Understand and recognise the Dunning–Kruger effect. https://en.wikipedia.org/wiki/Dunning–Kruger_effect

 

[Baluncore]

After some thought, I would use external conductive rigging and/or place and connect stainless steel conductors on the outside corner/edges of the vessel/cabin above the waterline.
I would remove the internal mast to keel electrical connection as it poses real liabilities. If it did conduct it would generate a strong magnetic field. If insulated it could start a fire or ignite fuel vapour in the hull. Lightning causes flash burns where current flows over a resistive surface. It will happily jump 18” outside insulated wire or from chain plate bolts to sea water rather than flowing on the wet hull surface.
Take a look at what determines the initial arc path in a gas. https://en.wikipedia.org/wiki/Paschen's_law

 

[jim hardy]

I would remove the internal mast to keel electrical connection as it poses real liabilities.

In our plant the lightning protection was large bare cables about 1" dialmeter from lightning rods atop the structures down to dedicated lightning earthing electrodes sunk forty feet in the ground. Plant ground mat was laid half that deep, so as to make a "magic carpet" ground plane that "floats" atop whatever disturbance lightning currents impose on local Earth potential.

Basic premise was encourage lightning current to flow around the plant structures not through them.
It worked very well, the plant was quite tolerant of lightning strikes.

So your external conductive rigging agrees with the intuition I've picked up from talks with industrial design engineers.

That mast to keel connection is probably embedded in the fiberglass hull and not easily removed. I'd make sure it has huge cross section so it won't explode when asked to carry a couple tens of kiloamps , and provide a "preferred path" around it outside the hull. I would find deployable cables with trolling weights reassuring, affix them to standing rigging in electrifying weather. That should wrap the entire vessel in a comparatively equipotential surface.

two cents worth (if that much) from a non expert...

 

[anorlunda]

Wow, this dicussion is going in many directions.

Baluncore said, "I would remove the internal mast to keel electrical connection as it poses real liabilities" I like your arguments, I think they are well founded, but they violate the ABYC standards that Neely mentioned. Every 5 years my vessel must be inspected by an ABYC surveyor for compliance, or else I become uninsurable. I said in #29 that I think the ABYC standard is idiotic, but I'm supposed to comply with it. ABYC is the floating equivalent to NEC, the National Electrical Code.

The ABYC standard also implies that I need 50 or so (#8 AWG minimum) cables, and Jim Hardy just recommended an extra thick ground cable from the mast able to carry "a couple tens of kiloamps". My ground is not the keel, it is a 1x2 foot bronze plate below the water line, that has a single 1/2 inch bronze stud that extends 1 inch inside. All these cables are supposed to terminate on that stud. Let's get real.

I also have to reject any suggestions about dangling loose objects overboard when underway. That would be extremely unseaworthy.

 

[jim hardy]

as i said I'm neither a lightning expert nor a good mariner. and I should have said 'plate' not keel...

I'm under the impression it takes ~10*9 amps/m^2 to explode a wire in a millisecond
for your half inch bronze bolt that would be 190 kiloamps which is a HUGE lightning bolt... i doubt your bolt is the weak link.
but i'd run a big wire over to the mast...

sorry for butting in - just it's interesting. I like to learn something every day.

Thanks for letting me play in your sandbox, guys...

old jim

 

[anorlunda]

Your butting is always welcome Jim. You have goo instincts.

I wasn't worried about the stud melting. I can't see fitting 50 of these connectors on a one inch long stud. Even 2 or 3 is tight. I also can't imagine one of these carrying 20 KA.

 

[Jeff Rosenbury]

The purpose of the "keel plate" is to allow the lightning to ground directly under the mast. It was added when, in many cases, lightning didn't follow the bends in the conductors toward the sides of the boats but instead went straight through the hull. It should be noted that the current "keel plate" designs are usually sufficient for salt water, but not fresh water. Boats struck by lightning in fresh water typically have a hull blow through even with a plate. (source)

This source also suggests upping the down wire from 8 to 4 gauge, since the 8 gauge apparently came from a study on rural buildings in Poland with only a 95% safety factor. It indicates the "keel plate" would need to be unreasonably large in fresh water (bigger than the keel).

Baluncore, what you think I might know many years from now depends a lot on your beliefs in the afterlife. I'm not some wet behind the ears kid. I'm also not recommending ignoring established standards.

 

[nsaspook]

as i said I'm neither a lightning expert nor a good mariner. and I should have said 'plate' not keel...

Thanks for letting me play in your sandbox, guys...

old jim

Don't worry Jim, you won't be Keelhauled.

 

[Jeff Rosenbury]

Your butting is always welcome Jim. You have goo instincts.

I wasn't worried about the stud melting. I can't see fitting 50 of these connectors on a one inch long stud. Even 2 or 3 is tight. I also can't imagine one of these carrying 20 KA.

I always enjoy Jim's posts. It is kind of him to make the effort.

I agree your plate's lug isn't big enough. And I suspect the connector is seriously undersized, particularly if you go with the IEEE recommended 4 gauge wire. Remember these plates aren't required by shipbuilders and are only put into make insurance companies happy. Given that about 3% of boats are struck each year in South Florida, a clever man would figure something out.

 

[anorlunda]

I too like Jim's posts.

The 4AWG connectors are slightly better.

But this whole discussion stretches my imagination by several orders of magnitude. Especially when the ABYC standard wants me to give all large metal objects above and below decks the same treatment as the mast. (I think of a sauce pan in the cupboard or a spare anchor stored under the floor as absurd examples).

I'm an old power plant guy. To safely handle 20 KA, I think of things like isolated phase bus ducts as shown below.

Add to that the opinion of a knowledgeable engineer like Baluncore that I should eliminate the ground plate entirely, and you see how far ranging the advice is.

I believe these questions can only be answered statistically. My insurance company, BOATUS, publishes a safety magazine. I asked them to publish an article based on their claims database about what works and what doesn't for lightning protection.

But I must say, this has been one the most enjoyable PF discussions I've seen. Divergent opinions, but all rational. Thanks all.

 

[jim hardy]

Remember these plates aren't required by shipbuilders and are only put into make insurance companies happy.

Look at a copy of National Electrical Code. Publisher isn't IEEE or some such august academic outfit, but NFPA - National Fire Protection Association.

I guess experience is the best teacher. It teaches us which fundamentals to apply for given situations.

Thanks again, guys .

old jim

PS that Isophase photo is nostalgic. I measured flux in vicinity or ours, 0.1m² gave 0.3 volts per turn. ~8 milliTeslas ?

 

[Jeff Rosenbury]

I started a thread on the general physics forum about how predictable lightning is.

 

[anorlunda]

My hopes that the BOAUS insurance company could shed more light are dashed. But the reply I got does add some interesting stuff.

On the ABYC, lightning protection is no longer a standard, just a “technical bulletin” which means it is there to provide information to interested parties. I wasn’t part of the discussion, but I would guess that objections like this had a lot to do with that decision.

Unfortunately, our claims files can only be of limited assistance on the topic of lightning protection. First, there’s only a very few boats (relatively speaking) that get hit, so the data is thin to start with. Historically, our files did not include information on the grounding/bonding system in the boat even after a lightning strike, and our application doesn’t ask about it, so we definitely don’t know anything about lightning protection when it comes to boats that did NOT get struck. We’re working on all of this, so we might have really good data on how lightning protection affects claims in a decade or so…​