Again, most of this is wrong.
There are single point grounds, and there are multi-point grounds.
There are benefits and drawbacks to both systems, and I'm not getting
into them here until we correct a few misconceptions first.
1 - WHEN LIGHTING STRIKES, THERE IS A SIGNIFICANT FAULT CURRENT
FLOWING IN THE EARTH. ANY CONDUCTOR LOCATED IN THAT EARTH WILL
DEVELOP A CHARGE. THE MAGNITUDE OF THE POTENTIAL IS DEPENDANT ON MANY
THINGS - NOT ALL OF THEM KNOWABLE IN ADVANCE.
Period. End of story.
2 - JUST LIKE HOUSE CURRENT, LIGHTNING SUFFERS FROM SIGNIFICANT SKIN
EFFECT LOSSES.
Therefore, you are always better off to use thin copper sheets (or
copper braiding) for grounding, rather than big honkin' ground rods.
Only the surface-area of the rod conducts, and it exhibits a great
deal of inductance at VHF (which is what lightning mostly is.) So,
worthless. (or at least, never as good as you might expect / hope. :
( ...Unless you can't properly solder copper flashing, then you're
usually better off with a ground rod and a clamp.
3 - THE CURRENT IS TREMENDOUS. It will conduct across air, just as
easy as copper.
It depends on where the leader, or M-return strokes, have been
established before the lightning actually struck. (You can use a Boys
Camera to test this theory, if you are so inclined...) So, if the pre-
strike ionized path did not happen to include a nice conductive path
to your ground system(s) (i.e., rods, halo, X-it Rods, whatever...)
you are S-O-L. A typical strike averages approx 20,000 to 40,000
Amps. Individual strikes can be MUCH larger.
4 - EVERYTHING LOOKS LIKE "GROUND" WHEN YOU ARE CHARGED TO THAT
POTENTIAL.
5 - GROUND RODS AT HOUSES ARE THERE PRIMARILY TO CLEAR FAULTS (i.e.,
to ensure that a fuse or circuit breaker trips quickly!) THEY ARE NOT
THERE FOR LIGHTING PROTECTION PER SE, BUT BECAUSE THEY REPRESENT A LOW
CURRENT PATH TO "GROUND", THEY CAN BE QUITE EFFECTIVE.
6 - The NEC says drive 1 ground rod. If < 25 ohms cannot be
established with 1 rod, drive ONE more rod at least (x)-feet away. I
don't recall how far?, but it's like 6 to 10 feet. Just enough so
that the "earth" it "sees" is different "earth" from what the other
rod "sees". Now, no matter what the dual-rod connection is reading,
EVEN IF IT IS 200-OHMS!, it meets NEC Code!!!! - And that means
you're either at the beach with crap sandy soil, or on a volcano with
lava rock, or maybe trying to gound to solid rock. Note: You really
should drive some more ground rods, or switch to a different method of
grounding, even though the NEC does not require you to do so, and even
if the building inspector says you're good....
In either of these situations, you could try Bentonite, or salted
rods, or both.
You could trench out long radials of copper flashing and cover them
with salted soils, etc...
50-50 Flake Calcium Carbonate and Rock Salt works wonders by the
way,and much easier to get than Copper Sulfate. (But neither of them
are particularly good for the grass for all you do-it-yourself
homeowners out there!!) Plus its gets quite hot as it is an
exothermic reaction. Wear PPE.
As for Halo ground systems, these are usually (effectively) multi-
point grounds, and the idea is to reduce the path length to ground,
and try to maintain the same potential when the ground does elevate.
The inductance effectively limits a discharge path, so you get the
idea why some folks use them. I find the problem with them is that it
is (expensive) and next to impossible to keep this grounding separate
from the Neutral coming in to the structure. The famous Motorola
"R-56 Standard" was notorious for violating this, but I guess they
were more concerned about radio transmitter antenna / lightning issues
than compliance with Code.
As we all know (because we are all NEC experts here), you MUST not
bond the neutral to ground past the service entrance (main breaker in
most situations) becuase doing so creates a lethal ground fault
condition IF the Neutral to the building ever gets disconnected or
damaged. All the conduits - which should be at ground potential -
would then be carrying current.
And if lighting struck in that situation..... Good luck!!!
I would not endorse a design that sacraficed personnel protection for
equipment protection.
When this is done, I believe other engineering controls are prudent to
protect against bodily harm, such as a prohibition against working
during electrical storms, etc....
Oh, I forgot one. Separately derived grounds (such as one building
feeding another) can have
"common earthing" much greater than 10 feet. "common earthing" is a
made up term anyway, so one can say whatever they want about that....
It is meaningless.
Another one. The guy who got injured --.
The equipment DOES NOT MATTER!. He could have been standing with his
feet apart. Lighting strikes nearby, and a current starts to flow in
the Earth. Now, Earth, like everything else, has resistance. By ohms
law, there can be a SIGNIFICANT potential between the person's two
feet. The right foot could be at 1000 Volts potential, and the left
at 5,000 Volts. The difference is enough to cause electrocution
irrespective of the location (or even existence) of the machinery!!
It is also possible a current path to/through the equipment occured;
we just don't know.
You can sum up the "valid" portions of Tom's reply thusly:
Each level of protection tends to peel off some of the energy of the
stroke. With enough suitable protection, the remaining energy can be
brought down to a level where (hopefully), gizmos and people can live
peacefully. Unfortunately, as we know, this is not always the case,
even with well engineered ground systems - single point or
otherwise.
(Not talking about Faraday cages here, which is a special case.)
Also not talking about ball lightning, which again is a special case.
-mpm
