The farm house we are looking to buy in Westport is 100 years old and has ridge top lighning rods. The ground cable is connected to the copper plumbing, upstream from the meter.
Owners agent says this is fine. Inspector says it’s a no no and should be connected to a grounding rod going into the ground. Sure would hate to be taking a shower if lightnng hit the water pipes?
Let’s not confuse the issue with facts!
Smart construction decisions and material choices can significantly improve occupant safety and survival in the event of a fire in the home.
"I have learned so much thanks to the searchable articles on the FHB website. I can confidently say that I expect to be a life-long subscriber." - M.K.
Dig into cutting-edge approaches and decades of proven solutions with total access to our experts and tradespeople.
Start Free Trial NowGet instant access to the latest developments in green building, research, and reports from the field.
Start Free Trial Now
Replies
My understanding is that it's a no-no. The lightning rod system should have it's own dedicated ground rods, with as direct a path between the air terminals and the ground rods as possible. Not sure whether it should be connected to the rest of the grounding system or not, but if so it should be with a relatively small and long wire, so wire impedance minimizes lightning pulse transfer to the rest of the system.
Did a metal room tower addition on a house in WV and the local lore said dedicated ground.. .looked at some old barns and homes with lighting rods and they had dedicated ground. Put a metal roof on my two story and use a dedicated ground.
Without a doubt the lightning rods should run to their own ground (they may be connected to each other) and not to anything else in the building. If lightning were to strike the lightning rod, everything connected to it would instantly have on it the millions of volts that the bolt had. That's enough to jump from everything connected to it to objects and people nearby.
I've only had a lightning system put in once, but the guy I hired must have put in 20 ground rods around the house,and connected absolutely everything metal in the house to the grounding system.
He also installed surge protectors in all the electical panels.
Then he came through with a state inspector who apparently only inspects lightnig systems, who OKed it.
You absolutely DO want to tie the lightning rod to its own string of ground rods spaced more than 6' apart from each other AND tie that with a low impedance connection to the water pipe and electrical and telephone and cable TV grounds. A length of #4 copper and some ground clamps between the various pipes and ground rods will do the trick. You want all grounds to rise and fall together in the event of any one of them rising. If you don't the voltage difference between ground on one system will jump to the lower voltage on another ground system through whatever it can such as you and/or your tv. If all the grounds rise and fall together there is no voltage difference. Assuming it is an unbroken length of metal pipe buried in the ground that doesn't get replaced with a piece of PVC at some point in the future, the cold water pipe is actually one the best grounds you can find, but you want to supplement it with a ground rod system.
I understand voltage potentials wanting to jump. Teachng automechanics for years helped me appreciate what 20,000 volts can do, albeit low current.If I understand correctly, the lightning ground cable should be solidily connected to both the pipes and the earth itself, so there can be no sigificant potential between the two. So if the lightning ground cable I saw solderd to 1 1 1/2" pipe also connects elsewhere to a rod or rods driven in the ground I'm likely okay?D I correctly infer that it should not be directly connected to the main panel? Apprecite the feedback so far. Any further clarification would be most helpful as we would like to get this straightened out before we close on ''this old house'.Let's not confuse the issue with facts!
There are some very complex high-frequency issues at work. Lightning behaves a lot like high-frequency (near microwave) electricity, and the usual rules of thumb don't apply. The ground wires are really acting almost like waveguides and as such things like sharp turns need to be avoided.This also means that what "looks" like a low-impedance path at 60Hz may not be to lightning (and vice-versa). And given the high voltage of the lightning burst anyway, you could have hundreds if not thousands of volts potential at the top end of a well-grounded lightning rod.Normally lightning rods are connected with a loosely-twisted stranded conductor which has a metal cross-sectional area of maybe a #4, but "looks" to lightning like about an 00, due to it's surface area. Wires of this nature should be run to the lightning rod grounding system. One might then run maybe a #6 between that system and electrical system ground, preferably connecting at the electrical system ground rod or Ufer connection. This smaller wire, if more than a dozen or so feet long, would effectively attenuate the lightning voltage by a factor of ten or so, and hence not cause the electrical system ground to spike too badly.But overall, lightning protection is a specialty all unto itself, and much of the body of knowledge is from experience vs science, since science still doesn't fully understand the phenomenon.
If Tyranny and Oppression come to this land, it will be in the guise of fighting a foreign enemy. --James Madison
>>"There are some very complex high-frequency issues at work.
I agree with you, but to put a finer point on it, it's not the frequency but the extremely short risetime and falltime of short duration DC pulse. It usually behaves like a high frequency wave due to the rise fall times, and usually short pulse duration, but it's still DC.
It's important to keep this in mind, IMHO, because if the DC pulse is of longer duration ("hot lightning") it also behaves as high voltage, high current DC during that interval. Pulses have been recorded as long as 500 ms. A stroke of that duration will behave like DC given that amount of time (and admittedly likely melt all the wires regardless of size). Another reason this is important is because if your protection "cage" can't handle high DC currents, it will fail if hit by a long duration stroke.
"Let's get crack-a-lackin" --- Adam Carolla
Where's that hammer? I'm obviously over my head on this one. I think we have a decent system, but the wife may need reassurance. Thanks one and all for all the great info.Let's not confuse the issue with facts!
>>"Where's that hammer?
I'm a carpenter too. Just like to stick my 2 cents in and try to keep those ancient memories from stuff I picked up long ago, at least two careers ago from completely fading. ;-)
"Let's get crack-a-lackin" --- Adam Carolla
Yeah, like I said it's a bit of an occult science. The high-voltage peak of a typical strike is fairly short, with lower voltage peaks before and after the main strike. There are of course atypical strikes that are much longer, but protecting a home from those is probably not worth the money. From the standpoint of the electrical system you hope to attenuate the high-voltage peaks substantially before they reach the electrical system, and you also hope to maintain them as more-or-less "common mode" across all the AC lines vs being in only one line. You're aided in this by the generally high-frequency characteristic of the voltage pulse, and the inductance/capacitance of the home wiring at that effective frequency.From a practical matter, though, the electrical grounding system will be involved in any reasonably large strike, since virtually everything metal in the home is connected to it. For instance, steel siding is supposed to be grounded, and a spark can easily jump from the lightning rod cable to the siding. You're also likely to have a substantial jolt come in via any rooftop penetrations such as a CI drain vent or metal furnace flue (both of which are generally bonded to the electrical ground, if only indirectly). The metal drip edge along the gable end of the roof will energize the gutters which will in turn energize the downspouts.Even if there's no direct connection, inductive coupling between the lightning rod ground wire and wiring or other metal objects inside can be substantial.
If Tyranny and Oppression come to this land, it will be in the guise of fighting a foreign enemy. --James Madison
Forget the high voltage aspect, it is high current that is the main design factor.
Typical lightning is about 20 kA, with largest recorded a whopping 3 million amps.
Lightning rist time is about 6 microseconds, and duration typically only 50 to 100 microseconds (about 10 kHz equiv), but can be as long as 0.07 seconds for the biggies, which makes the biggies even slower than 60 Hz. Lightning is nowhere near microwave frequencies GigaHz, as in 1e9Hz.
Like Dan said it is a specialty. Tie every thing together. Solidly, bolted - soldering is no good, it will melt and magnetic forces blow the solder away and the wires will seperate. Have seen that happen. One of the best grounds is if someone had the foresignt to weld the rebar together in your footing and bring out a stub.
Some tidbits:
A 300 kA stroke will break 4/0 wire from high magnetic forces way before it fuses it. Once did a 4 million amp test and that broke 250 MCM cable.
20 kA on a 10 ohms ground rod = 200 kV, will go to any nearby other ground.
100 kA across 10 feet of 4 AWG will have a 250 V potential , so forget about "no sigificant potential between the two"
> 20 kA on a 10 ohms ground rod = 200 kV, will go to any nearby other ground.
You beat me to it. Those electrons don't just magically disappear into the dirt. They can do strange things underground, like fusing sand into glass, or following drain tile to a septic tank and causing a flash steam explosion. In one ground rod and out another is a real possibility.
-- J.S.
I was looking forward to your input on this one. I figured that as often as you mention the voltages you routinely work with and the business you're in, you'd be in on this one. ;-)
I'm curious about the magnetic forces. Is the energy high enough that "normally" non-magnetic materials like copper would be broken (I'm assuming it will because you mention solder being blown out)? If it will, what's going on in there in terms of the physics?
"Let's get crack-a-lackin" --- Adam Carolla
If you take a loop of copper and run a high current through it, the magnetic forces will tend to stretch the loop into a larger circle, IIRC. Two parallel wires with currents in the opposite direction will be driven apart, while two wires with currents in the same direction will be drawn together.There are actually some industrial uses for this effect, for forming metal in certain cases.
If Tyranny and Oppression come to this land, it will be in the guise of fighting a foreign enemy. --James Madison
Run an electric current thru any wire, copper included, and you get a magnetic field. Move a wire and a magnetic field relative to each other, and you get electricity back again.
Arrange things so that the magnetic fields cause physical motion, and you can make electric motors and solenoids. Use mechanical power to move the wire and fields, and you have a generator. Go from electricity to magnetism and back in a neat little package, and you have a transformer.
In the vast majority of practical devices, the effect is increased by using long wires coiled up. Given the current, voltage, and rise/fall times of lightning, one wire is enough to get plenty of magnetic effect.
-- J.S.
Basic physics is that a current in a wire will generate a magnetic field,
Mag field = .5*mu* turns* amps/)distance away) ; mu = 4piE-7
So, 2 wires are like 2 magnets, current the same way attracts, different ways repels. Force is thus proportional to the current square. A solenoid wants to get bigger in diameter and squash itself like a pancake.
First pix is an open air coil that got about 50 kA, 0.2" by 0.4" copper bar wound in a 2.5" solenoid. Flying shrapnel.
2nd pix for ref is lighning occurence reference.
Sorry for not taking the time to reduce the sizes.
Have seen 20 ft ground rods at substations pulled out of the ground by lightning forces and wound up into wierd gyrations.
Interesting sidelines are electromatic and railguns. 'Co-worker' missed the target and sent a cm cube of lexan thru the concrete block wall of the building. Used to have a 400,000 uF, 6 kV capacitor bank to play with.
I would imagine that folks with arc welders have seen the cables jump when the arc is struck. I know I've seen cables jump with applied current under several different circumstances, but can't remember just where at the moment.
If Tyranny and Oppression come to this land, it will be in the guise of fighting a foreign enemy. --James Madison
Now I get it. Took a minute but if I understand you correctly, it's not the copper reacting to the magnetic field; it's one field reacting to another field; is that it?
"Let's get crack-a-lackin" --- Adam Carolla
Yes.
the "arcs and sparks" links are always good for an education.
http://205.243.100.155/frames/longarc.htm
The arc movement on the mpg files is primarily due to 2 forces - the magnetic fields generated by the arc current and bouyant air forces on the hot lower density plasma.
Edited 5/12/2006 10:31 am ET by junkhound
Or even the field reacting to itself.
If Tyranny and Oppression come to this land, it will be in the guise of fighting a foreign enemy. --James Madison
> Have seen 20 ft ground rods at substations pulled out of the ground by lightning forces and wound up into wierd gyrations.
What were they made of, how were they driven?
-- J.S.
I don't know guys, you may be big dealing this, after all the owners AGENT said this was fine.LOL
Ideally the protected structure would be ringed with rods and the the rods would be bonded together so that the protected structure is inside of a loop. Actually rods in the ground aren't strictly necessary if the loop around the structure is used. The net effect would be to minimize the potential difference across the protected structure, which is what you want. In a typical lighting strike the current rises from zero to maximum in around a microsecond so the loop should be all at the same potential for a loop on a residential structure or even a reasonable sized barn. This sounds like what you're describing, but, if you bond the structure's internal grounds to the lighting protection ground you're asking for trouble.
To keep thing simple, consider just the electrical service. Suppose you bond the ground to the lightning protection system at the service panel. When you take a hit the voltage on the ground in the building will jump (say 10 KV). Meanwhile the voltage on the hot leg will stay about where it was due to the relatively large impedance (f>1 MHz) between the source (think the transformer down the block) and the ground in your panel. So for some number of microseconds you've got a 10 KV potential difference between hot and ground. Kiss your home entertainment system goodbye. Factor in other systems (e.g., phone) that may not be at the same potential and there's potential for danger. My old E&M and lightning instructor always gave the example of yacking on a phone (this was before cordless phones, IIRC) and leaning against a fridge. As I recall he had a documented case of a person killed in this manner. I also remember PSAs telling people not to be on the phone during electrical storms. Bottom line, while you can bond the various electrical grounds and the plumbing you can't bond the various electrical 'hots' so you're going to get the potential difference between the ground and the hots.
In big buildings things get a little more complicated and electronic transient protection devices that are designed to eat both common and normal mode transients become *highly* desirable. These work essentially by introducing a low impedance path between the two (or more) devices that are exhibiting a voltage differential that is out-of-spec -- essentially a temporary bonding -- and like any short circuit dissipate a lot of energy.
"Bottom line, while you can bond the various electrical grounds and the plumbing you can't bond the various electrical 'hots' so you're going to get the potential difference between the ground and the hots."
Aren't the secondaries in the transformers floating? If this is the case, the reference is defined by the ground stake at the residence, and the service transformer is just delivering differential voltages. If the voltage of the secondary neutral is at enough of a different voltage from the ground of the transformer, then there will be an arc in the transformer.
OK, I musta seen wrong. LOL!
If I am correct, and I am no expert just seen a few, it is a small matter of six or so foot copper rod driven into the ground near teh foundation. Thirty minute fix.
.