Years ago I recall watching a lightning study documentary... and IIRC, they suggested that arrays of smaller sharp devices like aircraft static discharge wicks might be superior.
Not that they dealt with strikes well, but were efficient at quickly bleeding off local atmospheric electric charges, thus preventing a local strike in the first place.
Erik
You're 100% right. Thanks for posting the NFPA780 reference, which can be googled for some good information. I was thinking of the grounding conductor at the service entrance panel, and just guessed that it would be sufficient for a lighning strike.... I wuz wro..wrr.. wrooo...... mistaken......
I think the idea that lightning rods bleed off atmospheric charges has been disproven. Conceptually it doesn't make sense to me. Consider two huge metal plates, say a city block size in area, seperated by 2000 ft of air.. You put some metal points about a foot higher on two spots on the bottom plate. Now you start applying a charge to the two plates. Are those two points going to do anything to lessen the charge in any material way? Air is a pretty good insulator, until it becomes ionized. And once that occurs, we know what's going to happen.....
Something else to consider: There's cloud-to-ground discharges and ground-to-cloud discharges. On a ground-to-cloud would the discharge be from the roof rods? I've felt a ground vibration build up to where it rattled dishes before the discharge happened. Thats a definite ground to cloud discharge!
Another point though is discussion of the charge conductor - solid or cable. Electrical charges travel on the surface of the conductor, not through the center. A cable has multiple strands, each strand with it's own surface. So a stranded cable provides more surface area & more electrical discharge than a solid copper wire.
Another point is the roof lightning rods. The protection provided is a 45 degree downward cone from each rod. So height of the rods above the roof is important - short ones give very little surface area protection whereas multiple higher ones can protect most of the roof.
On Friday 25 January 2013 02:50 Red wrote in alt.home.repair:
Isn't that applicable for high frequency AC (the skin effect)? For DC, bulk matters.
Now, I agree that the duration of a lightning strike is short so the skin effect may be relevant - but just for the sake of being precise...
However, bulk does provide thermal mass to prevent overheating for a given I^2.t
That is true for high frequencies, not DC. The depth the current penetrates below the surface of a conductor decreases as the frequency increases. At say 60hz, the penetration is deep enough that it's not a factor in the conductor gauges used for typical wiring and the current can be assumed to be carried throughout the entire conductor.
I don't know what the frequency profile of a lightning strike looks like, but would suspect it has a broad range of frequency components to it, so there likely is some skin effect involved, but I doubt anywhere near all the current travels only close to the surface.
=A0A cable has multiple strands, each strand with
de quoted text -
OTOH, like charges repell. At these current densities it does matter.
Not true. Purge that thought from your memory. If the stranded cable is made of uninsulated conductors, the electromagnetic forces will successfully move all the conduction to the outside of the cable. Thus, the conductor areas on the inside will still carry almost no current, similar to what happens in the solid conductor. If the 'wires' all all insulated, wrapped in a simple twist pattern, and connected in parallel at their ends; the impedance is NOT as low as your thought process would expect. The impedance IS a bit lower with insulated conductors placed randomly into a bundle, called Litz wire. However, if the goal is to connect two points together with the lowest impedance possible; make certain the connection is wider than long. Then you have a chance at low impedance between the two points.
If you want to explore for yourself and not suffer through all the equations, download a free copy of a finite element analysis program called femm 4.2 Any trouble trying to use the program, the user group is extremely knowledgeable, fast, and helpful.
It's not my thought process, it's the knowledge I was taught when I installed lightning protection systems. And the cable wasn't just a standard twisted strand, it was more of a woven type which might have had some different characteristics. I would assume the thought process came from engineers who designed the materials used in lightning protection.
Hide quoted text -
Then kindly provide us a formula that shows that skin effect applies to DC moving through a conductor at high currents. Or a reference that shows that this phenomena exists. FYI, high current does not imply a build-up of charge within a conductor. It's just the faster movement of electrons that are already there.
I agree with Robert Macy.
I believe hams like flat braid (like a flat stranded wire) for conductors that may carry high lighting currents.
Far as I know, the current design technique is to roll a sphere with a radius of 30m over the building and surroundings. The sphere stays on top of the rods. If the sphere touches the building, lightning can strike there.
You likely need fewer rods with the sphere design than the cone design. But if you transport the Empire State building to the middle of Nebraska, with no surrounding buildings, lightning can hit the side of the building. Sides of buildings may also need lighting protection.
Correct, Red. Robert has obviously not installed them and likely has neve seen an installation up close.
ALL of the farm lightning rod systems I've ever seen have used the looase braid copper cable.(Class 1)
See:
I with you on the above. It's exactly what I said in another post.
quoted text -
A lightning strike doesn't add electrons to the wire? Go figgr.
r.- Hide quoted text -
Not in any amounts that create the effect you're claiming. Electrons moving through a wire are like water flowing through a pipe. If you double the flow rate of the water, do you increase the amount of water in the pipe? No you still have the same amount of water in the pipe at any point in time, it's just moving through it faster Same thing with current through a wire. You're pushing electrons in one end and some other electrons come out the other end of the wire.
If you still believe you're correct, ie that skin effect exists in a conductor for DC as current increases, it should be easy to find the formula that expresses it. The only skin effect I know of is always discussed as only applying to AC because you need a changing field to create it. The changing field doesn't exist with DC. And the skin depth effect varies as a function of the frequency and can be expressed in a formula. Put zero in for frequency in the formula and you get infinite skin depth.
Disclaimer: I'm not saying skin effect isn't of importance in a lightning strike. I said that I believe it is, because lightning has a fast rise time, so clearly it has high freq components to it. I'm only saying DC current in a conductor is evenly distributed.
went to ground
ubt they would
ter.- Hide quoted text -
e quoted text -
You are absolutely correct. DC Current demonstrates NO skin effect. The conductance carriers are distributed uniformly across the profile of the conductor. of course, uniform in the absence of an external field. The earth's field at 50uT is too small to notice much.
However, it is worth noting that DC current is an approximation. Cannot truly exist because it was not always flowing. Actually, every current is AC, just a matter of the degree of how slow you want to consider the frequency to be.
Thank you for those two URL's
ALL the cables from these two sources appear to have bare, uninsulated strands.
Near DC, these cables are probably only 40% higher impedance than solid cable of the same diameter. Being a bundle of uninsulated strands, they will still have high impedance caused by skin effect. At initial inrush, during the fast transient of a strike, the conducting carriers will still move to the outside of the stranded bundle and be restricted to a small region around the outside edges. Picture a 'tube' of current going along the cable.
There will NOT be an overall lowering of impedance at higher frequency because the cable has 17 strands in it.
The main advantage of stranded cable? Easily made, easily bent. As in good luck working with a cable that is that large in diameter and is therefore a solid rod.
The large diameter has an advantage of strength. Its strength will help during a discharge to prevent the cabling from launching itself off the building as it tries to straighten out.
An example of wire straightening out during high current flow of a discharge was very noticeable in a piece of test equipment I once built for testing telecom equipment. Telecom equipment usually takes
-48Vdc. Four 12 V batteries. To test telecom equipment against surges in their power supply, simulating the event of someone dropping a wrench across the bare distribution cables, I put a 50 foot length of
0000 cable loosely coiled on the floor between the batteries and the telecom equipment. Then in parallel to the equipment I placed a starter solenoid in series with a 3AG 1A FB fuse [small glass tube with a tiny wire in it] Activating the solenoid put all 48 volts across the little fuse [no pun intended] and it quickly blew out, going off much like a flash bulb. The momentary short would draw up to 500 Amps that after the fuse had blown would then try to go through the telecom equipment attached next to it. If the current couldn't go through, the voltage would quickly climb to over 300 to 500 volts and literally blast its way in trying to supply 500A. The extremely heavy cable would jump up off the floor as it tried to straighten out. Now that was only 500A. Imagine the 'straightening' power of 20,000 Amps!
I understand. But remember that *if* you take 17 strands of a small diameter wire, combine them into a bundle, near DC the overall cable will act like 17 wires in parallel and have lowered impedance, but at high frequency the new bundle of 17 wires will not act like 17 wires in parallel, but rather act much like a large diameter, somewhat solid cable.
The only way to connect two points with a really low impedance is to make the connection WIDER than LONG. Then you have a shot at lowered impedance. And, that even means making the connection wider than you think necessary, like 2X wider. Then your connection becomes very low impedance acting much like a ground plane, very low impedance. Else, if connection is longer than wide, you have an inductor. And as you know, for an inductor it will not pass any appreciable current for a short bit of time, no matter how large a voltage you slap across it.
You can kind of cheat a bit and approximate a wide connection by using several connectors. That is why a well bonded wire grid structure protects buildings/structures so well. The connections, albeit not solid over the whole width, help lower the impedance of the connection by making it wider than long.
I think the thing to remember, DC has no skin effect with a steady state or slow rate changes. A spike has frequency content. Instantaneous changes have frequency. Never thought about this before except in the case of electronic filtering.
Greg
The only way to have a bunch of conductors bundled, is to insulate each to prevent interaction. Except it must withstand, how many volts !!!! Even insulated, stranding close to other strands will have interaction nullifying what your trying to do.
The cables I have seen on buildings are stranded aluminum at least 3/4 inch in diameter.
Greg
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.