Wireless Versus Ethernet

glen herrmannsfeldt hath wroth:

Yep. A survival book I read at one time suggested that walking in a lightning storm is a bad idea because of the voltage gradient problem. It suggested I either take small shuffling steps (to reduce the gradient) or hop like a rabbit. Running was apparently a bad idea as there is a small interval when both feet are on the ground and widely seperated.

Quiz question: Fair weather atmospheric potential works out to about

200 volts between my head and my feet. If I connect a wire between my head and my feet, why don't I get a spark?
Reply to
Jeff Liebermann
Loading thread data ...

A true "DC" signal (i.e., theoretically pure DC) is constant *for all time*, from -infinity to +infinity. The spectrum (Fourier transform) of such a signal has a component at zero frequency only.

Any other signal (including an "almost pure DC" that is constant for long periods of time, short of infinite), will always have components at non-zero frequencies, due to the transition from zero current to the specified DC when the signal appears, and back to zero when the signal is removed. The amplitude of these non-zero frequency components is determined by the time since the transition (longer time since transition yields lower non-zero frequency amplitudes), the amplitude of the resulting current, and the slew rate (faster transitions yields higher non-zero frequency amplitudes).

Although we commonly speak of "DC" when discussing practical circuits and systems, we never encounter true DC, since the signal is never constant for infinite duration; however, a signal that is constant for long times (relative to the time periods we care about) can be approximated as DC for most engineering purposes. Lightning is NOT such a signal.

When he referred to the concept that lightning is DC, I believe that Jeff L. meant that the current in a lightning strike never changes

*direction*. However, it clearly changes amplitude during the time of interest, on the order of thousands of amperes in tens of milliseconds. Thus, the resulting spectrum will have significant components at non-zero values of frequency, as the various posted data indicate.

The amplitude of the lightning signal will be non-zero for all values of frequency from -infinity to +infinity (as for any impulse-like current function). Thus, while it contains "RF" (i.e., signals in the RF spectrum), it will generally not have narrowband (CW) components at any particular frequency; thus, a narrowband filter (such as a 1/4 wave transmission line shunt) will not be effective to direct or contain lightning.

However, if the signal you are trying to *protect* is narrowband (e.g., a 2.4 GHz signal), a transmission line shunt can appear as a low impedance for all of the lightning components other than the desired 2.4 GHz. Thus, to the extent that the shunt is really a low impedance (good connection to earth, good conductivity, etc.) it *may* help somewhat. Of course, the filtering characteristics of a simple transmission line shunt are not spectacular; IIRC it appears as a single-pole filter, rolling off at only 6 dB/octave. Finally, if the lightning current melts the shunt conductor, all is again lost.

-- Rich Seifert Networks and Communications Consulting 21885 Bear Creek Way (408) 395-5700 Los Gatos, CA 95033 (408) 228-0803 FAX

Send replies to: usenet at richseifert dot com

Reply to
Rich Seifert

Earthing of lightning energy is why some use a J pole type antenna.

The RF nature of lightning is why short wire length, no sharp bends, not inside metallic conduit, etc are all important for that earthing connection. Due to the RF nature of lightning, wire impedance (not just wire resistance) is an essential part of a lightning protection 'system'. Not only is that copper wire 'heavy'. A shorter length and how it routes to earth is also critical for better protection.

Due to lightning's RF characteristics, an earthing wire bundled with other cable wires may induce transients on those adjacent wires. Just another RF effect that must be considered when installing an effective lightning protection 'system'.

Reply to
w_tom

w_tom wrote: (snip)

Not only RF. Look at the 60Hz impedance of a copper wire in a steel conduit. In the case of a single wire running through a steel conduit carrying 60Hz, how much current travels in the wire, and how much in the conduit?

-- glen

Reply to
glen herrmannsfeldt

Thanks Tom. Those J-Poles look pretty nice from a grounding standpoint. (One plan I found was just a J-shaped antenna made from soldered copper water pipe.)

formatting link
As an aside, I did notice that a company is selling 1/4-wave stubs for lightning protection. (They act as a normal N-to-N coupler at the freq of interest and act as a short at anything off frequency.)

formatting link

-wolfgang

Reply to
Wolfgang S. Rupprecht

My idea wasn't to tune anything for the lightning. What I think you can do is to put a 2.4Ghz 1/4-wave "short" in your antenna cable. The

1/4-wave "short" by the magic of out-of-phase addition looks like an open at the frequency of interest (and any harmonic where the stub is N+1/4 wave long). At very low frequencies the stub looks almost like a dead short. The thing that is enticing to me is that the shorting stub can be made out of very beefy parts and can be clamped directly to a copper rod pounded into the ground. That should give you as very good grounding.

-----------------+---------------------- from antenna | to radio -----------------|-+-------------------- | | | | 1/4 wave shorted stub. | | (1/4 wavelength of 2.4Ghz) | | +-+

(Pretend the above is coax. I'm too lazy to draw all those ascii lines and curved sections.)

-wolfgang

Reply to
Wolfgang S. Rupprecht

"Wolfgang S. Rupprecht" hath wroth:

That's done all the time with antennas. Most mountain top VHF/UHF antennas use one of numerous designs that offer a grounded feed or grounded elements. The main reason is not for lightning protection, but too reduce noise pickup from nearby frequencies. The grounded antenna is by it's very nature a resonant bandpass of sorts. There's also a big advantage as a grounded antenna does not pickup much of a static electricity charge.

More common is a half wave "stub", which acts as an open circuit at the resonant frequency, and a short at all other frequencies. Oh, it's the same as what you drew below.

The problem with that scheme is that there's still a small possibility for some of the energy to leak past the stub. The stub might also decide to become a fuse and there goes your protection. The stub would also need to be shielded as it would act as quite a good antenna or unintentional radiator.

More common is a bandpass filter, where there is no physical connection between input and ouput. Something like this:

formatting link
two wires are a 1 section interdigital filter at 800-900Mhz. It could easily be done at other frequencies. Of course, there's some losses involved. Incidentally, the reason the photo shows 4 spark gaps in series is that gives you 4 lightning hits before your protection is gone and the line shorts out.

Well, there's:

formatting link
This is cute:
formatting link
ASCII art tools:
formatting link

Reply to
Jeff Liebermann

I did the math back in 74 but I forget the answer. :)

I think most is in the wire. But after about 1" or 2" of wire diameter there's only trivial amounts being carried in the center. Which is why they use pipes in many high current situations and I think high tension lines use a center core for strength and the outer windings for current.

Reply to
DLR

DLR wrote: (snip, I wrote some months ago...)

You might have missed an important point. I mentioned a steel conduit, that is, a high permeability material.

The answer is very different for aluminum.

Some National Electrical Code rules are based on this. For steel, it is mostly in the conduit.

-- glen

Reply to
glen herrmannsfeldt

Cabling-Design.com Forums 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.