Brownouts, shorts, explosions and the unix pc.

[Floyd Davidson]

In article <1991Jan7.084738.7441 at yenta.alb.nm.us> dt at yenta.alb.nm.us (David B. Thomas) writes:
>gil at limbic.ssdl.com (Gil Kloepfer Jr.) writes:
>
>>The best protection from lightning is unplugging your machine from
>>the power **AND** phone lines.

Good advice.

>My friend Mark (no credentials at all) reports that tying your cords
>in knots actually helps protect the equipment against lightning.
>He said he sustained an actual hit to his power lines, and the
>cord caught fire, but the equipment, which was plugged in, was
>unharmed.

First off, what is a direct hit?  Well a real direct hit involves
on the order of 100,000 to 300,000 amps of current at its peak.
NOTHING will protect you from that.  And it would melt even the
power line, never mind the telphone line.

Actually what we all get hit with is the current induced indirectly
into power or telephone lines.  The spectrum (frequency range) for
the energy that comes into your house on the power or tel lines
depends on how good a transmission line and how far the current has
to travel.

By the time it gets to your equipment there probably has been
quite a bit of rolloff in the upper frequency ranges due to the
simple fact that neither power or tel lines are designed to be 
used at anything other than relatively low frequencies.

Not that higher frequency components won't be there, just that they
will have been attenuated to some degree that will vary depending
on a number of things, including the distance involved.  There
is a difference  whether it comes from down the road a couple
miles or if it is in your backyard.

>
>My first thought was to balk....but consider this:
>
>The knots form inductors which have a fairly large reactance
>at rf.  Lightning has lots of rf energy (switch on a radio
>during a thunderstorm to prove this!).

Two problems:  1) it really is a very low reactance, and
2) the higher ranges of rf energy may or may not be what
comes down the line.

One tight turn in your tel line modem cord does not have even the
inductance of say a half mile of cable.  Several turns doesn't
either.

>So... a knot looks to lightning like an inductor with non-
>negligible inductance, and it will absorb some of the energy
>before the equipment gets any.
>
>To be honest, I haven't taken any of this seriously enough
>to have tied all my cords in knots yet, but hey --- it's
>probably worth a try.  I'd trust it no less than a box
>that costs $1000 and says "lightning eliminator" on it! :-)
>

All it would do is very slightly attenuate what has already
been attenuated the most.  Three or four turns might affect
frequencies above say 100 Mhz a little and above 200 Mhz a
lot.  I'm not sure what range is most common, or most
damaging.  But I'm sure it is lower than that.

If you do want to do something along these lines, wrap a couple
turns through a high Q toriodal core.  I don't know if it will
help with lightning protection, but it sure will help cut the
rf radiation from your computer down!

Floyd
-- 
Floyd L. Davidson                                  floyd at ims.alaska.edu
Salcha, AK 99714                    paycheck connection to Alascom, Inc.
 When I speak for them, one of us will be *out* of business in a hurry.