QUESTION:
>How do I eliminate the danger of a lightning strike? I can't ground the
>copper pipe because the center conductor of the coax is connected.
>Naturally, I am concerned about damage to any electronic equipment
>connected to this antenna, but I am also concerned about actual physical
>damage to my house from a strike. I live in a coastal region, and we
>sure get our share of thunderstorm and lightning activity. There must be
>some way that this danger is handled, since so many vertical antennas are
>in daily use by SWLs and hams.
You have to provide a path for lightning currents to be diverted to
Earth in such a fashion that they don't go into your house and your
equipment. This is, of course, easier when you use an antenna design
which is bonded directly to Earth, the so-called plumber's delight
beam is one example, a shunt fed vertical is another. Sometimes there
are reasons why you want to, or have to, isolate the antenna from
direct Earth connection. That makes the lightning mitigtion job
a bit more challenging.
In cases where the antenna must be electrically isolated from Earth,
you have to use a controlled discharge device. This is basically a
spark gap designed so that normal operating voltages won't fire it,
but so that the higher voltage of a strike will. The simplest spark
gap is just an air gap, but because of varying humidity and air
pressure, an air gap's firing potential is ill controlled. The
better approach is to use a special gas tube. These tubes are
manufactured with precise gas mixtures and pressures to ensure
a very predictable firing potential. You want to select a gas tube
which has a firing potential just above the normal transmitting
potential at the point of insertion. The suppressor should fire
very quickly, in less than 2 nanoseconds, and should become a
*very* low impedance when firing (typically on the order of
a few milliohms). Polyphaser makes suitable suppressors.
Now lightning energy is *RF*. The energy of the pulse is distributed
over a very wide frequency span from ELF up through UHF, but most of
the energy is concentrated in the range below 2 MHz. There is so
much energy in the lightning discharge, however, that you can't
ignore the higher frequency components. This presents quite a
challenge for antenna protection because of the many possible
resonances in the antenna, antenna support structure, feedlines,
and lightning downleads.
There will be nodes of high voltage and nodes of high current
distributed over these conductors by the lightning discharge.
Since the lightning energy spans such a large frequency range,
the number of such nodes is huge, and it is impractical to
analyze them all. However, there is one which is of interest.
The suppressor you install should have a firing time on the
order of 2 nanoseconds, so we are interested in the voltage
and current distribution at a frequency of f = 1/t or 500 Mhz.
If practical, we'd like to install the suppressor at a point
which is a multiple of a halfwave at 500 MHz from the antenna
tip. This will put it at a high voltage node for the strike
during the period when the gas tube is building to firing
potential, assuming it is the antenna tip which is struck.
That's good because it will allow the largest differential
between the required hold off voltage of the suppressor due
to your transmitted signal and the potential generated by the
lightning discharge. In other words, it maximizes the margin
for the suppressor.
In your case, it is more important to put the suppressor directly
at the antenna feedpoint, because that will be a current node for
your transmitted signal on 20m, and will maximize the margin to a
greater extent than if we ignored that and located the suppressor
back down the line where it would be at a lightning maximum. There
is a "sweet spot" a few inches back from the antenna base where
the highest differential will occur, but anywhere within those
few inches will suffice, and we might introduce extra unwanted
inductance in the lightning path by locating the suppressor back
from the feedpoint too far.
Now inductance is the killer in the path to Earth. We want to
minimize that to every extent possible because it will act to
elevate the suppressor above Earth during the strike. And that
will elevate your coax shield, which will elevate your equipment,
and that allows a large current to flow through your equipment
to *its* Earth connection.
Ideally, your feedpoint will be directly at ground level, with
an essentially zero length path to Earth for the suppressor.
Even a few feet of straight wire will have sufficient inductance
to elevate the coax shield. So keeping this path short and straight
is vitally important. (Obviously, any bends in the downlead will
add inductance.) The lightning downlead must be at least #6 wire,
but a flat copper strap has less self-inductance and a greater
skin area for a given amount of copper, so that's even better.
Note do *not* use braid. It has a high self-inductance and a
high skin resistance compared to smooth solid strap.
OTOH, we'd *like* to have some inductance in the feeder going
into the shack. This will raise the impedance of the path for
lightning currents trying to reach your station ground. Coil
a few turns of the coax into a choke balun near the feedpoint
to provide this inductance. The phone company uses a half turn
loop, called a "drip loop" for this purpose in their installations.
We'll use a few more turns here.
Now this is all standard stuff, but you have another problem with
your installation. As I understand what you said, you're going to
stand the element up next to your house. This introduces the
possibility of *sideflashes*. Sideflashes occur when the
impedance to Earth is lower through some other structure than
through your intended downlead path. This can happen because of
the resonance effects noted above, and because of plain inductance
in the downlead path. The rule of thumb is that all conductors
carrying lightning surge should be at least 5 feet from any
structure not intended to carry surge currents. So be sure to
stand your antenna off at least that far from the house.
If I've misunderstood, and you intend to mount the antenna on
the roof, then you've got to be *very* concerned about downlead
inductance. Keep it straight, keep it wide, and keep it short.
Lightning can be considered to be a constant current source
with a current on the order of 8000 amperes. The potential
backing the constant current source, IE the cloud to ground
potential, can be many millions of volts. There is no practical
way to insulate against that. When you try to impede that
current flow, due to resistance or inductance in the current
path, those millions of volts will express themselves across
the impedance and drive sideflashes. That's why it is so vital
to minimize inductance in the downlead, and why you need
a large standoff distance to other structures.
As a final barrier, be sure to use a suppressor in the coax
at the station entrance bulkhead. This bulkhead should form
the *ground window* for your station. A ground window is the
point to which *all* conductors entering or leaving the station
area are tied, either bonded directly if nominally of ground
potential, or tied via a suppressor if they carry operating
potentials. This includes *all* conductors, including station
AC power, the telephone line if a telephone is in the station
area, any CATV line, etc. The ground window serves as the
station *single point ground*. It is then tied in turn to
a good Earth ground via a direct low inductance path.
The purpose of the single point is to ensure that if the
station's nominal ground potential becomes elevated during
a strike, all conductors become elevated roughly equally.
If there is no potential difference, there can be no damaging
current flow through equipment attached to those conductors.
Note that all conductors entering the station area must be
tied to the ground window *first* before going on to the
station equipment. And don't create daisy chains inside
the station. All equipment grounds should go directly to
the ground window. Bypassing the ground window with just
one wire will negate its protection.
The latter can happen all too easily. For example, if you
run an extension cord into the station area "temporarily"
instead of using conditioned station power which has entered
the station area through the ground window, a strike can fry
all your equipment, even though you've done everything else
right.
Gary
