I disagree.
1. The NEC is based on presuming that the MINIMUM ground conductor be
only large enough to carry any current the hot conductor would and so
trip the protection with damaging the ground conductor. Its not
anticipating ground currents from outside sources, such as RF or
lightning. And the NEC is the MINIMUM for safety, not necessarily for
good engineering. There have been time periods where the NEC accepted
safety ground conductors smaller than the phase wires. E.g. 14 in 12
romex.
2. Length and width are both important for an RF ground. Looking at a
ground wire as a transmission line, the impedance seen at the radio end
is Z0 * tangent(length in wavelengths). Which means that unless exactly
at 1/4 or 3/4 wavelength (or other odd multiples of a wavelength) the
impedance is directly proportional to the CHARACTERISTIC impedance of
the transmission line. A flat sheet has a much lower characteristic
impedance than a round wire, hence for all practical purposes and except
at exactly the 1/4 wave resonance, the impedance to ground is smaller.
We accept the small wire in the open wire transmission line because the
characteristic impedance of the line is high and so except at the dipole
load, the transmission line currents are small and so the loss is small.
We correct for the changes in impedance using a tuner.
3. Both the effective conductance AND the current carrying capabilities
of the lightning ground are very important. Lightning currents can be a
few to several kiloamps. It can require more than a wire to handle them.
I've seen wires from equipment and houses after lightning where the
lightning had melted the surface and the magnetic fields had expelled
that melted copper out in a fan showing the magnetic field's shape when
the copper then froze preserving that shape. My comments in part 2 for
characteristic impedance apply to lighting protection.
The trouble with depending on grounds for lighting protection is that
the earth, wet or dry dirt is a rotten ground. So the ground rod (and
any rod under 8' long is probably more decoration than safe ground)
supplies much local resistance to the ground system. Multiple ground
rods are important, and series isolation is important. I prefer to use
several ground rods for the antennas (solid guy wires to screw in
anchors are a benefit to spread the grounds out) outside the shack, then
only a single wire to a patch panel connecting antennas and ground to
inside and when there's lightning about, I disconnect ALL including the
ground and maintain a significant (feet) air gap. Times when I have not
done that and have been hit, there's been damage all over the house.
Transmission lines wound into inductors between the tower grounds and
the interior grounds may have some benefit in diverting lightning
currents to the earth, but when a lightning strike can be 10,000 amps,
cutting it down by a factor of say 10 still leaves a current to the
radios capable of severe damage. And small air gaps are not much
benefit. If there's 10 KA strike to a 10 ohm ground system (and 10 ohms
for ground resistance is very difficult to achieve), there's still 100
KV voltage difference to other ground systems (e.g. the power line and
phone line) and that can jump 6 or 8 inches... So any isolator that
depends on a shorter air gap is false protection.
RF filters in power lines do nothing to dissipate peak energy, however
they do integrate short pulses into longer pulses of lower amplitude.
The lower amplitude pulse after the filter contains virtually the same
destructive energy. Because of that, MOV protective elements that are
based on voltage alone need to be on the line side of power line
filters. This is inconvenient with the filters that have an integral
line cord connector. I have conducted experiments using line impulse
generators to prove that a power supply protected by MOV outside the
filter would be protected from the magnitude of pulse that would
perforate the transformer insulation without the MOV and filter
combination.
73, Jerry, K0CQ
--
FAQ on WWW: http://www.contesting.com/FAQ/tentec
Submissions: tentec@contesting.com
Administrative requests: tentec-REQUEST@contesting.com
Problems: owner-tentec@contesting.com
|