I could draw an arc off the end of the coax connector about an inch long.
The rate of build/up and then discharge (how long it took between SNAPS)
correlated very nicely with the rate of snowfall. The harder it snowed, the
more rapidly the arc would happen, the more slowly it snowed, the longer in
between arcing from the connector.

Both of these, I would call "precip static"...one was dust, the other was
snow. I'll leave it to you guys to figure out the mechanism.

Hasan,

I think the problem here is there are two things at work.

The first is a slow (compared to radio frequencies) build up of charge
in an antenna that eventually breaks down a dielectric. I' like many
people, have seen that many times.
As a matter of fact that is common trigger for relay failures or SWR
detector failures in systems without a leak impedance that leaks off
charge faster than it builds.

The second is what is commonly called P-static. As described earlier
we had severe problems on tall tower and on the roofs of buildings.
This is a noise that starts as a slow rapid hissing or buzzing that
quickly builds almost to a musical tone as the weather gets more
severe. This is the mechanisim I disagree with being caused by
particles striking the antenna. The reason I disagree was clearly
stated:

1.) Taller structures are affected much more than lower ones that are
in the same weather

2.) Even one exposed protruding point seems to be the cause, since I
could cover that point in Station Master antennas and make it go away,
and it would go away by adding a taller blunt ended structure near the
affected antennas.

3.) With identical antennas stacked on a single tower, the uppermost
antenna is by far the most affected.

4.) I could get near the antennas and actually hear the accoustical
noise that matched the noise on repeaters and actually see the corna
ball.

....and on and on.

Cecil then proposed, if I am not mistaken, that P-static was caused by
particles striking the antenna, each one making a noise as it
discharged into the antenna, and that noise could be reduced by
grounding the element at DC. That is really the only point I disageed
with.

No one should even have a tall antenna or large antenna or any antenna
that is subject to high levels of charge that does not have a leakage
path to ground. That's just a given. But the idea a closed loop is
somehow quieter than an open element that also has a leakage path makes
no sense at all (especially when we are talking about noise from each
particle) unless the antenna is:

1.) Lower
2.) More blunt
3.) Has less protruding ends

My reasoning is the same number of charges hits each antenna, and the
greater the potential difference between those changes and the antenna
the greater the discharge power transfer to the antenna will be. So if
the antenna were allowed to float to avarage potential of space around
the antenna the noise from any energy transfer would actually be less.

This excludes the big pop when a dielectric breaks down.

My mobile antenna is matched by a shunt inductor, and that inductor
prevents the coax from charging up with dc. However, that inductor does
NOT make the antenna any more useful in weather like you described. It
is every bit as noisy, absent the sporatic loud pops when a dielectric
breaks down.

Say I leak that antenna with a 20K resistor, and replace that resistor
with a nearly zero ohm inductor that has high reactance at the
operating frequency. There will be no noise change.

The same applies to repeater antennas. We could use folded dipole
elements, elements in fiberglass radomes, and so on. The only thing
that helped was a blunter element or making the antenna lower in
relation to some other object on the roof.

We looked at dozens of systems hunderds of times, because no one wants
communications to vanish in inclement weather. We never found a "dc
grounded antenna" any help at all, although some leakage path probably
would have stopped the diode failure you had...but not the noise.

73 Tom