On Tue, 09 Nov 2010 14:41:49 -0500, Bob wrote:

I have the typical ICE lighning arrester at the junction (outside) where
the antenna joins the coax leading to my radio in a room upstairs.
The arrester is connected to a ground rod that I laborously banged into
the soil.

This has the potential (pun intended) to be the source (pun intended)
of considerable grief. An isolated ground (not connected to the
service ground as mandated by code) will lift during a strike and the
current at that point will seek the better ground by going through
your equipment - NOT pretty.

a. Can "nearby" strikes, say 1/8 or 1/4 mile away actually induce
meaningful voltages in the antenna (about 10 feet above the ground,
running horizontally) ? How large ?

You can get killer potentials on a summer's day with full sunlight and
blue skies. It comes from the accumulated charge carried by particles
in the air. As you describe your antenna as "end fed" and it is
feeding a "high resistance front end"; you are guaranteed high voltage
- n'est pas?

What kind of "protector" can I put right at the radio to try to protect
against any voltage surges (not direct hits) ?

Put a 50 Ohm to 5KOhm resistor across the antenna terminals at least
and let the rest of the (properly implemented) protection you
described from polyphaser do its own work.

Better yet, ground the antenna and couple to it through a gamma feed
(or delta match) or direct connect to a folded monopole.

I've also read about using neons back to back, and 10 meg ohn resistors,
etc.

You've done some lousy reading if the neons are back-to-back (they
have no "back" to be back-to-back). Neon bulbs in such a described
configuration are ADDING the potential where you are trying to REDUCE
it. What you describe is DOUBLING and GUARANTEEING the frontend
lethal potential. You may as well plug your antenna connector into
the AC socket.

Now, having said that, neon bulbs have a classical application in the
Collins R-390 (I know, because I used to teach the R-390 in the Navy);
and I know full-well that the frontend there was a tube, not a Bipolar
transistor nor a FET. This sort of creates the imprimatur of a magic
amulet that should be hung everywhere....

Another practice was Zener diodes reversed in parallel (the correct
topology of your poor reading source for neon bulbs) - which also
guarantee lethal potentials, or massive intermodulation problems in
the normal day-to-day use of your receiver. Looks good on paper - but
IFF you have a tuned front end (as was the R-390 in three stages) and
you don't (almost no radios are).

73's
Richard Clark, KB7QHC