[BC] More thoughts about AM transmitter lightning sensitivity issue

[Gregory Muir]

First of all, I wanted to thank you and everybody else for the timely 
response and wide array of comments and suggestions.  From this I have 
developed some more ideas about what to consider.

I had an opportunity to discover some additional important information while 
talking with our director of engineering a little earlier today.  Prior to 
my arrival at this firm, he happened to be at the site during the onset of a 
thunderstorm.  At this point the storm was many miles off in the distance 
and clouds were just beginning to form over the site.  At that point, the 
lightning strikes were accompanying the center of the storm.  And it was 
stressed that the strikes were miles from the site with no activity in the 
immediate region.

He walked out to the active daytime tower and observed arcing across the 
spark gap concurrent with each lightning strike.  The transmitter responded 
with a power cutback.  This started me to remember what I had learned about 
studying cloud physics some time ago regarding cloud potential equalization 
and the fact that even though actual strikes may take place dozens of miles 
away from the location in question, the re-equalization of cloud potential 
throughout the entire cloud layer after each strike still happens.  This in 
itself could induce an instantaneous  potential in the tower high enough so 
as to cause momentary breakdown of the spark gap and, as expected, a VSWR 
event which the transmitter would see.

Mind you, this is no gradual buildup of static on the tower.  Instead it is 
a rapid transient.  The slow buildup of charges is effectively handled by 
the DC path to ground afforded by the static drain choke.  In that light, I 
must scrutinize any application using a static dissipation array since it's 
intended purpose is to minimize general static charges in the tower, not 
rapidly induced ones (unless someone can effectively instruct me as to how 
such an array could minimize a quickly induced charge before the spark gap 
breaks down.)

Secondly, it was suggested that the spacing of the tower spark gaps be 
checked and adjusted for optimum distance.  This was added to my list. 
Widely spaced balls would not only cause the charge to build up to a higher 
potential before conducting, that potential could also reflect back towards 
the transmitter feed at a much higher magnitude creating more problems for 
the transmitter.

Another suggestion involved adding an in-line high-power attenuator between 
the transmitter and the line during storms.  This is similar to adding an 
attenuator to the output of a signal generator or network analyzer when 
performing tests requiring high precision source impedance matching to the 
load.  Not only does it guarantee a 50 ohm source impedance, it also helps 
isolate the load from the source and vice versa.  But the main drawback here 
is the power wasted while such a device is in use.

I have also heard about use of ferrite products around the transmission line 
at the transmitter.  This is a good approach to use when one wants to 
restrict stray RF currents from circulating in the sheath of the coax and 
may serve to minimize return transients from lightning events especially if 
they are caused by ground bounce (ground potential differences).  Something 
to consider.

A comment was made regarding tower guy wire grounding.  No attempt was ever 
made to ground any of the wires at the anchors because the first set of 
insulators are actually located about six inches above each turnbuckle eye. 
In addition, the anchor rods extend nearly 10 feet into the ground before 
coming in contact with the concrete dead man.  A physical inspection of the 
anchor rods about two years ago revealed no corrosion and a layer of paint 
that was applied prior to burial of the rods was found to be still intact at 
several inspection points.

The ground system for this site is quite impressive.  Having been 
constructed in 1947 it has a welded copper mesh ground plane over the entire 
property with extremely heavy 4 inch copper strap running between each of 
the towers and also to the transmitter building.  The ground system is in 
very good shape and was inspected nearly two years ago.  At present I don't 
have much concern about it's integrity but will be taking a closer look if 
some of the "easier" fixes don't work out.

Thanks to all,

Greg