[RFI] Question re: Whole House Surge Protection

Tom Rauch w8ji at contesting.com
Thu Jul 22 06:15:04 EDT 2004

> One more thing to be aware of in the fine print of most of
those power
> company power protection services: They probably won't
cover lightning
> energy coming from anywhere else except the power feed.

There's a problem with that. It is almost impossible to
prove where the lightning came from and where it went.

In my case lightning almost always hits the tall tower but
the huge currents can melt smaller conductors that run
between "ground systems". An example is the really hard hit
hit I took in springtime during an ice storm. It blew a
control line and F-11 (like RG-11) cables completely in two
about 2000 feet from the tower, and melted the telephone
wires between my house and the pole.

There probably was enough voltage drop along the earth that
lighter cables even a long distance from the tower failed at
weak points.

How would I prove where the lightning actually hit, other
than my word that it was one hit on the tower?  The same
would be true if it hit the power line, and grounded through
the cables out to the tower. As a matter of fact the utility
company grounds are so poor that is often what happens.
Lightning hits the big target of the power lines, and seeks
ground by flowing as a COMMON MODE current on the drop wires
to the house, through house wiring, and eventually the path
stops at a good ground like a well or something else.

Suppression across the power line does very little, because
all it does is clamp the conductors together after they
reach several hundred volts difference. The real potential
difference isn't that way in the first place, it is between
all three drop wires and something else connected to the

> Tom is correct in that the way in which you ground and
bond things and
> route cabling can account for whether or not you'll have
problems due to
> lightning.  I'm amazed by Tom's boldness to go without a
single surge
> suppressor, but that shows his confidence in his
installation practices.

I just grew up before we had all these off-the-shelf
lightning cures, and had to be sure repeaters and BC
stations stayed on the air during and after storms. I have
old habits, and I'm sure had I not learned early on how to
ground things and route cables I would be singing a
different tune.

I'm not recommending anyone else NOT use protection devices,
I'm just saying how the cables enter the building and how
once in the building they enter the equipment means more
than anything else you can do.

The very same things that clean up RFI in my consumer
devices also cleans up lightning damage. I have a
four-square element for my 160 meter system about 80 feet
from the end of my house, so the common mode RF currents
induced in house wiring are huge. When a telco line enters a
room, it enters next to the wallplate that feeds any
transformers that run a phone. There are good HV bypass caps
between the telco line and series RF chokes, and the
capacitors "ground" to the power line safety ground wire.
The same thing that makes the telephone have no RF potential
between the outlets powering it reduces lightning potential.

My TV sets and audio/VCR equipment all has ONE entrance
cable to a multiple strip outlet. The metal shell of that
strip is grounded to bulk-head feedthroughs for ANY cable
feeding that system, and speaker wires leave after bypassing
to that same point. While I do that for RFI, it also
effectively cures lightning. There are off-the-shelf outlet
strips made that accomplish this same thing.

The entrance for all my feedlines and cables has bulkhead
feedthroughs, and a 4" copper flashing (maybe 30ft long)
runs to the utility ground. Outside the house I have 3/8th
inch diameter copper circling the entire house, and
everything bonds to it.

Every cable entering the operating desk does the same thing.
It all enters at one point. I have ONE power line feed for
everything, and the telco lines all bypass to the same point
and have series RF chokes to the outside wires.

To show how well this works, my receiving antennas all enter
through preamplifiers with small 12V relays for bypassing
right on the preamp boards. That is the only "disconnect". I
haven't changed or repaired a preamp in years.

The repeaters in a building at the base of a tower all have
similar wiring. The cables, including power feed, enter
through a wall plate mounted LOW on the building. That plate
is grounded to the tower ground. Inside the building cables
enter the racks the same way. They all enter low on the
rack, and all in a group through bulkhead feedthroughs.
Power lines enter at the same point, and the safety grounds
are grounded at that point.

All it takes is ONE conductor entering the hub incorrectly,
and that becomes a point of repeated failures.

This is the same scheme I used at commercial stations, and
we almost never had lightning failures other than antennas
occasionally blown into pieces. In the70's we had spark gaps
for lightning protection, so I certainly don't have a
problem with coaxial line arrestors. It probably is a good
idea to use them. But almost all of the protection comes
from eliminating ground loops through sensitive devices by
proper wiring. That always has to be done first.

Oh, one word of caution. Be SURE any lightning protection
device following a linear amplifier  has enough voltage
breakdown. Some protection devices are designed too close to
working voltage! You need actual worse case peak voltage
plus adequate headroom. Most radios have overshoot, so
transient voltages can be higher than from the 1500 watts
you think you are running.

Also if a distant hit causes the a voltage spike, it can
trigger a gas tube. Once it fires, your amplifier is dumped
into a short until the RF stops and the tube recovers.

At Ameritron, we found at least one or two cases a month of
gas tubes firing on voltage peaks causing equipment
problems. To see what happen if you lightning protection
devices short a feedline, look at the "practical
demonstration" at


this shows what happens to tank voltages when the load is
removed from an operating PA.

73 Tom

More information about the RFI mailing list