[Mldxcc] My QST article on high voltage relays

[Richard (Rick) Karlquist]

FYI, QST published my article on high voltage relays in
QST this month.  Here is some additional
background on my application:

I wrote an article for Hints and Kinks on antenna relays that
may be of interest to low band opearators.  It appears in the current
issue of QST (May 2009, page 66).  It describes how to modify
commonly available open frame relays to allow them to withstand
very high RF voltages.

These modified relays are useful for replacing traps in
multiband verticals.  For example, I have a 90 ft top loaded vertical
made out of 3 thirty foot lengths of aluminum tubing.  The
tubing is joined with PVC pipe so that the joints are insulated.
Relays are installed across the joints so that the height of
the vertical can be 30, 60 or 90 ft.  Then it can operate on
40, 80, or 160 meters.  The lower relays not only control RF
conduction across the joint, but also convey DC for the coils
of the upper relays.  RF chokes are across all joints to keep
the whole mast at DC ground.

This system was originally implemented with vacuum relays.
Initially, it worked perfectly, but within a year, the 60 foot
relay failed.  Several times, I replaced it, only to see it
fail within 6 months to a year.  Perhaps it arced over and
fouled the vacuum or the contacts welded.  The relay was
rated at 8000 volts, so it should have been OK with anything
my 1.5 kW could throw at it.  Also, we basically don't have
lightning at our QTH.  Whatever the mechanism, the relay always
failed eventually.

The new relay is a modified open frame relay.  The relay has
been modified to spread the contacts apart by 1/2 inch.  This
should withstand any conceivable voltage short of a lightning
strike.  And if there is an arc, no harm will be done.  With
paralleled 30 Amp contacts, contact welding should be eliminated
as well.

I haven't had time to install these relays to try them out.
But QST wanted to publish the article anyway (they approached
me).  I plan to get the relays on the air as soon as that
"round tuit" turns up.

Of course, the question will arise as to "why not just use
traps?".  Four reasons.  First, building and tuning a trap is
a lot more work than modifying a relay.  Second, traps act as
unwanted loading coils on the lower bands.  I don't want shortened
loaded verticals.  Third, they decrease the bandwidth on the
higher bands.  Without traps, it is possible to cover the whole
40 meter band without a tuning adjustment, and 80 meters can
be covered in 2 or 3 settings.  With traps, it would take at
least 2 settings for 40 meters, and many settings for 80 meters.
Fourth, in a phased array on 80 meters, I don't want current
flowing in the upper section (above 60 feet).  This could
distort the pattern.  A relay will do a better job of stopping
this current than a trap.  Especially away from the exact
resonant frequency of the trap.

It is important to use the relays in a configuration that does
not put high voltage between the relay coil and the moving
contacts.  This is not a problem when using them as trap
replacements as long as the lower section goes to the moving
contacts and the upper section goes to the fixed contacts.

Rick N6RK