[Elecraft] KAT1 (K1) & end-fed halfwave wire antenna

[Ron D'Eau Claire]

You are right about the improved efficiency of the end fed dipole (i.e. half
wave) antenna. From the standpoint of radiation, it's identical in all
respects to a center fed dipole. The only difference is that you're feeding
at a voltage loop instead of at a current loop. For portable use, that means
no feeder is needed and you can still get the high-current portion of the
antenna up "in the clear" - at least 1/4 wave up if you're able to run it
vertically or as an "inverted L" with the bend at the center of the wire.

There are two problems with an end fed dipole. First, most automatic antenna
tuners (including the Elecraft tuners) cannot handle the very high impedance
of an end fed dipole. Secondly, such tuners do not isolate the rig from the
high RF voltage present at the end of the antenna. The whole rig tends to
"float" up to the high RF voltage at the end of the antenna which makes the
rig "hot" with RF. In modern rigs filled with logic chips, the RF can raise
havoc by producing false inputs to the logic here and there at random. In
extreme situations you can even feel the "bite" of the RF voltage when you
touch the rig, even at QRP power levels. Also, at a voltage loop,
capacitance to ground affects the tuning. Your body touching the rig easily
detunes the antenna system. 

The hot RF issue is the easiest to deal with, but makes the antenna more
complex. Ground the rig. It doesn't have to be a very efficient ground; just
something to reduce the RF voltage to an acceptable level. A 1/4 wave length
of wire isolated from ground and hooked to the case of the rig is perhaps
the simplest although all sorts of lengths of wire laid on the ground
usually work too. 

Matching to the high impedance at the end of a dipole is a little more
difficult to deal with using an automatic tuner. I currently use an end-fed
dipole in a temporary setup with my rig on the second floor of the house. I
use a manually-adjusted link coupled tuner that effectively decouples the
antenna from the rig. Unfortunately, such tuners are very difficult to
automate. It has two variable capacitors that must be adjusted and five coil
taps to adjust when moving from one band to another. Most automatic tuners,
including the Elecraft tuners, us an "L-network". It is an efficient and
flexible design, but because of the very high RF voltages that can be
present when the impedance is very high, the compact, automated versions
aren't designed to match the very high impedance at the end of a dipole. 

That's the bad news. The good news is that it's not necessary to use an
antenna exactly 1/2 wave long for good efficiency. While the 1/2 wave is
ideal, lengths longer or shorter, within reason, can still provide an
efficient antenna. As you move away from the half wave length, the impedance
drops and ground losses increase, so the idea is to find the length closest
to 1/2 wave that your ATU will still handle and which don't produce any
detuning effects when you touch the rig. 

Many folks do what you are considering: use a coil or a variable capacitor
in series with the antenna where it connects to the rig. A tapped coil is
more common because coils are less easily damaged in the field. The coil or
capacitor will change the impedance at the feed point so the ATU can find a
match. For highest efficiency, you want to start with an antenna a close to
1/2 wavelength long as possible, and then add only the reactance
(capacitance or inductance) needed to get a match; no more. The farther you
go from 1/2 wavelength, the lower the antenna efficiency. That is, unless
you can set up a good ground system. And that's why some portable operators
choose to go to the trouble to rig a 1/4 wave wire, insulated from
surrounding objects, as  a "counterpoise" connected to the rig case. You
need one for each band. You can use multiwire cable with each wire cut to
the right length, but be sure to separate the last few feet of each wire
from the others. The point about a 1/4 wave wire is that if you force one
end to be "hot" (at a high impedance) by insulating it from everything
around it, the other end you connect to the rig will show an impedance of
about 35 ohms: low enough to be a decent RF ground. You must separate the
ends from the other wires (and other objects) to keep them from detuning the
wire.  

BTW, your friend is right about the "switch". That's part of an L-network
like the Elecraft tuners use. In the Elecraft ATUs the switching done
automatically. It's part of the relay chatter you hear going on while the
ATU is searching for a match. 

Ron AC7AC

-----Original Message-----

The other day I posted an antenna question about making a K1 antenna 
that avoids halfwave multiples (& thanks again to the guys who replied 
offlist).  From the calculations in that note and some modeling by 
another op it seems the ~18m/60' is probably the length to try there.

But end-fed halfwave antennas have some advantages, especially the fact 
that as a voltage fed antenna they have less of a need for a low 
impedance ground to keep rf losses low.  On the other hand, they offer 
very high impedance to the tuner.  The KAT1 manual says to look out for 
"rf problems" at exact half wavelengths but doesn't come out and say if 
it can or can't handle it (even if just at certain ranges of length).

A longtime ham told me some ATUs used to have a switch to move a cap to 
one end of the matching network or another to offer a wider matching 
range.  So if the KAT1 can handle the impedance, great.  If not, maybe I 
can put a coil in series and tap it at the 50 Ohm point of the coil. 
I'm not sure if it's worth the effort, especially since it'd be nice to 
keep portable ops as simple as possible.  But if there's a noticeable 
performace increase...  Has anyone tried either method?  (That is, 
direct connection of halfwave end-fed or tapping an inductor.)  Thanks 
for any ideas.

73,
Mike  ab3ap