[Antennas] End-fed half-wave religion

[Jim Glover]

Danny--

The classical matching scheme for an end-fed half-wave is a resonant
circuit consisting of a coil and a variable capacitor, tuned to the
operating frequency, with either another coil coupled to the tuned
circuit, or a tap about 20% up the tuned circuit's coil.  For an
example, see:

http://www.g3ycc.karoo.net/endfed.htm

Standard web searching techniques can produce many pages with
variations on the same theme.  You might try google with search
parameters such as:

"end fed" "half wave" coil

Some suggest connecting a counterpoise at the low-voltage end of the
tuned circuit.  Others do not.

If it's going to be a ground-mounted vertical, then a ground plane
might be provided, to prevent ground losses.  According to analysis
I've read, which makes sense to me, the situation is no different
than if one were to mount a center-fed half-wave (a dipole) in the
same vertical configuration, with one end near the ground.  The
proximity of the ground to the center-fed half-wave would cause the
same sorts of losses that would result in a ground-mounted end-fed
half-wave.  Those losses, and the resulting need for ground radials
to prevent them, are, however, peculiar to the situation where the
end of the half-wave is near the lossy earth ground.  There is no
direct parallel to an end-fed half-wave in free space (or high above
ground, measured in wavelengths).

If, however, one chooses to mount an end-fed half-wave in a vertical
orientation near the ground, and not provide a counter-poise or
ground plane of any sort, one can do so, with a loss of around
3 dB or so.  One certainly could not choose to connect a quarter-wave
vertical (regardless of proximity to ground, or not) in a similar
fashion and expect it to work at all (except in the same manner in
which a dummy load "works").  In some situations (for example,
portable operation, or where radials are not feasible) it might be
considered a pretty good deal to be able to put up a ground-mounted
end-fed vertical which can operate without radials, at a penalty
of a mere 3 dB loss.

There are other options for matching/connecting an end-fed half-wave.
For example, there is the J-Pole... it has a matching mechanism, and
at the end of that matching mechanism (normally the "top" of it, but
that is not a requirement) one side is attached to the end-fed half-wave
portion of the J-Pole, and the other is attached to... "nothing".  Or,
ARRL publications such as the handbook and the antenna handbook show
horizontal end-fed half-waves fed with open line balanced feedline. 
Again, one side is attached to the end of the half-wave, and the other
side of the feedline simply terminates.

One may argue that an antenna feedline must in all cases be attached
to "something" on both sides, using the "complete circuit" argument.
That argument starts off a bit shaky, however, since in many types
of antennas, there's no clear path from the antenna element(s) 
connected to one side of the feedline, back to the element(s)
connected to the other side of the feedline.  People do argue that
there is still some kind of path, at least from an RF point of view...
but I have never seen any such argument that convinces me that an
actual antenna element must be present on the "return" side.  Perhaps
in the voltage-fed situation, it is sufficient for the other side of
the feedline to just...well... be there.

On some level, it is difficult to successfully argue that the other
side of the feed mechanism *must* be connected to something else,
because, hey... lots and lots of people are using antenna systems
in which there is no such connection, and many of them are well
experienced and educated amateurs who are very happy with the
performance of their antenna systems.  It's really hard to dismiss
them all as a bunch of dimwits who little realize that once in a
while, one can even make contacts on a dummy load.

I have seen people point out that there may be some slight skewing
of the pattern of some kinds of end-fed half-wave antennas from the
ideal version of the pattern when the "other" side of the feed
mechanism isn't connected to anything.  Nevertheless, such antenna
systems do, by and large, work.  Some of them, I hear, work quite
efficiently.

As far as I've been able to gather, there exist quite a few antenna
designs which are based on some sort of end-fed half-wave element,
and a feeding/matching mechanism which involves the "other" side
of the feed mechanism being connected to "nothing."  I gather
these antennas (J-Pole, just to cite one common example) work not
merely adequately, but downright impressively, at least in the
clear.  I also gather there are some losses (though not the huge,
dramatic losses other antenna systems would suffer) operating
these end-fed half-waves with one end on or near the ground,
without providing radials to improve the ground's ability to
reflect the signal.  But the evidence seems to suggest that there
are schemes for feeding/matching end-fed half-wave antennas which
involve connecting the "other" side of the feedline to "nothing",
and, notwithstanding arguments that such is impossible, do a fine
job anyway.

73,
Jim  WB5UDE
> 
> At 11:46 AM 7/14/2002 -0500, you wrote:
> >There's no need for a counterpoise of any sort to make any
> >of these function correctly and efficiently, is there?
> >
> >73,
> >Jim  WB5UDE
> 
> Please explain how you go about connecting a transmitter to that?
> 
> If you connect one conductor of your transmission line (or one terminal 
> from the transmitter) to the end of your single element where do you 
> connect the other end (or the transmitter's other terminal). And, most 
> importantly, what path does for the current use to get to the end of the 
> transmission line (or the transmitter's other terminal) which is not 
> connected to the antenna element?
> 
> 73
> Danny
> 
> 
> 
> 
> 
> 
> 
> - - - 
> 
> Your moderator for this list is:
> Larry Wilson KE1HZ antennas-owner@mailman.qth.net
> _______________________________________________
> Antennas mailing list
> Antennas@mailman.qth.net
> http://mailman.qth.net/mailman/listinfo/antennas
>