[TowerTalk] Feeding an End fed 1/2 wave.

Chuck Counselman ccc at space.mit.edu
Fri Mar 14 08:50:33 EST 2003


At 12:19 PM +0000 3/14/03, Robert Thain wrote:
>I would like to build an end fed 1/2 for 20M....But what I dont get 
>is how to connect the BALUN to the vertical element. There is only 1 
>element and you have 2 wires coming from the BALUN. 1 of the 
>connections
>goes to the vertical element, where does the other one go ?

The best way, which has worked extremely well for me with a 40-m 
half-wave antenna, is as follows:

As a combination balun and impedance transformer, use one-quarter 
wavelength of roughly 600-ohm open-wire transmission-line; i.e., two 
parallel wires, 14 AWG or 16 AWG, spaced about 5 inches apart.  I 
used Plexiglass rods as spacers.

Looking into the transmitter end of this line, you'll see about 50 
ohms resistive.  You can connect a 50-ohm coaxial cable directly to 
this end without a discrete balun transformer.  However, a coaxial 
common-mode choke, or "1:1 current balun" here would do no harm.

At the antenna end of the quarter-wave, 600-ohm, open-wire line, one 
of the two wires is connected to nothing at all.  Its end should be 
cut a few inches short and suspended in mid-air about five inches 
from the antenna conductor (and five inches from the second wire of 
the transmission-line).  I used a Plexiglass rod here, too.

The second wire of the transmission-line should be connected to the 
antenna conductor.

The point on the antenna conductor at which this one 
transmission-line wire is connected should not be the end; it should 
be about one-third of the way up from the bottom, in other words 
two-thirds of the way down from the top.

Changing the position of this connection point will change the 
impedance seen at the transmitter end of the quarter-wave, 600-ohm, 
open-wire line; so you can adjust it if necessary to get exactly 50 
ohms at the transmitter end, for a perfect match to 50-ohm coax.

Old-timers (or those who read old antenna books) will recognize this 
off-center, single-wire-fed antenna as a "Windom" (or "Wyndom" in one 
old book I've read); and they will recognize the quarter-wave, 
parallel-wire, feedline with one wire end floating at the antenna as 
a "Zepp" feed.  The name comes from "Zeppelin," because this 
arrangement was used to feed the end of a half-wave wire trailing 
behind, or hanging below, the Zeppelins.  All that I've done is to 
move the antenna-connection point away from the end  of the antenna 
in order to change the impedance at the transmitter end of the 
feedline to 50 ohms.

I modeled this antenna with NEC-4 before actually building it, 
measuring its input impedance with a noise bridge, and then using it 
on the air.  The NEC-4 simulation was so accurate that, after cutting 
all lengths in accordance with the simulation, I did not have to trim 
anything.

The NEC-4 simulation showed that radiation from the feedline (due to 
unbalance, in other words due to common-mode current on the feedline) 
was very small, more than 20 dB down IIRC.

If anyone wants them, I have and will happily email my NEC-4 input 
and output files.

73 de Chuck, W1HIS



More information about the TowerTalk mailing list