[Top] [All Lists]

Topband: Ewe, Flag, and Pennant questions for the experts

To: <>
Subject: Topband: Ewe, Flag, and Pennant questions for the experts
From: (
Date: Thu, 21 Mar 2002 20:36:33 -0800
Mark, WA1ION asked a lot of questions about Flag and Pennant types of

"Ken Alexander did some Pennant optimizations for medium-wave (AM
broadcast) use.  He came up with a vertical-section height of 16.4 ft.,
slanting members lengths of 54.6 ft. each (for a horizontal length of
52.1 ft.).  By comparison, the standard 
Pennant dimensions (as developed by K6SE for 160 m use) are 14 ft.
vertical, 30 ft. slanting members, 29 ft. horizontal length.  If a
medium-wave Flag was built, would a 16.4 by 52.1 ft. one have any
advantage in front-to-back ratio over one with dimensions of 14 by 29
ft.?  Obviously it would have more signal output, but it's directivity
that usually matters most. "
The Pennant as described in the July 2000 issue of QST is a broadband
antenna, optimized for use on the 160, 80 and 40-meter amateur bands. 
With the proper termination, the feedpoint Z is close to 900 + j0 (purely
resistive) for that frequency range.  It is highly directional for the AM
BC band also, although some slight juggling of the dimensions and
termination resistor value might improve the antenna somewhat for the 550
to 1610 kHz range.

When optimizing a Flag or Pennant for a specific frequency range, one
should always strive for the best possible F/B ratio at the desired
signal arrival angle, and the final design dimensions and termination
value should yield a purely resistive Z at the feedpoint of the antenna.

"Deltas (including Neil Kazaross's squashed-delta) are fed at one of the
lower corners and terminated at the other.  How would a Flag shaped
antenna fed and terminated at lower corners do 
compared to a conventional Flag fed and terminated halfway up its
vertical sides ?  This would be like a Ewe with a wire connecting across
the bottom instead of using ground rods at each end.  Is 
this a workable design if the bottom wire is, let's say, about 5 ft. off
the ground ?  Would it be as ground-independent as a regular Flag ?  How
about immunity to pattern influence by nearby antennas and metal objects:
better, worse, or same as a regular Flag ?  Conventional Ewes have
dimensions (vertical height, horizontal length) of 10 by 27 ft. for the
80/160 m model and 15 by 38 ft. for the 160 m optimized model (this would
presumably also be good at medium wave).  Bruce Conti in NH is using a 50
by 75 ft. Ewe with good medium wave results. It seems to work a bit
better when the bottom wire scheme is used in addition to, or in place
of, the two ground rods.  That ratio seems to lopsided in favor of the
vertical sides, especially considering that Neil's "Kaz antenna" (delta)
has a 1 to 4 height to base length ratio instead of 1 to 1.5.  The Koontz
Ewe ratios in the 1 to 2.6 range and the Cunningham Flag ratio of 1 to
2.1 seem to be the happy medium values here.  The Ewe-Flag hybrid
(groundless Ewe) - if it works - may have its best performance around a 1
to 2.4 ratio such as 20 ft. side height by 48 ft. base and top length for
a decent medium wave and 160 m set-up."

First, a Flag terminated and fed at its bottom corners was tried when I
designed the Flag, but it could only be made to work over a reletively
narrow frequency range, which was contrary to my design criteria.  Stay
with the feedpoint and termination in the centers of the vertical
sections and the antenna will work quite well over several octaves.

The Delta-shaped version of this antenna I designed for the FO0AAA
Clipperton Island DXpedition falls into this category.  It was designed
to be used on the 160-meter amateur band and its dimensions/termination
value are optimized for that frequency.  It is the only design which
requires only one support and can be easily rotated mannualy, a very
desireable feature for DXpediton use.  The Clipperton team found the
antenna to be very successful and many other subsequent DXpeditions have
also used the Delta configuration.

Second, the dimensions that Mark cites for the Ewe are correct for only
one soil conductivity type.  Over soil of better or worse conductivity,
that Ewe will perform poorly unless its demensions and termination are
adjusted for the soil conditions.  This anomaly with the Ewe is the
primary reason that EA3VY and I designed the ground-independent Flag and
Pennant types of antennas.  A properly designed Flag or Pennant will
exhibit a high F/B ratio over any type of soil and at virtually any
height above ground without need to change its dimensions or termination
value.  If Bruce Conti in NH finds that his Ewe works better with a
ground wire beneath it, that means the dimensions are incorrect for his
soil conditions there.

Third, the height by length ratio of the Flag is not a "happy medium"
value.  The correct dimensions were arrived at by designing the antenna
to be broadband (i.e., exhibit zero reactance at its feedpoint over a
very wide frequency range).  This required that the termination resistor
value be equal to the feedpoint resistance.  To meet these criteria, the
height and width dimensions had to be juggled until everything fit into
place.  Hardly a "happy medium".  The Ewe's dimensions and the K9AY's
dimensions are also not "happy mediums".

Flags, Pennants, Ewes, etc. are vertically-polarized antennas and their
proximity to a vertical transmit antenna will degrade their performance
dramatically in the range of frequencies that the transmit antenna is
resonant.  The only cure is to either move the Flag at least a wavelength
away from the TX antenna, or detune the TX antenna while receiving.

"I have done some initial tests that indicate that feeding both ends of a
Pennant, Flag, or Kaz delta through 16:1 transformers (Mini-Circuits
T16-6T or a d-i-y equivalent) effectively loads each end with 800 ohms
and gets you reasonably good front-to-back ratio when you switch from one
coaxial feedline to the other.  W7IUV amps can be used on these lines in
the "shack" or out in the same boxes as the transformers with not too
much difference in signal-to-noise ratio. The front-to-back may be
slightly less than could be had by feeding only one end and precisely
terminating the other, but the two-feedlines scheme actually gets
you superior nulling when the two feedlines are presented to the inputs
of a phasing unit such as an MFJ-1026, Quantum Phaser, or homebrew model.
 30 dB may be close to best-case cardioid null with a standard Flag or
Pennant.  The dual feedline version with phasing unit can easily do 40-50
dB of null on groundwave and often 30 dB on skip.  Two phased
quarter-wave spaced Ewes or 
Flags can probably beat this, but they won't fit in the typical small to
medium urban / suburban backyard either. 
The Mini-Circuits T16-6T core will saturate in an environment where there
is /are strong RF fields (i.e., in proximity to a BC xmtr).  The result
is intermod-sounding "birdies" on receive over the frequency range of the
antenna.  One DXpedition to the Caribbean used the T16-6T on their
Pennant and found it unusable due to the RF from a SW BC station
elsewhere on the island.  They replaced the xfmr with one on an Amidon
FT140-43 core and it cured the problem.  The Mini-circuits transformer is
definitely not recommended.  Either wind your own on an Amidon FT140-43
toroid core or purchase the correct item from the two manufacturers who
make transformers specifically for the Flag/Pennant.

73, de Earl, K6SE

--- StripMime Report -- processed MIME parts ---
  text/plain (text body -- kept)

<Prev in Thread] Current Thread [Next in Thread>