Hi,

I created a "omni-directional" vertical antenna that NEC-2 reports to
have a free-space gain 4 dBi. The shape of the antenna looks like:


------+
A |
| B
|
+-----+
C * (* = feed point)
+-----+
|
|
|
------+

The lengths can be adjusted to give the antenna a 50 ohm feedpoint
impedence. The overall length of wire forming the antenna (4A+2B+C)
is on the order of 1.5\lambda and the height (2B) is something like
\lambda. I built this antenna for 2-meters and it seems to perform
quite well. The .nec files and parameters are available from my
antenna pages at http://www.jedsoft.org/fun/antennas/omni.html.

I am sure that I am not the first to create this simple antenna,
nevertheless a google search has turned up nothing similar. Have you
seen such an antenna before and if so, what is it called? I suspect
that it belongs to some class of antennas (antennae?). I would
like to give the proper credit and name for it on my web page.

Also, if you can find a flaw in my NEC modeling of the antenna, please
tell me. The prototype that I built does have an SWR of 1.05:1 as
given by my uncalibrated meter at the design frequency.

Thanks,
--John

On 25 Sep 2006 17:40:22 GMT, (John E. Davis)
wrote:

I am sure that I am not the first to create this simple antenna,
nevertheless a google search has turned up nothing similar. Have you
seen such an antenna before and if so, what is it called?

Hi John,

A Stub Fed Doublet. A Stub Fed Dipole. A Stub Fed Short Dipole. A
Stub Fed Short Dipole with End Loading. Take your pick.

The end loads can be removed and the stub lengthened to 18.5 inches to
re-obtain resonance at a higher Z to demonstrate two principles:
1. End loading increases current into the structure to lower its
resistance;
2. Stub tuning works.

Several comments. What's with the curious cogging of the SWR computed
from NEC2? EZNEC predicts a quite smooth curve. Your measured values
suggest nearby losses. What do you do to snub common mode currents of
the nearby transmission line that is precariously close, and co-linear
with the polarization of your vertical dipole?

73's
Richard Clark, KB7QHC

John E. Davis wrote:
I created a "omni-directional" vertical antenna that NEC-2 reports to
have a free-space gain 4 dBi.

I created a somewhat similar antenna with 20+ dBi gain.

http://www.qsl.net/w5dxp/SUPRGAIN.EZ
--
73, Cecil http://www.w5dxp.com

"John E. Davis" wrote in message
...
Hi,

I created a "omni-directional" vertical antenna that NEC-2 reports to
have a free-space gain 4 dBi. The shape of the antenna looks like:


------+
A |
| B
|
+-----+
C * (* = feed point)
+-----+
|
|
|
------+

The lengths can be adjusted to give the antenna a 50 ohm feedpoint
impedence. The overall length of wire forming the antenna (4A+2B+C)
is on the order of 1.5\lambda and the height (2B) is something like
\lambda. I built this antenna for 2-meters and it seems to perform
quite well. The .nec files and parameters are available from my
antenna pages at http://www.jedsoft.org/fun/antennas/omni.html.

I am sure that I am not the first to create this simple antenna,
nevertheless a google search has turned up nothing similar. Have you
seen such an antenna before and if so, what is it called? I suspect
that it belongs to some class of antennas (antennae?). I would
like to give the proper credit and name for it on my web page.

Also, if you can find a flaw in my NEC modeling of the antenna, please
tell me. The prototype that I built does have an SWR of 1.05:1 as
given by my uncalibrated meter at the design frequency.

Thanks,
--John

Hi John

I really got interested in the configuration you show for the antenna.
But, when I looked more closely to the Elevation Plane Pattern that looks so
narrow (high gain), I realized that the antenna is very much the same
pattern as a basic full wave center fed wire.
The graduations on the plot graph was misleading to me.
Perhaps there is something special about this antenna that I am missing.

Jerry

On Mon, 25 Sep 2006 11:19:25 -0700, Richard Clark
wrote:
A Stub Fed Doublet. A Stub Fed Dipole. A Stub Fed Short Dipole. A
Stub Fed Short Dipole with End Loading. Take your pick.

I think I prefer "Stub Fed Doublet".

Several comments. What's with the curious cogging of the SWR computed
from NEC2? EZNEC predicts a quite smooth curve. Your measured values

I was also wondering about that. The SWR values are computed by
xnecview. It is conceivable that the choppiness of the curve is due
to numerical stability issues, e.g., using the difference of 2 small
numbers. Tonight I will dig out the xnecview source code and
investigate further.

suggest nearby losses. What do you do to snub common mode currents of
the nearby transmission line that is precariously close, and co-linear
with the polarization of your vertical dipole?

For the prototype, I tried to run the coax perpendicular to the
polarization to minimize the issue. Eventually I will use something
like a 1-1 choke balun. In fact, at the bottom of the web page I
suggest that something like that should be used. Of course I am
open to other suggestions.

Thanks,
--John

In article %4WRg.9158$Wi1.6469@trnddc06,
Jerry Martes wrote:

Hi,

I created a "omni-directional" vertical antenna that NEC-2 reports to
have a free-space gain 4 dBi. The shape of the antenna looks like:


------+
A |
| B
|
+-----+
C * (* = feed point)
+-----+
|
|
|
------+

The lengths can be adjusted to give the antenna a 50 ohm feedpoint
impedence. The overall length of wire forming the antenna (4A+2B+C)
is on the order of 1.5\lambda and the height (2B) is something like
\lambda. I built this antenna for 2-meters and it seems to perform
quite well. The .nec files and parameters are available from my
antenna pages at http://www.jedsoft.org/fun/antennas/omni.html.

I am sure that I am not the first to create this simple antenna,
nevertheless a google search has turned up nothing similar. Have you
seen such an antenna before and if so, what is it called? I suspect
that it belongs to some class of antennas (antennae?). I would
like to give the proper credit and name for it on my web page.

I really got interested in the configuration you show for the antenna.
But, when I looked more closely to the Elevation Plane Pattern that looks so
narrow (high gain), I realized that the antenna is very much the same
pattern as a basic full wave center fed wire.
The graduations on the plot graph was misleading to me.
Perhaps there is something special about this antenna that I am missing.

I believe that this antenna can probably be placed in the general
class of center-fed collinears. Other antennas in this class include
the center-fed fullwave, the classic Franklin antenna, and the EDZ
(extended double Zepp). The "Super-J" is a somewhat-similar design,
but is end-fed rather than center-fed.

The center-fed collinears of this sort tend to have a high (and/or
rather reactive) feedpoint impedance. They're usually fed through a
section of transmission line - often shorted at the end and fed via a
tap partway up the section... the "universal stub".

Based on the dimensions you posted, it looks to me as if this antenna
is pretty close to being an EDZ, but with the ends of the radiators
bent back sideways. I'd guess that by bending the ends sideways, and
fiddling with their lengths (and that of the matching section) you've
been able to match the 50-ohm feedline impedance without needing a
shorted/tapped matching section.

The elevation pattern of the antenna shows a hint of the high-angle
secondary lobes which characterize an EDZ.

So, I'd conclude that you've developed a variant on the EDZ (or
something partway between an EDZ and a center-fed fullwave) which
yields slightly lower gain than an EDZ but has a simpler matching
section.

The one thing I'd watch out for, with this design, is the folded-back
ends of the radiating arms. This design puts these high-voltage,
high-impedance points right at the mast, and this might make this
antenna more subject to mast/antenna coupling and de-tuning than a
traditional EDZ or full-wave center-fed.

The old ARRL VHF handbook has quite a bit of information on these
sorts of collinears, and has a nice writeup on the "universal stub"
matching technique (not very well known these days, but quite useful).

--
Dave Platt AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!

John E. Davis wrote:

Hi,

I created a "omni-directional" vertical antenna that NEC-2 reports to
have a free-space gain 4 dBi. The shape of the antenna looks like:


------+
A |
| B
|
+-----+
C * (* = feed point)
+-----+
|
|
|
------+

The lengths can be adjusted to give the antenna a 50 ohm feedpoint
impedence. The overall length of wire forming the antenna (4A+2B+C)
is on the order of 1.5\lambda and the height (2B) is something like
\lambda. I built this antenna for 2-meters and it seems to perform
quite well. The .nec files and parameters are available from my
antenna pages at http://www.jedsoft.org/fun/antennas/omni.html.

I am sure that I am not the first to create this simple antenna,
nevertheless a google search has turned up nothing similar. Have you
seen such an antenna before and if so, what is it called? I suspect
that it belongs to some class of antennas (antennae?). I would
like to give the proper credit and name for it on my web page.

Also, if you can find a flaw in my NEC modeling of the antenna, please
tell me. The prototype that I built does have an SWR of 1.05:1 as
given by my uncalibrated meter at the design frequency.

Thanks,
--John

Congratulations John, you appear to have re-invented a version of the
colinear vertical antenna on your own. I looked at your implimentation,
and it seems to be quite well done.

Keep it up, and have fun.

tom
K0TAR

On Mon, 25 Sep 2006 11:19:25 -0700, Richard Clark
wrote:
Several comments. What's with the curious cogging of the SWR computed
from NEC2? EZNEC predicts a quite smooth curve. Your measured values

I looked into this. The feedpoint impedence values Z(f) that NEC2
reports have discontinuities or discrete jumps, causing the predicted
SWR to have the same. I imagine that this is a result of the
numerical approximations and the segmentation used. Does EZNEC report
the oscillations when the frequency increment is on the order of 0.01
MHz? What version of NEC does EZNEC use?

I also tried using the extended thin wire kernel, but it did not help.
Nor did increasing the segmentation.

Thanks,
--John

On Mon, 25 Sep 2006 20:02:11 -0000, Dave Platt
wrote:
The elevation pattern of the antenna shows a hint of the high-angle
secondary lobes which characterize an EDZ.

This is definitely noticable on a log plot.

So, I'd conclude that you've developed a variant on the EDZ (or
something partway between an EDZ and a center-fed fullwave) which
yields slightly lower gain than an EDZ but has a simpler matching
section.

What sort of gain is expected from an EDZ? I tweaked the topology
of my design a bit in the hope of getting a broader bandwidth, but
instead got even more gain. For example, the version represented
below has a minimum gain of more than 4.7 dBi (at the side) and more
than 5.25 dBi in the forward direction. Previously I reported gain
values that ran from 4.2-4.7 dBi. So this one represents a bit of an
improvement. I may try building it this weekend. Thanks, --John

CM Model: expt5
CM parms = [4.68255, 3.41606, 36.4806, 7.22131, 11.9807, 3.02775, ];
CM
CM A1: 4-11/16 in.
CM A2: 3-7/16 in.
CM B1: 36-1/2 in.
CM B2: 7-1/4 in.
CM C: 12 in.
CM D: 3 in.
CM Wire diameter: 0.0640837
CM COM: (0 in., 9-1/16 in., 0 in.)
CM BBOX: dX=0, dY=11.9807, dZ=90.4316
CM COM Turning radius: 9.06615
CM Min Turning radius: 5.99035
CE
GW 1 3 0 0 9.10555 0 0 9.18245 0.000813863
GW 2 7 0 0 9.18245 0 0.30431 9.18245 0.000813863
GW 3 19 0 0.30431 9.18245 0 0.30431 10.1091 0.000813863
GW 4 5 0 0.30431 10.1091 0 0.30431 10.2925 0.000813863
GW 5 3 0 0.30431 10.2925 0 0.185373 10.2925 0.000813863
GW 6 3 0 0.30431 10.1091 0 0.217542 10.1091 0.000813863
GW 7 7 0 0 9.10555 0 0.30431 9.10555 0.000813863
GW 8 19 0 0.30431 9.10555 0 0.30431 8.17894 0.000813863
GW 9 5 0 0.30431 8.17894 0 0.30431 7.99552 0.000813863
GW 10 3 0 0.30431 7.99552 0 0.185373 7.99552 0.000813863
GW 11 3 0 0.30431 8.17894 0 0.217542 8.17894 0.000813863
GE 0
FR 0 201 0 0 144 0.02
EX 0 1 2 0 1
GN -1 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0
RP 0 1 73 1001 90, 0, 1, 5
EN

In article ,
John E. Davis wrote:

The elevation pattern of the antenna shows a hint of the high-angle
secondary lobes which characterize an EDZ.

This is definitely noticable on a log plot.

So, I'd conclude that you've developed a variant on the EDZ (or
something partway between an EDZ and a center-fed fullwave) which
yields slightly lower gain than an EDZ but has a simpler matching
section.

What sort of gain is expected from an EDZ? I tweaked the topology
of my design a bit in the hope of getting a broader bandwidth, but
instead got even more gain. For example, the version represented
below has a minimum gain of more than 4.7 dBi (at the side) and more
than 5.25 dBi in the forward direction. Previously I reported gain
values that ran from 4.2-4.7 dBi. So this one represents a bit of an
improvement.

For an EDZ in isolation (no mast) I believe it's around 3 dBd or 5
dBi, plus or minus a hair.

I modelled a copper-pipe EDZ, mounted at the top end of a conductive
mast. This one came out with 5.85 dBi in the favored direction, and a
2.5 dB front-to-back ratio.

CEDouble extended zepp for 2 meters
GW 1 7 0 0 2 0 0 6.5 1.25 # Upper segment of mast
GW 2 7 0 0 2 0 0 -2 1.25 # Middle segment of mast
GW 3 100 0 0 -2 0 0 -112.5 1.25 # Lower segment of mast
GW 4 15 0 0 2 13 0 2 0.5 # Upper support bar
GW 5 15 0 0 -2 13 0 -2 0.5 # Lower support bar
GW 6 7 13 0 2 13 0 -2 0.5 # Shorting bar
GW 7 7 13 0 2 17 0 2 0.5 # To top of balun
GW 8 7 13 0 -2 17 0 -2 0.5 # To bottom of balun
GW 9 15 17 0 2 28 0 2 0.5 # To top arm
GW 10 15 17 0 -2 28 0 -2 0.5 # To bottom arm
GW 11 40 28 0 2 28 0 46 0.5 # Upper arm
GW 12 40 28 0 -46 28 0 -2 0.5 # Lower arm
GW 13 5 17 0 2 17 0 -2 0.5 # Feedline attachment
GS 0 0 0.0254
GE 0
EX 0 13 3 0 1.0
FR 0 1 0 0 145.27 0.00E+00 0.00E+00 0.00E+00 0.00E+00
0.00E+00
RP 0 61 72 1101 0.0 0.0 3 5
XQ
EN

A stacked pair of them on a single mast yields 8.85 dBi in the favored
direction, with 2.38 dB front-to-back ratio.

--
Dave Platt AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!