|
At the end of a month of testing - I conclude "there is no such thing as a purely end-fed antenna" (hi) using UNUNs.
All end-feds using UNUNs to match impedances are really very "off-center feds" with a counterpoise of some sort.
I wanted to see where the natural resonances occur on my new end-fed: about 155 ft wire (#12 stranded) with 9:1 Palomar UNUN
It is fed with 100 ft RG213 coax with a choke at 90 feet from UNUN.
So perhaps it has a 90 foot "counterpoise" (whatever that means)
Here is the graph plotting RETURN LOSS (dB) and SWR from 3 to 20 MHz.
Looks like the biggest RETURN LOSS (almost 50 dB) and lowest SWR occurs at 9.3 MHz
If we use the formula L (feet) = 468 / f (MHz) = 468 / 9.3 = 50.3 feet for a 1/2 wave length
and for the 3/2 wave resonance that gives about 151 feet (close to my measured 155 ft)
So the formula works.
The shape of the wire is a giant horseshoe (running around my own yard) about 20 ft high.
I am only interested in NVIS for 80m and 40m NBEMS nets (and CW).
Might I mess with the counterpoise length or the wire antenna length to get a lower SWR on 80m and 40M?
I can live with a 2:1 SWR on 80m and 40m, or even a 3:1 SWR on the 100ft RG213 feed line, since the "added" attenuation due to higher SWR (3:1) on 100 ft RG213
is only about 1 dB of "additional loss" on 80 meters - only a fraction of one S-unit.
This 155 ft length is what was recommended by Palomar Engineering in their handouts.
It can also tune up nicely on the 30m (10.1 MHz) and 20m (14.1 MHz) and 17 meter (18.1 MHz) bands, which is great for DX
de k3eui barry
Feb 25, 2024
|
Thanks for posting, Barry!
So you are discovering there really is no "end fed" resonant antenna. The
proof is in the measurements as you show. So why is this abortion so
popular??? ONLY, and ONLY because it's so easy just to throw up a wire
into the trees.....
Sure, a wire thrown up into the grees is what I typically use out camping
and boondocking, but I would never consider it for home installation where
I have the options of optimizing efficiency and fitting the HF antenna(s)
to my local environment.
Dave - W?LEV
toggle quoted message
Show quoted text
On Sun, Feb 25, 2024 at 7:55?PM Barry K3EUI <k3euibarry@...> wrote:
At the end of a month of testing - I conclude "there is no such thing as
a purely end-fed antenna" (hi) using UNUNs.
All end-feds using UNUNs to match impedances are really very "off-center
feds" with a counterpoise of some sort.
I wanted to see where the natural resonances occur on my new end-fed:
about 155 ft wire (#12 stranded) with 9:1 Palomar UNUN
It is fed with 100 ft RG213 coax with a choke at 90 feet from UNUN.
So perhaps it has a 90 foot "counterpoise" (whatever that means)
Here is the graph plotting RETURN LOSS (dB) and SWR from 3 to
20 MHz.
Looks like the biggest RETURN LOSS (almost 50 dB) and lowest SWR
occurs at 9.3 MHz
If we use the formula L (feet) = 468 / f (MHz) = 468 / 9.3 =
50.3 feet for a 1/2 wave length
and for the 3/2 wave resonance that gives about 151 feet (close to
my measured 155 ft)
So the formula works.
The shape of the wire is a giant horseshoe (running around my own yard)
about 20 ft high.
I am only interested in NVIS for 80m and 40m NBEMS nets (and CW).
Might I mess with the counterpoise length or the wire antenna length to
get a lower SWR on 80m and 40M?
I can live with a 2:1 SWR on 80m and 40m, or even a 3:1 SWR on the
100ft RG213 feed line, since the "added" attenuation due to higher SWR
(3:1) on 100 ft RG213
is only about 1 dB of "additional loss" on 80 meters - only a fraction
of one S-unit.
This 155 ft length is what was recommended by Palomar Engineering in
their handouts.
It can also tune up nicely on the 30m (10.1 MHz) and 20m (14.1 MHz)
and 17 meter (18.1 MHz) bands, which is great for DX
de k3eui barry
Feb 25, 2024
-- *Dave - W?LEV*
--
Dave - W?LEV
|
If you are using a 9:1 unun then it's not an end fed. My EFW4010 gets me around the world on 100W SSB. My measured SWR is the same as simulated by EZNEC. If you're interested I can send you details when I get home.
--
Mike G8GYW
|
It should be noted that an antenna can be an "end fed" without being an
"end fed half wave." Antennas that are not half wave but are, in fact, end
fed, are often used with a 9:1 balun.
73,
Jim KX0U
toggle quoted message
Show quoted text
On Mon, Feb 26, 2024 at 2:09?AM Mike <mail@...> wrote: If you are using a 9:1 unun then it's not an end fed. My EFW4010 gets me
around the world on 100W SSB. My measured SWR is the same as simulated by
EZNEC. If you're interested I can send you details when I get home.
--
Mike G8GYW
|
Sorry Barry and Dave, I read your posts while half asleep and now realise you are talking about an end fed, not an end fed half wave!
Anyway, I thought I'd include details of my own measurements to show that a wire antenna can be used at the home QTH.
I have a small garden so my antenna options are limited but I managed to fit an EFHW4010 by bending the wire.
Wire 1 is a 2m coax counterpoise. At the bottom is a common mode choke connected to the shack by 18m of Ultraflex 7 coax. At the top is a 49:1 unun. Wire 2 length is 15.2m and Wire 3 is 5.6m. Height above ground is 7 metres.
I measured the VSWR at the choke using a NanoVNA-H4 and an FA-VA5 and the results were identical. Attached is a plot of the simulated and measured VSWR.
The resonant frequencies on the higher bands could be improved with a series capacitor but my internal ATU tunes them up without.
I don't know about your end fed but the optimum counterpoise length for an end fed half wave is around 0.05 of a wavelength.
--
Mike G8GYW
|
Hello Mike, I am interested in that EZNEC Simulation. 73 Peter -----Original-Nachricht----- Betreff: Re: [nanovna-users] where is the end fed natural resonance Datum: 2024-02-26T09:10:04+0100 Von: "Mike" <mail@...> An: " [email protected]" < [email protected]> If you are using a 9:1 unun then it's not an end fed. My EFW4010 gets me around the world on 100W SSB. My measured SWR is the same as simulated by EZNEC. If you're interested I can send you details when I get home. -- Mike G8GYW ?
|
My measured SWR is the same as simulated by EZNEC. If you're interested I can send you details when I get home.
I am interested in that EZNEC Simulation. ** Yes, also Modeling an End Fed antenna in Eznec requires precisely defining the geometry of the counterweight which, in certain cases, is constituted by the outer sheath of the coaxial. -- F1AMM Fran?ois
|
On Mon, Feb 26, 2024 at 03:52 AM, Jim wrote:
Antennas that are not half wave but are, in fact, end
fed, are often used with a 9:1 balun.
These are often referred to as "end-fed RANDOM WIRE" to distinguish them from the "end-fed HALF-WAVE" type. The end-fed random wire user needs to avoid resonant wire lengths, the half-wave type requires them. 73, Don N2VGU
|
A half-wave antenna has the maximum current flow in the center and 0 current flow at the ends. Ideally, it would be fed in the center. Current flow is what produces an electromagnetic field, and at the ends there is nowhere for the current to flow. So in an ideal half-wave antenna, the impedance at the endpoints is infinite. An end-fed antenna would have to be either a little longer than a half wave or a little shorter than a half wave for it to work. That would give a very high (but not infinite) impedance at the ends. But the antenna will not be resonant. It will still work, for our purposes. But I'd rather put up a center-fed half-wavelength dipole. 73, Zack W9SZ On Sun, Feb 25, 2024 at 2:33?PM W0LEV <davearea51a@...> wrote: Thanks for posting, Barry!
So you are discovering there really is no "end fed" resonant antenna. The
proof is in the measurements as you show. So why is this abortion so
popular??? ONLY, and ONLY because it's so easy just to throw up a wire
into the trees.....
Sure, a wire thrown up into the grees is what I typically use out camping
and boondocking, but I would never consider it for home installation where
I have the options of optimizing efficiency and fitting the HF antenna(s)
to my local environment.
Dave - W?LEV
On Sun, Feb 25, 2024 at 7:55?PM Barry K3EUI <k3euibarry@...> wrote:
At the end of a month of testing - I conclude "there is no such thing
as
a purely end-fed antenna" (hi) using UNUNs.
All end-feds using UNUNs to match impedances are really very
"off-center
feds" with a counterpoise of some sort.
I wanted to see where the natural resonances occur on my new end-fed:
about 155 ft wire (#12 stranded) with 9:1 Palomar UNUN
It is fed with 100 ft RG213 coax with a choke at 90 feet from UNUN.
So perhaps it has a 90 foot "counterpoise" (whatever that means)
Here is the graph plotting RETURN LOSS (dB) and SWR from 3 to
20 MHz.
Looks like the biggest RETURN LOSS (almost 50 dB) and lowest SWR
occurs at 9.3 MHz
If we use the formula L (feet) = 468 / f (MHz) = 468 / 9.3 =
50.3 feet for a 1/2 wave length
and for the 3/2 wave resonance that gives about 151 feet (close
to
my measured 155 ft)
So the formula works.
The shape of the wire is a giant horseshoe (running around my own yard)
about 20 ft high.
I am only interested in NVIS for 80m and 40m NBEMS nets (and CW).
Might I mess with the counterpoise length or the wire antenna length to
get a lower SWR on 80m and 40M?
I can live with a 2:1 SWR on 80m and 40m, or even a 3:1 SWR on the
100ft RG213 feed line, since the "added" attenuation due to higher SWR
(3:1) on 100 ft RG213
is only about 1 dB of "additional loss" on 80 meters - only a
fraction
of one S-unit.
This 155 ft length is what was recommended by Palomar Engineering in
their handouts.
It can also tune up nicely on the 30m (10.1 MHz) and 20m (14.1 MHz)
and 17 meter (18.1 MHz) bands, which is great for DX
de k3eui barry
Feb 25, 2024
--
*Dave - W?LEV*
--
Dave - W?LEV
<> Virus-free.www.avast.com <> <#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>
|
I think your assumption is wrong.
When we feed an antenna, made of a stretched wire, from one end, in fact we build an antenna which looks like a dipole whose feed is off-center. The second strand is made up of the outer conductor of the coaxial which powers the antenna.
Attached to what Roy W. Lewallen says on page 80 of his Eznec notice (excerpt).
Afterwards, we are not obliged to believe R W Lewallen either; I believe it :)
--
F1AMM
Fran?ois
toggle quoted message
Show quoted text
-----Message d'origine----- De la part de Zack Widup Envoy¨¦ : lundi 26 f¨¦vrier 2024 14:50
|
I am not speaking of the method of feeding, only of the characteristics of
a wire antenna itself. Antenna engineering was one of my minors as an
electrical engineering student. The other one was lasers.
If you do the integrals of the current distribution on a wire antenna from
end to end to determine the electromagnetic field at a point in space, you
can get the Poynting vector, which shows what the antenna pattern is. As I
mentioned, the current distribution along a half-wave wire antenna is
sinusoidal with the current at maximum at the center and zero at the ends.
You have to use an antenna that is slightly longer or shorter than an exact
half-wavelength to get any current to flow at the feedpoint, no matter what
kind of feed you use.
Zack W9SZ
toggle quoted message
Show quoted text
On Mon, Feb 26, 2024 at 8:16?AM Fran?ois <18471@...> wrote: I think your assumption is wrong.
When we feed an antenna, made of a stretched wire, from one end, in fact
we build an antenna which looks like a dipole whose feed is off-center. The
second strand is made up of the outer conductor of the coaxial which powers
the antenna.
Attached to what Roy W. Lewallen says on page 80 of his Eznec notice
(excerpt).
Afterwards, we are not obliged to believe R W Lewallen either; I believe
it :)
--
F1AMM
Fran?ois
-----Message d'origine----- De la part de Zack Widup
Envoy¨¦ : lundi 26 f¨¦vrier 2024 14:50
|
Not exactly. You can feed any length of wire and it will radiate. A very short wire will have a very high impedance, so you¡¯ll need a lot of voltage to get some current to flow.
And the current distribution along the wire won¡¯t be sinusoidal - more likely a linear decrease from feed point to end. It has to be zero at the end.
A very popular antenna ¡°back in the day¡± (1920s) was a flattop - which is basically a short (relative to wavelength) wire with a big capacity hat of single or multiple wires, with the vertical connected in the middle. The radiating part of the antenna was the vertical wire - typically much shorter than a half wavelength (given that the frequencies in use had wavelengths >500 meters). The capacity hat, being symmetrical, doesn¡¯t contribute to the radiation.
Half wavelengths are just ¡°convenient¡± because they have a mostly resistive feed point impedance (even if fed at the high Z end, as opposed to the low Z middle). And not having a lot of reactance makes the feed network easier.
To relate this back to the NanoVNA - you can do some experiments with smaller antennas at higher frequencies. And then compare what you measure to what you model with a program like NEC (or EZNEC or 4NEC2, etc.). The deviations from the model are either ¡°effects of the surroundings¡± or ¡°measurement/fixture artifacts¡± and that¡¯s useful to get a feel for in its own right.
Some years ago I had a lot of fun fooling with making antennas using copper foil tape on paper (like the stuff you use for EMI shielding, or the stuff you use to do PCB repairs), and then hook it up to the VNA (an 8510 in my case, had access at work).
It¡¯s kind of fun to be able to ¡°see¡± how the feed point impedance of a dipole changes as you change the included angle from 180 degrees (70-ish) ohms to 120 degrees (50-ish ohms)
toggle quoted message
Show quoted text
On Feb 26, 2024, at 7:03?AM, Zack Widup <w9sz.zack@...> wrote:
?I am not speaking of the method of feeding, only of the characteristics of
a wire antenna itself. Antenna engineering was one of my minors as an
electrical engineering student. The other one was lasers.
If you do the integrals of the current distribution on a wire antenna from
end to end to determine the electromagnetic field at a point in space, you
can get the Poynting vector, which shows what the antenna pattern is. As I
mentioned, the current distribution along a half-wave wire antenna is
sinusoidal with the current at maximum at the center and zero at the ends.
You have to use an antenna that is slightly longer or shorter than an exact
half-wavelength to get any current to flow at the feedpoint, no matter what
kind of feed you use.
Zack W9SZ
On Mon, Feb 26, 2024 at 8:16?AM Fran?ois <18471@...> wrote:
I think your assumption is wrong.
When we feed an antenna, made of a stretched wire, from one end, in fact
we build an antenna which looks like a dipole whose feed is off-center. The
second strand is made up of the outer conductor of the coaxial which powers
the antenna.
Attached to what Roy W. Lewallen says on page 80 of his Eznec notice
(excerpt).
Afterwards, we are not obliged to believe R W Lewallen either; I believe
it :)
--
F1AMM
Fran?ois
-----Message d'origine----- De la part de Zack Widup
Envoy¨¦ : lundi 26 f¨¦vrier 2024 14:50
|
On Mon, Feb 26, 2024 at 12:43 PM, Fran?ois wrote:
My measured SWR is the same as simulated by EZNEC. If you're interested I
can send you details when I get home.
I am interested in that EZNEC Simulation. ** Yes, also
Modeling an End Fed antenna in Eznec requires precisely defining the geometry
of the counterweight which, in certain cases, is constituted by the outer
sheath of the coaxial.
--
F1AMM
Fran?ois
Hi Fran?ois, I have attached the EZNEC model. Remove the .txt suffix before running. -- Mike
|
Be careful when measuring wire antennas with a NanoVNA because static build-up can damage the input.
--
Mike
|
But to reiterate, the majority of "End Fed" antennas are not what the name implies. Electrically they are "off center feed" antennas with frequently, an indeterminate" length of "counterpoise". As a consequence, the actual performance, resonance, and feed point impedances, of most "end feed" antennas will vary as a consequence of the intentional or the inadvertent counterpoise made by the installation.
For more predictable results, install with an actual counterpoise from the balun to the end support insulator. Resonance, of course, with be a (now predictable) frequency, based on the total wire length, end to end, between the insulators. Feed point impedance will be a function of the off center ratio.
And B.T.W., now would be a good time to consider the advantage (electrically) of, instead, feeding with open wire feeder rather than (lossy at high VSWR) coax.
Obviously, most "End Feed" installers are looking for simplicity of installation (and an imaginary performance), so the use of open wire feed line is "beyond the Pale".
Lester B Veenstra K1YCM M?YCM W8YCM 6Y6Y W8YCM/6Y 6Y8LV (Reformed USNSG CTM1)
lester@...
452 Stable Ln
Keyser WV 26726 USA
GPS: 39.336826 N 78.982287 W (Google)
GPS: 39.33682 N 78.9823741 W (GPSDO)
Telephones:
Home: +1-304-289-6057
US cell +1-304-790-9192
Jamaica cell: +1-876-456-8898
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] [mailto: [email protected]] On Behalf Of Jim Sent: Monday, February 26, 2024 3:52 AM To: mail@... Cc: [email protected]Subject: Re: [nanovna-users] where is the end fed natural resonance It should be noted that an antenna can be an "end fed" without being an "end fed half wave." Antennas that are not half wave but are, in fact, end fed, are often used with a 9:1 balun. 73, Jim KX0U On Mon, Feb 26, 2024 at 2:09?AM Mike <mail@...> wrote: If you are using a 9:1 unun then it's not an end fed. My EFW4010 gets me
around the world on 100W SSB. My measured SWR is the same as simulated by
EZNEC. If you're interested I can send you details when I get home.
--
Mike G8GYW
|
Sorry for straying off the OP's topic but ...
I've never known an antenna polarise amateurs like the EFHW (pun intended).
Those who've built one properly love it. Those who haven't say it doesn't work. I've had QSOs from Texas in the west to Jakarta in the east on 100W SSB with my antenna at my QTH west of London.
--
Mike G8GYW
|
I always need to laugh when I hear such argument for an "antenna". I several times worked a VK station on the long path who was using a whip on his bicycle and 20W in SSB. Is that antenna better or worse considering the much greter distance ? 73 Peter -----Original-Nachricht----- Betreff: Re: [nanovna-users] where is the end fed natural resonance Datum: 2024-02-26T18:18:45+0100 Von: "Mike" <mail@...> An: " [email protected]" < [email protected]> Sorry for straying off the OP's topic but ... I've never known an antenna polarise amateurs like the EFHW (pun intended). Those who've built one properly love it. Those who haven't say it doesn't work. I've had QSOs from Texas in the west to Jakarta in the east on 100W SSB with my antenna at my QTH west of London. -- Mike G8GYW ?
|
And remember an EFHW is a 1/2-wavelength at only one frequency. Other than
that it's just an EFW: End Fed Wire.
Dave - W?LEV
toggle quoted message
Show quoted text
On Mon, Feb 26, 2024 at 8:52?AM Jim <jim.kx0u@...> wrote: It should be noted that an antenna can be an "end fed" without being an
"end fed half wave." Antennas that are not half wave but are, in fact, end
fed, are often used with a 9:1 balun.
73,
Jim KX0U
On Mon, Feb 26, 2024 at 2:09?AM Mike <mail@...> wrote:
If you are using a 9:1 unun then it's not an end fed. My EFW4010 gets me
around the world on 100W SSB. My measured SWR is the same as simulated by
EZNEC. If you're interested I can send you details when I get home.
--
Mike G8GYW
-- *Dave - W?LEV*
--
Dave - W?LEV
|
That sounds like confirmation bias. When conditions are good you can work the world on a wet noodle. When they are not good you need a kilowatt and an aviation hazard stack in the back yard. That's like saying "works great!" Only if you have not tried something that works greater. :) 73 -Jim NU0C On Mon, 26 Feb 2024 18:45:34 +0100 (CET) "Peter Voelpel" <dj7ww@...> wrote: I always need to laugh when I hear such argument for an "antenna".
I several times worked a VK station on the long path who was using a whip on his bicycle and 20W in SSB.
Is that antenna better or worse considering the much greter distance ?
73
Peter
-----Original-Nachricht-----
Betreff: Re: [nanovna-users] where is the end fed natural resonance
Datum: 2024-02-26T18:18:45+0100
Von: "Mike" <mail@...>
An: "[email protected]" <[email protected]>
Sorry for straying off the OP's topic but ...
I've never known an antenna polarise amateurs like the EFHW (pun intended).
Those who've built one properly love it. Those who haven't say it doesn't work. I've had QSOs from Texas in the west to Jakarta in the east on 100W SSB with my antenna at my QTH west of London.
-- 73 -Jim NU0C
|
?If an antenna is fed at the end it¡¯s an end fed regardless of length. There are 1/4 wave end fed AM broadcast and 1/2 wave end fed AM broadcast antennas.
The BALUN is not part of the antenna it¡¯s just a fixed antenna tuner with maybe a common mode choke
If the wire is a half wave then the feed point impedance will be high, 2000 ohms or more. The BALUN turns ratio depends on the antenna feed point impedance.
Ideally the feed point impedance would be measured and the appropriate BALUN constructed.
|
Barry K3EUI
Donald S Brant Jr
Jim Shorney