Curtain Antennas?
SMB
I was looking to try and find some information on this type of antenna to
try and construct one for 6 or 10 meter.
Any thoughts?
SMB
Scott
of the October 1991 QST has one for 10 meters. It is simply made
up of several phased 1/2 wave elements with the phasing lines
make up of 450 ohm twin lead feed line. It would be easy to
calculate one for 6 meters using the info in the QST article. Of
note here is that the 10 meter Sterba curtan is 66 feet long!
The 6 meter antenna would almost be half that size. Good luck!
-----------------------------------------------------------
Got questions? Get answers over the phone at Keen.com.
Up to 100 minutes free!
http://www.keen.com
Tom W8JI
<n4zouN...@netzero.net.invalid> wrote:
>You are looking for a classic "Sterba curtan" antenna. Page 26
>of the October 1991 QST has one for 10 meters. It is simply made
>up of several phased 1/2 wave elements with the phasing lines
>make up of 450 ohm twin lead feed line. It would be easy to
>calculate one for 6 meters using the info in the QST article. Of
>note here is that the 10 meter Sterba curtan is 66 feet long!
>The 6 meter antenna would almost be half that size. Good luck!
One thing to remember about Sterba curtains (and Bruce arrays) is that
all the sections are in series. That means any errors in length add as
each section excites the next section. They are sensitive to frequency
change no matter how you build them, and are very sensitive to
moisture if you use ladder-line or twinlead in the phasing sections.
They never work near as well as predicted or modeled when in the "real
world", and they were frequency sensitive. That's why they fell out of
favor in commercial stations.
"USIA Curtains" solve this problem by bringing all the feedlines to
common points. The antenna is much more forgiving for construction
errors or a less-than-clear location. The system is less affected by
wet weather. These are the curtains used by SWBC stations like VOA and
other large SWBC stations. They have more gain than Rhombics even
though they occupy much less physical space.
Additionally, a simple distributed-feed curtain (like the USIA
Curtain) will work perfectly fine with nearly 3:1 frequency ranges. A
single curtain can easily be used on ten and six meters.
The price you would pay is extra ladder-line blowing in the wind, but
the payback is a multiband antenna that will work more like you
plannned or expect.
73 Tom
SMB
Thanks for the info.....
I looks like I have some more research to do about the how too's to
construct one....
Hey Tom, what do you think, should I just go buy a Fractenna?
;-)
hehehehehehehhe
NOT!
Thanks for the info guys,
SMB
KD5KZN
Tom W8JI <2w...@contesting.com> wrote in message !
Richard Karlquist
Tom W8JI <2w...@contesting.com> wrote:
>
>"USIA Curtains" solve this problem by bringing all the feedlines to
>common points. The antenna is much more forgiving for construction
Sounds interesting. Can you possibly give us a pointer to some
detailed information on these curtains?
Rick N6RK
Tom W8JI
wrote:
>Sounds interesting. Can you possibly give us a pointer to some
>detailed information on these curtains?
>
>Rick N6RK
I'm not aware of any amateur publication articles describing the
engineering behind these antennas, but you can see a commercial design
in the Antenna Engineering Handbook by Jasik.
In my first edition copy of that book...
Chapter 21 describes engineering, section 21.6 shows an actual USIA
Curtain.
An antenna for amateur use could be made more simple, since power
levels are lower. In an amateur installation impedance controlling
schemes (such as the very fat elements and low impedance lines) could
be eliminated.
Basically the feed is a distribution-type feed, where all elements are
fed in parallel. This eliminates the main design flaw of Sterba and
Bruce type arrays, where the system design connects the elements in
series.
73 Tom
Richard Harrison
"These are the curtains used by SWBC stations likke VOA and other large
SWBC stations. They have more gain than Rhombics even theugh they occupy
less physical space.----Additionally, a simple distributed-feed curtain
(like the USIA Curtain) will work perfectly fine with nearly 3:1
frequency ranges. A single curtain can easily be used on ten and six
meters."
Not without substantial adjustments for matching, unless the laws of
physics have been repealed. We had to make substantial adjustments to
switch a curtain between 6 MHz operation and 7MHz operation. You don`t
have to do that with a rhombic!
The curtains I`ve seen were (4) 1/2-wave dipoles, (1) in each quadrant
in a plane perpendicular to the direction of transmission. All are
driven in-phase. These are backed with an identical (4) 1/2-wave dipole
parasitic array.
A dipole has 0 dbd gain. (2) have 3 dbd gain. (4) have 6 dbd gain. (8)
have 9dbd gain. A rhombic has 14 dbd gain.
The dipoles in a curtain array use (2) conductors for broadbanding. The
reflector leads are brought down, just like the driven element
feedlines, for tuning adjustments. All this wire means that the curtain
takes as much as a rhombic with 4 or 5 wavelengths per side. The ringer
is that transmitting rhombics usually use triple wire, spread at the
side towers for broadbanding the drivepoint impedance. This isn`t done
on receiving rhombics, usually.
(4) towers support a curtain, (2) each for the radiator array and (2)
each for the reflector array. They are about twice the height of the (4)
towers needed to support a rhombic.
The advantage is that curtains are usually installed in rows so that
adjacent curtains share towers. The practice with rhombics is several
wavelengths of lateral spacing for isolation. So, Tom is right about
less physical space for curtains.
SWBC curtain antennas are good for two adjacent SWBC bands. Ours were
specific to 6 & 7 MHz, 9 & 11 MHz, and 15 & 17 MHz. They were
definitely no good over a 3:1 frequency range. A rhombic is needed for
that job.
The reason for choosing a curtain over a rhombic is to get wider
beamwidth to cover a target area from side-to-side without consuming
extra energy in a dissipation resistance (the choice is still close to a
toss-up) and to save real estate. For point-to-point you would choose a
rhombic to avoid all those tuning adjustments when changing frequency.
Best regards, Richard Harrison, KB5WZI
Tom W8JI
(Richard Harrison) wrote:
>Tom, W8JI wrote:
>"These are the curtains used by SWBC stations likke VOA and other large
>SWBC stations. They have more gain than Rhombics even theugh they occupy
>less physical space.----Additionally, a simple distributed-feed curtain
>(like the USIA Curtain) will work perfectly fine with nearly 3:1
>frequency ranges. A single curtain can easily be used on ten and six
>meters."
>
>Not without substantial adjustments for matching, unless the laws of
>physics have been repealed. We had to make substantial adjustments to
>switch a curtain between 6 MHz operation and 7MHz operation. You don`t
>have to do that with a rhombic!
Of course you'd have to re-adjust the LTU (antenna tuner).
A Rhombic, on the other hand, would have nearly the same impedance
over a wide frequency range, which is the Rhombic's main selling
point.
My point is the antenna would not require re-adjustment, a single
antenna would work very well and have the same directional
characteristics on both bands...and everything in between. Actually it
would be easy to cover 21 through 50 MHz with a single antenna.
>The curtains I`ve seen were (4) 1/2-wave dipoles, (1) in each quadrant
>in a plane perpendicular to the direction of transmission. All are
>driven in-phase. These are backed with an identical (4) 1/2-wave dipole
>parasitic array.
Curtain come in various flavors. You obviously saw a simple low-gain
single-band curtain.
>A dipole has 0 dbd gain. (2) have 3 dbd gain. (4) have 6 dbd gain. (8)
>have 9dbd gain. A rhombic has 14 dbd gain.
So what? The typical curtain at VOA fits well inside the area of the
Rhombics, and has several dB more gain.
>SWBC curtain antennas are good for two adjacent SWBC bands. Ours were
>specific to 6 & 7 MHz, 9 & 11 MHz, and 15 & 17 MHz. They were
>definitely no good over a 3:1 frequency range. A rhombic is needed for
>that job.
You obviously haven't seen a USIA style curtain. You've seen a single
band low-gain design that wastes materials but is definitely
physically small.
Now take your simple curtain, and put a wire reflector behind it
instead of a parasitic element. Or just feed it as a bi-directional
array. Bring all feedlines in through an equal length branching feed
system to a common point.
No matter what frequency excites the antenna, all the dipoles are in
phase. At some frequency less than the frequency where the dipoles are
1/2 wl long, efficiency drops. At a frequecy where each dipole is over
1-1/4 wl long, the antenna develops what might be objectional grating
lobes (although the main lobe continues to be in the same direction.
Gain increases with frequency until at some frequency above 2.5 times
the lowest frequency gain starts to fall off again.
So much for the claim of single-band or close-spaced band operation!
>The reason for choosing a curtain over a rhombic is to get wider
>beamwidth to cover a target area from side-to-side without consuming
>extra energy in a dissipation resistance (the choice is still close to a
>toss-up) and to save real estate. For point-to-point you would choose a
>rhombic to avoid all those tuning adjustments when changing frequency.
That simply isn't true. The curtains at VOA and other BC stations
(Radio Canada, etc) have more gain AND a cleaner pattern than
Rhombics...even though the Rhombics are much bigger antennas. Not only
is very little power lost as heat, very little power is lost in minor
lobes. Tuning is as simple as switching in a different stub.
Of course Rhombics have advantages. They have wide VSWR bandwidth, and
are very easy to build. They are also a cheap way to get some gain, if
you ignore the price of real-estate.
Unfortunately the disadvantages outweigh the advantages in many cases.
You can't electically steer a single Rhombic, and they waste power in
ground losses and termination losses. They never have uniform current
because of losses from radiation and coupling to earth, and that
prevents them from ever reaching the gain theory predicts. They have
multiple lobes that waste power, and a very narrow beamwidth for the
amount of actual gain produced.
But then we weren't talking about 1920's technology, like Rhombics.
The question was about a curtain.
73 Tom
Siegfried Rambaum
> The curtains I`ve seen were (4) 1/2-wave dipoles, (1) in each quadrant
> in a plane perpendicular to the direction of transmission. All are
> driven in-phase. These are backed with an identical (4) 1/2-wave dipole
> parasitic array.
>
> A dipole has 0 dbd gain. (2) have 3 dbd gain. (4) have 6 dbd gain. (8)
> have 9dbd gain. A rhombic has 14 dbd gain.
Worded this way, the statement about Rhombics is nonsense. The gain of a
Rhombic hinges -- among other things -- on the length of wire used for each
"leg".
So, what wire length (expressed in multiples of wavelengths) is behind your
14 dBd ?
Richard Harrison
"Worded this way, the statement about Rhombics is nonsense."
Siegfried is ignoring earlier posts in this thread.
I wrote: A rhombic has 14 dbd gain.
This was a fair comparison for the rhombic we were using for
broadcasting to the same target area as the curtains. It could best the
curtains when both were doing their best to the center of the target
area at the middle of the antennas` operating frequency range. That
means a standard rhombic against a 9 & 11 MHz curtain. 14 dbd gain for
the rhombic. Fact is, a rhombic would beat any curtain at any frequency
we used for broadcasting in the center of the target area, but the
signal declined from the rhombic at the fringes of the target due to the
rhombic`s higher directivity.
At 9 & 11 MHz The rhombic is about 5 db better than the curtain. The
advantage is less at 6 & 7 MHz. and more at 15 & 17 MHz. This is one
rhombic versus three different curtains, each cut for its own operating
range.
At the frequency where the rhombic has 14 dbd gain, its sides have about
3.7 wavelengths per side. At the frequency where the curtain antenna has
about 9 dbd gain, its dipoles are 1/2-wavelength.
This is not nonsense. We actually made many fieldstrength measurements
of our curtains versus a reference dipole at the drivepoint height of
the curtains. These were made and recorded in and around the target over
many days and averaged. The rhombics have been measured many times and
are well documented.
The Signal Corps rhombic has a side length of 375 feet and a height of
65 feet. We had many of these kits and used them. We also designed our
own for specific frequency ranges, path lengths (elevation angles),
etc..
This thread was about curtains. I didn`t think it necessary to expound
on rhombics other than to point out that they are the real wide-band
antennas and that curtain antennas usually are not.
Tom W8JI
You keep saying curtains are narrow banded antennas.
I certainly agree they can be narrow banded.
However, the dozen or so installations using curtains I've seen were
not narrow band antennas. Just because we see one poor or limited
design example, it doesn't mean they are all built that way.
>means a standard rhombic against a 9 & 11 MHz curtain. 14 dbd gain for
>the rhombic.
Comparing a very small Curtain to a large Rhombic isn't indicative of
their performance when both are identical sized. A much smaller area
curtain with outdo a Rhombic hands-down. Rhombics and other long-wire
arrays are poor efficiency antennas, both for gain vs beamwidth and
space requirements.
They have their benefits, which are broad VSWR bandwidth and simple
construction, but those are the only benefits.
>Fact is, a rhombic would beat any curtain at any frequency
>we used for broadcasting in the center of the target area, but the
>signal declined from the rhombic at the fringes of the target due to the
>rhombic`s higher directivity.
Maybe at a Ma and Pa SWBC station, where the curtain wasn't optimized.
A 4 half-wave dipole curtain with a resonant parasitic reflector is a
modest antenna by any stretch of the imagination. Simply operating
that antenna at the second harmonic would have almost doubled the
power gain. If the parasitic reflector was replaced with a screen,
that antenna would have covered a 2.5:1 frequency range or larger.
Don't blame the antenna sysle when the design is marginal!
VOA, Radio Canada, Radio France, and many other stations easily get
more gain from their multiband curtains than from Rhombics occupying
many times the physical area.
>At 9 & 11 MHz The rhombic is about 5 db better than the curtain. The
>advantage is less at 6 & 7 MHz. and more at 15 & 17 MHz. This is one
>rhombic versus three different curtains, each cut for its own operating
>range.
Most stations use broadband curtains to save material and space. An
almost 3:1 frequency range is quite possible with a single curtain.
>At the frequency where the rhombic has 14 dbd gain, its sides have about
>3.7 wavelengths per side. At the frequency where the curtain antenna has
>about 9 dbd gain, its dipoles are 1/2-wavelength.
14 dBd is the lossless freespace gain of the Rhombic given in many
tables in textbooks. Factually, that gain can not be obtained.
Long wire arrays, even those spaced 1/3 of their leg length above
ground, have significant losses.
The losses not only reduce gain, they also cause the antenna to reach
gain limiting at about 3 or 4 wavelengths per leg. The reason for gain
limiting, and the inability to achieve the theoretical gain, is the
inability to force significant current out on remote areas of the wire
due to losses in the system (conductor loss, earth related losses, and
radiation).
That loss is well known and well documented. All you need to do is
measure the heat in a terminating resistor, and you'll see that much
less than half the transmitter power makes it to the far end of a
large Rhombic.
>This is not nonsense. We actually made many fieldstrength measurements
>of our curtains versus a reference dipole at the drivepoint height of
>the curtains. These were made and recorded in and around the target over
>many days and averaged. The rhombics have been measured many times and
>are well documented.
The Rhombic was as large as possible, the curtain was small. It is
one case where a small limited design curtain was compared to a large
Rhombic, and free-space lossless gain spec's were used for the
Rhombic.
Rhombics have a special place in many people's hearts, but they really
aren't an efficient antenna. They have high gain, but very poor gain
performance when the beamwidth is compared to other antennas with the
same beamwidth.
73 Tom
Richard Harrison
"---the dozen or so installations I`ve seen were not narrow band
antennas."
I`m curious. A description and performance numbers would be interesting.
Richard Harrison
"If the parasitic reflector was replaced with a screen, that antenna
would have covered a 2.5:1 frequency range or larger."
Maybe a bedspring array could be broadbanded by cutting different
dipoles to different frequencies.
Fact is, a single dipole close in front of a large solid reflecting
sheet can produce 5dbd gain, but has a 20% of frequency bandwidth.
Multiple identical elements narrow the bandwidth.
Tom W8JI
I'm sorry, that isn't correct Richard. With a distributed feed system
a single curtain can easily cover a 2.5:1 frequency range.
If you have any modelling programs, I can help you model one.
73 Tom
Tom W8JI
(Richard Harrison) wrote:
>Tom, W8JI wrote:
>"---the dozen or so installations I`ve seen were not narrow band
>antennas."
>
>I`m curious. A description and performance numbers would be interesting.
>
>Best regards, Richard Harrison, KB5WZI
I've described one example earlier in this thread.
If you want to see actual numbers or a real antenna, visit Radio
Canada or any VOA site. You can also see photos and examples from time
to time in SWL magazines.
73 Tom
Richard Harrison
"Any thoughts?"
I installed many curtain arrays in the then largest shortwave broadcast
station almost 50 years ago. These were state of the art and a good
choice for the job. They replaced rhombics which pin-pointed energy in a
too-small area and took-up too much land area.
Curtain antennas are well known to many amateurs, some of whom are
experienced in their commercial use.
Were curtain antennas economical and apt for amateur application, you
would have seen them often in ham publications.
Richard Harrison
"Simply operating that antenna at the second harmonic would have almost
doubled the power gain."
Where would the added power in the center of the target come from?
From narrowing the beamwidth of the signal and depriving other parts of
the target. Remember, gain and directivity are sides of the same coin.
The curtain antennas were selected to cover the entire target area. This
is precisely why the curtains were never used on harmonics.
Richard Harrison
"Heck, even a Hadley Fishbone TW is reported in Jasik to have more bang
for the buck per acre than a rhombic."
A lot more hams erect rhombics than erect curtains or fishbones. The
rhombic is more economical and practical if you have the space for it.
Tom W8JI
(Richard Harrison) wrote:
>The curtain antennas were selected to cover the entire target area. This
>is precisely why the curtains were never used on harmonics.
>
>Best regards, Richard Harrison, KB5WZI
The curtain would always have much more coverage area than a Rhombic
if gain from the two were equal.
That's because the Curtain has almost 100% efficiency, while the
Rhombic has less than 50% efficiency.
73 Tom
Fractenna
>is precisely why the curtains were never used on harmonics.
>
>Best regards, Richard Harrison, KB5WZI
>
>
Excellent point Richard. Hits the nail on the head.
73,
Chip N1IR
mike....@ziplog.com
Heck, even a Hadley Fishbone TW is reported in Jasik to have more bang
for the buck per acre than a rhombic!
GDRFC
--> Sleep well; OS2's still awake! ;)
Mike....@ziplog.com
Mike....@f3000.n117.z1.fidonet.org
Tom W8JI
(Richard Harrison) wrote:
>A lot more hams erect rhombics than erect curtains or fishbones. The
>rhombic is more economical and practical if you have the space for it.
>
>Best regards, Richard Harrison, KB5WZI
....and you have that statistic from where Richard?
I assume you included antennas like the old Cushcraft curtain for 2
meters, the Bobtail curtain, and so on?
Richard Harrison
"...And you have that statistic from where---?"
I wrote: A lot more hams erect rhombics than erect curtains or
fishbones.
Observation is the source of my statement. Statisticians call it the
"sampling technique".
Ask yourself, what have I seen?
Best regards, Richard Harrison, KN5WZI
Tom W8JI
(Richard Harrison) wrote:
>Ask yourself, what have I seen?
I did. The answer was Bobtails and a few of the Cushcraft antenna.
Fractenna
>fishbones.
>
>Observation is the source of my statement. Statisticians call it the
>"sampling technique".
>
>
>
That is true, based upon approximately 1,000 ham antennas I have seen over the
years.
I have seen about 30 rhombics; 2 curtains, and no fishbones. In fact, most hams
have never even heard of it, and documentation on it is not easily found in ham
publications since 1960 or so.
Richard's statement is an accurate, and not anecdotal, estimate of the relative
useage of the three antennas in ham-dom.
73,
Chip N1IR
Richard Harrison
"---you can see a commercial design (curtain array) in "Antenna
Engineering Handbook" by Jasik. Tom has ignored a request for a
description, facts, and figures from Jasik. Maybe someone else has
access to a copy?
Tom W8JI
This whole thing is beyond silly Richard. You asked where to see an
antenna, and I told you the book and chapter. I also described one,
and offered to help you model it. I also pointed out that VOA has data
on their curtains, it is a standard handout.
While I'd like to help you, I'm not about to type in a whole chapter
of a readily available book.
You seem to have some mindset based on seeing one antenna that
Rhombics are good antennas, and curtains are always narrow banded low
gain antenna. That simply isn't true.
For the same gain on a target area, a curtain is:
1.) Smaller than a Rhombic in physcial area
2.) Generally requires only two supports compared to four
3.) Covers a wider area of the target
You may not like it, but that's how it is.
73 Tom
Richard Harrison
"Just because we see one poor or limited design example, it doesn`t mean
they are all built that way."
Tom also wrote:
"Maybe at a Ma and Pa SWBC station, where the curtain wasn`t optimized."
This was in response to my statement: "A rhombic would beat any curtain
at any frequency in the center of the target area, but the signal
declined from the rhombic at the fringes of the target area due to the
rhombic`s higher directivity."
I`ve learned that the curtain antenna Tom chose as a paragon in Jasik
comes from chapters written by Stephen W. Kershner of A.D. Ring and
Associates.
Tom has mistaken the bandwidth of some of the parts of the antenna for
the bandwidth of the complete design.
Andy Ring was our antenna consultant and Steve Kershner designed all our
curtain antennas. None were poor or limited designs.
The broadcast plant was no Ma and Pa SWBC station but was the world`s
largest broadcast facility when the curtains were installed.
Tom just jumped to the wrong conclusions.
Bart Rowlett
>
>
> I`ve learned that the curtain antenna Tom chose as a paragon in Jasik
> comes from chapters written by Stephen W. Kershner of A.D. Ring and
> Associates.
A better description is in the second edition of _Antenna Engineering
Handbook_, (Johnson & Jasik, Mc Graw - Hill 1984) in the chapter (26) on
High Frequency Antennas. The chapter is written by Ronald Wilensky and
William Wharton (TCI inc) and while clearly featuring TCI's commercial
products, better reflects modern practice compared to the first edition.
In fact, Rhombic and other traveling wave antennas have been eliminated
from the chapter.
> Tom has mistaken the bandwidth of some of the parts of the antenna for
> the bandwidth of the complete design.
Section 26-7 is dedicated to the topic of "Broadband Dipole Curtain
Arrays" and is a decent, though not very technical, description of
modern curtain arrays suitable for HF broadcasting applications. There
is a brief discussion of the methods used for broadbanding through the
use of folded 'fat' radiators incorporating built in impedance
compensation. The chapter also discusses how wideband transmission line
transformers can be integrated into the inherently broadband corporate
feed system used to obtain an VSWR bandwidth (VSWR < 1.5:1) over an
octave or more.
Chapter 26 also makes reference to earlier curtain array designs which
used thin dipole elements, "very-narrow-bandwidth" feed systems and
parasitic reflectors. These earlier designs were only capable of
operating only in one or two broadcast bands.
> Andy Ring was our antenna consultant and Steve Kershner designed all our
> curtain antennas. None were poor or limited designs.
Sounds like considerable progress has been made since then.
> The broadcast plant was no Ma and Pa SWBC station but was the world`s
> largest broadcast facility when the curtains were installed.
> Tom just jumped to the wrong conclusions.
Seems to me Tom didn't _jump_ to any conclusions, but accurately
described current practice.
bart
wb6hqk
Fractenna
>"Just because we see one poor or limited design example, it doesn`t mean
>they are all built that way."
>
>Tom also wrote:
>"Maybe at a Ma and Pa SWBC station, where the curtain wasn`t optimized."
>This was in response to my statement: "A rhombic would beat any curtain
>at any frequency in the center of the target area, but the signal
>declined from the rhombic at the fringes of the target area due to the
>rhombic`s higher directivity."
>
>comes from chapters written by Stephen W. Kershner of A.D. Ring and
>Associates.
>
>the bandwidth of the complete design.
>
>curtain antennas. None were poor or limited designs.
>
>largest broadcast facility when the curtains were installed.
>
>Tom just jumped to the wrong conclusions.
>
>
Thanks for being an authoritative voice here Richard.
73,
Chip N1IR
Tom W8JI
>The broadcast plant was no Ma and Pa SWBC station but was the world`s
>largest broadcast facility when the curtains were installed.
>
>Tom just jumped to the wrong conclusions.
>
>Best regards, Richard Harrison, KB5WZI
Richard,
It's simple elementary antenna theory. Put what you have "learned" out
of you mind, and consider this.
If you take a simple dipole array, and feed it with a distribution
system that uses equal length lines from common points, all them
elements are fed in-phase on every frequency from DC to light.
There simply is no bandwidth problem in the feed system, that you seem
to be stuck on insisting exists.
The array is useful from frequencies just below those where the
dipoles are 1/2 wl long, to frequencies just above where the dipoles
are 1.3 wavelengths long. They are useful over even wider ranges if
you tolererate increased sidelobes.
If you desire a unidirectional pattern, you simply back the arrays
with a wire grid of parallel wires spaced .1 wl or more at the highest
frequency.
This really isn't that hard to comprehend, and the system is commonly
used at many SWBC stations. It would make quite a nice amateur
antenna, because the bandwidth would be from about where the elements
are .4 wl long to where they are about 1.3 wl long. 0.4 divided into
1.3 is a ratio of 3.25.
If there is any part of that simple basic system you can't understand,
let me know and I'll try to explain it in more detail.
73 Tom
Tom W8JI
<ba...@wb6hqk.ampr.org> wrote:
>Seems to me Tom didn't _jump_ to any conclusions, but accurately
>described current practice.
>
>bart
>wb6hqk
Thanks Bart,
This whole thing started when I suggested a USIA type curtain would
cover 10 meters and six meters. By now the original party is probably
long gone, with the incorrect information that it can't be done.
Oh well. I tried. Hopefully someone learned something about more
modern Curtain design than the old narrow band Sterba and Bruce arrays
from the 20's and 30's.
73 Tom
Richard Harrison
"A better description is in the second edition of _Antenna Engineering
That is not the antenna Tom was referring to.
Tom, W8JI wrote:
"I`m not aware of any amateur publication articles describing the
engineering behind these antennas, but you can see a commercial design
in the Antenna Engineering Handbook by Jasik. In my first edition copy
of that book---Chapter 21 describes engineering, section 21.6 shows an
actual USIA curtain."
Tom`s paragon was clearly not TCI`s product as describrd by Ronald
Wilensky and William Wharton in the second edition.
It is clearly too soon to close the patent office. Better antennas will
be built.
Tom jumped to the conclusion that an antenna he saw in a book, "(like
the USIA Curtain) will work perfectly fine with nearly 3:1 frequency
ranges".
Tom jumped to the conclusion that the antenna I described was a "poor or
limited design example".
Tom jumped to the conclusion that the curtain I described was "maybe at
a Ma and Pa SWBC station".
If Tom were describing current practice, why did he reference the first
edition of the Antenna Engineering Handbook which Bart implies is
obsolete?
Why didn`t Tom answer a request for facts and figures?
Richard Harrison
"You seem to have some mindset---"
No desire on my part to see one antenna perform better than any other.
I am curious about how one gets a resonant antenna element to respond
evenly and efficiently over a 3:1 frequency range and, even more
surprisingly to me, get an array of multiple identical elements to do
likewise.
I would never say impossible, only how is it done?
Fractenna
>limited design example".
>
>Tom jumped to the conclusion that the curtain I described was "maybe at
>a Ma and Pa SWBC station".
>
>If Tom were describing current practice, why did he reference the first
>edition of the Antenna Engineering Handbook which Bart implies is
>obsolete?
>
>Why didn`t Tom answer a request for facts and figures?
>
Tom Rauch is destructive Richard. He takes one's comments and digresses them to
something that seems like what was posted, but is not. Then he contentiously
pontificates on the self-righteousness of something, hoping you'll say
something in error. It really is counterproductive to the purpose of this NG
IMHO. It fails to enlighten and directly attempts to question the veracity and
motivation of the other party.
He has done it many times with many others on many NG's.
73,
Chip N1IR
Tom W8JI
(Richard Harrison) wrote:
>Burt Rowlett wrote:
>"A better description is in the second edition of _Antenna Engineering
>Handbook_."
>
>That is not the antenna Tom was referring to.
>
>Tom, W8JI wrote:
>"I`m not aware of any amateur publication articles describing the
>engineering behind these antennas, but you can see a commercial design
>in the Antenna Engineering Handbook by Jasik. In my first edition copy
>of that book---Chapter 21 describes engineering, section 21.6 shows an
>actual USIA curtain."
>
>Tom`s paragon was clearly not TCI`s product as describrd by Ronald
>Wilensky and William Wharton in the second edition.
>
>It is clearly too soon to close the patent office. Better antennas will
>be built.
>
>Tom jumped to the conclusion that an antenna he saw in a book, "(like
>the USIA Curtain) will work perfectly fine with nearly 3:1 frequency
>ranges".
>
>Tom jumped to the conclusion that the antenna I described was a "poor or
>limited design example".
>
>Tom jumped to the conclusion that the curtain I described was "maybe at
>a Ma and Pa SWBC station".
>
>If Tom were describing current practice, why did he reference the first
>edition of the Antenna Engineering Handbook which Bart implies is
>obsolete?
>
>Why didn`t Tom answer a request for facts and figures?
>
>Best regards, Richard Harrison, KB5WZI
How odd. I did respond, even with an offer to help you model the
antenna. You must not have received all the posts for some reason.
Tom W8JI
I don't know why you are having such a problem with a simple concept
and system like this, but I'll give it one last shot. There seem to be
two points you don't follow, as indicated by your statement below:
On Wed, 2 Aug 2000 00:58:58 -0500 (CDT), richard...@webtv.net
(Richard Harrison) wrote:
>I am curious about how one gets a resonant antenna element to respond
>evenly and efficiently over a 3:1 frequency range and, even more
>surprisingly to me, get an array of multiple identical elements to do
>likewise.
1.) Resonance has nothing to do with efficiency.
2.) When the elements are indentical size and phasing is either
in-phse or out-of-phase...it is very easy to make a totally frequency
independent phasing system.
Consider a open-wire feeder center fed antenna, like the double Zepp
antenna. At a lower frequency, it is a half-wave long. As you increase
frequency, the directional response remains the same except
directivity increases. As the length passes a full wavelength, the
antenna starts to develope minor lobes, but the main lobe gain
continues to increase. At 1.25 wl the broadside gain is nearly
maximum, and then beyond that length broadside gain starts to
decrease.
If you feed an array of dipole elements with a low-loss feeder in a
distribution type feed system, they have steadily increasing gain as
frequency is increased up to the point where the gating lobes
eventually rob too much poower from the main lobe. That happens on
some frequency above the frequency where the individual dipoles are
1.3 wl long.
If you decrease frequency, eventually matching system and feedline
losses cause a decrease in gain. That happens below the frequency
where each dipole is 1/2 wl long.
It is quite easy to have appreciable gain from a frequency where the
antenna elements are less than .4 WL long, to a frequency where they
are over 1.3 WL long. That is more than a 1:3 frequency range.
UHF TV antennas were commonly built this way in the 60's. Although
they sometimes didn't use a distribution feed system in each stack of
dipoles, they always used a distribution-type feed system in
combination broadside-collinear arrays of elements.
Some early VHF TV antennas used the same technology to cover 54 to 212
MHz in one array.
This actually isn't new technology, it is pretty old stuff. It just
isn't as old and ancient as Rhombics and the early narrow band
curtains, like Sterba's and Bruse Arrays.
>I would never say impossible, only how is it done?
You could have read this all yourself in the reference I gave, but the
above is a plain language explaination of how the system is designed.
Of course you simply back the system with a non-resonant reflector to
make it unidirection.
I can easily build a single Curtain antenna that covers a 3:1
frequency range with excellent gain and pattern at amateur power
levels just using conventional wire dipoles.
73 Tom
Klein Gilhousen
the ham) by considering some issues specifically related to the ham
application. In ham radio, there is a desire to get the most gain
posible for the least real estate and dollars together with the most
flexibility re: direction, ease of changing direction, frequency band,
etc. Anyway, I have some real estate (a number of acres) and I'm
planning to plant an antenna farm. I've been considering building a
set of steerable curtain arrays with each set of curtains being
pointed at a different sector. If each curtain array can be made to
be switchably unidirectional, then not as many sets of curtains are
needed. I.e., we switch one of the curtains from covering NE to SW by
reversing the phasing of the elements.
I just got a copy of EZNEC 3.0 to use to analyze various
configurations. (Thanks Roy for a great piece of software. Talk
about a serious contribution - wow!) One of the first things I
noticed that was puzzling (and still is) is that when you have two
parallel dipole elements that are .1 to .2 wavelengths apart and you
phase them to get maximum front to back ratio, the back dipole really
has a strange impedance, sometimes even becoming negative. Hmmmm!!??
I suppose that what's happening is coupling of power from one element
to the other and power is actually being absorbed by this element. Of
course, it has to go somewhere, like back into the feedline and
hopefully find its way back to the other element. Or what?
My question is how does all this affect the design of the feedline
system?
BTW, a "reasonable" physical configuration seems capable of
approaching 20dBi gain and be steerable over maybe +/-25 degrees.
Maybe more.
Klein Gilhousen, WT6G
Roy Lewallen
> . . .
> I just got a copy of EZNEC 3.0 to use to analyze various
> configurations. (Thanks Roy for a great piece of software. Talk
> about a serious contribution - wow!) One of the first things I
> noticed that was puzzling (and still is) is that when you have two
> parallel dipole elements that are .1 to .2 wavelengths apart and you
> phase them to get maximum front to back ratio, the back dipole really
> has a strange impedance, sometimes even becoming negative. Hmmmm!!??
>
> I suppose that what's happening is coupling of power from one element
> to the other and power is actually being absorbed by this element. Of
> course, it has to go somewhere, like back into the feedline and
> hopefully find its way back to the other element. Or what?
>
> My question is how does all this affect the design of the feedline
> system?
Your speculation is correct. In some arrays, one or more elements can
deliver power to the feed system. This power is obtained from the other
elements via mutual coupling. The feed systems I've designed have
largely been designed to deliver a specified current ratio regardless of
the feedpoint impedance, so these work as planned. But I'm not certain
what you'd do with an impedance-dependent design. You might just include
the negative resistance in the calculations and see what sort of network
results. I had a brief discussion about this with Forrest Gehrke, K2BT,
not too long ago, and maybe he can offer more.
This phenomenon is discussed briefly in the EZNEC 3.0 manual. See
"Resistance, negative" in the index, then choose the Source Data topic.
Roy Lewallen, W7EL
Klein Gilhousen
wrote:
Thanks for the tips Roy. After re-reading the EZNEC manual and also
perusing Chapter 8 of the ARRL Antenna Book, I'm now convinced that a
parasitic element for directional steering is the way to go. I'll try
using relays at the drive points to switch in a short to select
direction. Seems a lot simpler than trying to make the feed system do
it.
The remainder of the design will be a number of side by side sets of
two such elements. I think the mutual coupling is a lot smaller in
this case and probably a lot easier to deal with. The techniques
given in the Antenna Book seem to be up to handling this. Or maybe
I'll just buy a big yagi and put it up high. (Naaaah!)
Klein, WT6G
Yuri Blanarovich
>
>How odd. I did respond, even with an offer to help you model the
>antenna. You must not have received all the posts for some reason.
>
>
>
Thanks for sheding some light on the curtains Tom. I saw some awesome Sterba
curtains at W2KW (3 towers/curtains) and played with Rhombics. I like Rhombics
for the ease of hanging on strategically located trees (and they work!) But
USIA type curtains deserve some consideration, especially on higher bands.
Just starting to play with excellent W7EL EznecW (thanks Roy!!!) and looking
forward to some new killer antenna designs.
73 Yuri, K3BU
Tom W8JI
>Thanks for sheding some light on the curtains Tom. I saw some awesome Sterba
>curtains at W2KW (3 towers/curtains) and played with Rhombics. I like Rhombics
>for the ease of hanging on strategically located trees (and they work!) But
>USIA type curtains deserve some consideration, especially on higher bands.
>
>Just starting to play with excellent W7EL EznecW (thanks Roy!!!) and looking
>forward to some new killer antenna designs.
>
>73 Yuri, K3BU
I was amazed to see Chip's claim he has seen around 30 Rhombics at
amateur stations, or thirty stations using Rhombics Yuri!
I've been racking my brain, and can only recall two or three people
using Rhombics. Lot's of V beams, and even more curtains of various
types (most popular are the "Bobtail curtains" and Sterbas).
I'm puzzled why contesters don't use broadband curtains, since it
would be easy to cover at least three bands with one curtain and gain
could be exceptional on at least one band and very good on others.
I'd be happy to work with anyone who wants to design a curtain and
write an article. I think it would be a very good thing for out
community to learn how to do that, instead of following most articles
that describe narrow-band antennas like Sterba's ( that only have a
few percent bandwidth).
73 Tom
Fractenna
>write an article. I think it would be a very good thing for out
>community to learn how to do that, instead of following most articles
>that describe narrow-band antennas like Sterba's ( that only have a
>few percent bandwidth).
>
>73 Tom
But Chas Thomas W8JI ! You don't WRITE articles. You write LETTERS! And when
you READ articles you call yourself a REVIEWER of ARTICLES HAHAHAHAH!
Why would anyone believe anything you say? What have you ever said on antennas
that wasn't out of someone oelse's book or article?
Hams needing curtains! Be sure to feed it with latter line ! Don't say anthing
that was new after 1950!
Watta joke! Why don't you show'em how to make a Tesla coil for 160 metres! At
least that would be something CLOSe to new.
Tom writing 'articles'. HAHAHAH!!! Better get the boss's ok!
And the BEst part-- WRITING IT WITH YURI!!! AHAHAHAHAHAHAHAHAHH! You guys
DESERVE each other! Maybe make April QST ish!
Phil
de N1ZKT
Yuri Blanarovich
>I was amazed to see Chip's claim he has seen around 30 Rhombics at
>amateur stations, or thirty stations using Rhombics Yuri!
>
I am not amazed at anything Freakbunch writes, I am amazed that people still
"argue" with him.
>I've been racking my brain, and can only recall two or three people
>using Rhombics. Lot's of V beams, and even more curtains of various
>types (most popular are the "Bobtail curtains" and Sterbas).
>
Same here, most notable W6AM farm, ZD8Z using commercial site and bunch of
individuals using one here and there.
>I'm puzzled why contesters don't use broadband curtains, since it
>would be easy to cover at least three bands with one curtain and gain
>could be exceptional on at least one band and very good on others.
>
I think the subject was overlooked to some part, part limited "turnability",
need for high towers. I can see putting up mother curtain for directions like
Eu for heavy contest population. All it would take for someone to cream some
records using curtains and the crowds will follow (if it works).
>I'd be happy to work with anyone who wants to design a curtain and
>write an article. I think it would be a very good thing for out
>community to learn how to do that, instead of following most articles
>that describe narrow-band antennas like Sterba's ( that only have a
>few percent bandwidth).
>
I would like to look into vertical curtains on the beach. I am mostly monoband
at the time, due to various limitations, but will do some investigating. Just
starting with EZNEC and trying to correlate the results with reality.
>73 Tom
>
Yuri
Fractenna
I remind hams, in general, that is was the frustration of curtain antennas that
motivated W8JK to devise the '8JK antenna'. Hams abandoned curtains in droves
for the smaller, hugh gain 8JK.
Set "DX records" with curtains and you will have the entire ham community
wondering when spark will make a comeback.
Perhaps it's better if you obsess on this. Go for it. Leave the innovation, in
good hands, to others.
73!
Chip N1IR
Tom W8JI
>I remind hams, in general, that is was the frustration of curtain antennas that
>motivated W8JK to devise the '8JK antenna'. Hams abandoned curtains in droves
>for the smaller, hugh gain 8JK.
With two elements, end-fire is always the way to go. Gain is always
higest in the end-fire configuration.
When some number more than two elements are used, broadside-endfire
combinations result in the highest gain for amount of material.
>Set "DX records" with curtains and you will have the entire ham community
>wondering when spark will make a comeback.
Only a person with little actual experience, or a poor understanding
of antennas would equate a broadside-endfire combo as an ancient
design. Most modern SWBC stations of any caliber use curtains. Very
few use Rhombics, except for small low-budget operations.
73 Tom
Cortland Richmond
I've been privileged to use the rhombics at one military post, and the results
could be awesome on the right band, in the right direction. Never tried a curtain,
but I did work one station who WAS using a curtain array on 15 - it was due to be
shipped to its end user in Abu Dhabi, and a DL who worked at the place it was made
brought his rig in; he had the ONLY signal on 15, that afternoon.
However... I wonder if a curtain is not inferior in cost versus gain, ease of
aiming, and utilization of antenna volume over the ground to a Yagi, or stacked
Yagi's? Doesn't a Yagi meet Kraus's condition for "supergain," showing more gain
than the effective area of the elements occupying the same space?
Cortland
(ka...@saber.net)
Tom W8JI wrote:
> On 03 Aug 2000 12:53:05 GMT, k3...@aol.com (Yuri Blanarovich) wrote:
>
> >Thanks for sheding some light on the curtains Tom. I saw some awesome Sterba
> >curtains at W2KW (3 towers/curtains) and played with Rhombics. I like Rhombics
> >for the ease of hanging on strategically located trees (and they work!) But
> >USIA type curtains deserve some consideration, especially on higher bands.
> >
> >Just starting to play with excellent W7EL EznecW (thanks Roy!!!) and looking
> >forward to some new killer antenna designs.
> >
> >73 Yuri, K3BU
>
> I was amazed to see Chip's claim he has seen around 30 Rhombics at
> amateur stations, or thirty stations using Rhombics Yuri!
>
> I've been racking my brain, and can only recall two or three people
> using Rhombics. Lot's of V beams, and even more curtains of various
> types (most popular are the "Bobtail curtains" and Sterbas).
>
> I'm puzzled why contesters don't use broadband curtains, since it
> would be easy to cover at least three bands with one curtain and gain
> could be exceptional on at least one band and very good on others.
>
> I'd be happy to work with anyone who wants to design a curtain and
> write an article. I think it would be a very good thing for out
> community to learn how to do that, instead of following most articles
> that describe narrow-band antennas like Sterba's ( that only have a
> few percent bandwidth).
>
> 73 Tom
Klein Gilhousen
are going to use the antenna. If your plan is to communicate only
along azimuth 47degrees +/-5 and you don't care about any other
direction, then maybe the rhombic is the best for you. However, if
you are a contester or DXer and want to have that high gain able to
squirt your signal in EVERY direction (well, one direction at a time),
then maybe the rhombic is not the right answer. Consider that curtain
arrays (some of 'em) can easily be steered +/-25 degrees or so around
the center of the beam with no loss in gain. Also consider that they
can be designed to be reversible. So a single structure can provide
very high gain over about 100 degrees of azimuth. With three or four
you can cover the world. This probably takes a lot less real estate
than the equivalent azimuthal coverage with rhombics would take.
Klein Gilhousen, WT6G
Tom W8JI
the land is "free".
Rhombics do offer less gain performance for a given beamwidth but they
have a low VSWR bandwidth. In amateur installations, we could use a
small cheap LTU (and we won't have feedline corona problems at our
power levels) so VSWR bandwidth is a non-issue.
73 Tom
Peter Bertini
> >>> I'd be happy to work with anyone who wants to design a curtain and
> >>> write an article. I think it would be a very good thing for out
> >>> community to learn how to do that, instead of following most articles
> >>> that describe narrow-band antennas like Sterba's ( that only have a
> >>> few percent bandwidth).
> >>>
> >>> 73 Tom
> >
Tom
I would be very happy to review an article on curtain antennas for the Tech
Notes column
in QEX written by you. I think it is something that would be immensely
popular and pracitical.
Please drop me a line if you are interested in doing an article, and we will
work out the details!
If you don't have the time, perhaps you and Rick could collaberate on the
project?
Always a pleasure reading your lucid explanations in these forums.
73
Peter K1ZJH
Editor, Tech Notes, QEX/Communications Quarterly Magazine
>
TyllEulenspiegel
TyllEulenspiegel
evil step-sister tomasina wears the glass slipper while chiperella scrubs the
fireplace
>> >>>
>> >>> I'd be happy to work with anyone who wants to design a curtain and
>> >>> write an article. I think it would be a very good thing for out
>> >>> community to learn how to do that, instead of following most articles
>> >>> that describe narrow-band antennas like Sterba's ( that only have a
>> >>> few percent bandwidth).
>> >>>
>> >>> 73 Tom
>> >
>