Steve
kb8viv
See URL:
http://www.n2aqs.com/facts.html
and URL:
http://www.n2aqs.com/
More can be found by typing "Vertical Dipole" into google
http://www.google.com/
--
73 From The Wilderness Keyboard
"Desmoface" <desm...@aol.com> wrote in message
news:20040108135102...@mb-m16.aol.com...
Yeah, exactly what i was thinking of building...was just gonna string wire up a
tree...will probably have to run the coax parallell to the lower half of the
antenna though...wonder what effect that will have?? Thanks again for the
info..
73's de kb8viv
steve
"Desmoface" <desm...@aol.com> wrote in message
news:20040108135102...@mb-m16.aol.com...
Hi Steve,
There are two ways to accomplish this.
1.) Simply run a length of coax directly up. Strip back the shield
to expose a quarterwave upper element and in the stripping back of the
shield, actually roll it back over the jacket below it to serve as the
lower element.
2.) Use a quarterwave length of tube as the lower half, run the coax
up its interior, connect the shield to it at the top of the tube, and
continue the inner coax up another quarterwave.
73's
Richard Clark, KB7QHC
steve
This sounds like the simplest way. He could use an 8 foot piece of 1/2 in
Al tubing for the lower element.
I would put the thing as high as you can. For locals, you *want* a low
radiation angle.
Tam/WB2TT
How about a J-Pole? Low angle radiation. The bottom end can be connected to
earth ground. In fact, I constructed one using a mast with the 1/4 wave
parallel section added 1/2 wave below the top. Real easy. No radials.
Just a thot.
John - KD5YI
thats also an option..i wonder if'n I'll need radials with a vertical dipole??
Steve
kb8viv
Hi Steve,
The lower dipole element IS(ARE) the radial(s).
You can follow my former instructions and simply make the upper
portion 5/8ths wavelength (if you have the room). If this makes it
too long (too close to ground) you will need radials. But for 10M
that is getting a little carried away if you can hoist a vertical
dipole.
> How about a J-Pole? ..one using a mast with the 1/4 wave
parallel section added 1/2 wave below the top. Real easy. No radials.
> John Smith - KD5YI
As others have already noted, vertical dipoles do work
whether center-fed or end-fed (a.k.a. J-pole) without radials.
As already noted, hang it as high as you can manage
- just the normal advice for any antenna except N.V.I.S.
If you want to pursue the J-pole route,
go to CQ magazine's web site
http://www.cq-amateur-radio.com/
and download George VE3ERP Murphy's freeware HAMCALC
ham radio software package. (lower left corner)
(Download it just for general principles.)
Program about-23 of 250+ programs
is Murph's version of my version of Gary N3GO O'Neill's
J-pole program from his Communication
Quarterly article on end-fed half-wave antennas.
Go to Steve AA5TB Yate's website
under "Antennas" for more info and a link
to Gary's article on how J-poles
and Zepps really work versus the popular view.
The program gives starting lengths.
For 28.5 MHz:
Using 300 ohm twinlead:
halfwave radiator wire = 196.7 inches
series matching 300 ohm section = 78.6 inches
to the 50 ohm tap point
shorted shunt stub 300 ohm section = 6.6 inches
Total length is about 23.5 feet if you hang it all up
vertically. The not-quite-quarter-wave matching section
can be horizontal with the halfwave radiator vertical
if you need/want to run it that way.
Gary's article, the program or a Smith chart show
why the commonly used 450 ohm ladder line is not
the as good a choice for matching as 300 ohm.
Have fun. Measure and cut metal and plastic,
hang it up, apply and withdraw RF energy.
Don't just do computer calculations
(as I do most of the time (sigh)).
Cheers, 73,
Ron McConnell
N 40º 46' 57.9" W 74º 41' 21.9"
FN20ps77GU46 [FN20ps77GV75]
http://home.earthlink.net/~rcmcc
As others have replied, no you don't.
I once made a 1/4 wave vertical and was listening to a local conversation on
my IC-2AT when there occured a lightning stroke from a thunderstorm a few
miles away. Although the mast and radials were grounded, the conversation
disappeared for about 5 seconds and then faded back in. Electrical charge on
the ungrounded vertical element. But I have never had this problem with a
J-Pole since all parts of it can be at earth ground. I've also used folded
vertical monopoles with the same results.
In my old age I have probably become paranoid about ungrounded driven
antenna elements. I know that some (maybe all) commercial antennas are
designed with networks or other devices which does the grounding. If I were
to build a simple ground plane today, I would make it a bit shorter than
resonant and add a shunt inductor across the feed point. That way there can
be no static accumulation on the element.
By the way, we sometimes get sand/dust storms here. I have seen and heard
the repeated discharges from the center pin to the shell of a PL-259 on a
RG-8 line coming down from the roof during a sand storm.
Anyway, this is not to convince anyone that a J-Pole is a cure-all or even
the best choice. I just wanted to tell about what drives my choices and tell
about a couple of my experiences. Anything you choose will be just fine, as
long as you are aware of its properties.
Have fun.
John - KD5YI
Good thinking. I worked for an oil and gas company which was assigned a
33 MHz frequency shared with a few others in the same industry but
available to us almost at any location within the U.S.A. We made good
use of the frequency for land mobile operations with dozens of fixed
stations and hundreds of mobile units. It worked very well for the radio
"line-of-sight. Usual FM base station power was 500 watts. The mobiles
were 50 watt units. Higher base station power is justified by noiser
receiving conditions in a mobile.
In our international operations we had many HF installations. Some were
AM and some were SSB.
I made a trip to Bolivia to relocate a HF station from a site which was
being seized by the government to accomodate homeless rural people who
were descending upon La Paz to seek their fortunes. Relocation of the
station was easier said than done but was ultimately successful. We had
offices in La Paz and in Cochabamba and we were drilling in the Chaco
Jungle.
While in Bolivia, I got a request from Argentina. They wanted an
aircraft beacon installed on Tierra del Fuego, local land mobiles that
worked, and radio between Tierra del Fuego and Buenos Aires. Strikes
frequently interrupted normal communications. The Peronistas wanted Juan
Peron back and the strikes were part of their agitation. So, I went to
Argentina.
In Argentina, we had RCA and RCA Radiomarine SSB equipment. Our aircraft
were equipped with Collins equipment which was dependale and our pilots
knew how to use it.
The RCA SSB was a problem. The automobile units had plastic coil forms
which melted in normal service. This was far removed from the equator.
It was near the antarctic and damn cold.
Operations were a problem. Base stations were used to communicate
between land bases, vehicles, boats, and tankers taking on oil we had
found and were producing. Multiple crystals meant operators often
couldn`t communicate because they were often switched to the wrong
channel. They also were mystified by knobs identified as "speech
clarifiers".
Fortunately we had some 5-watt Motorola Handitalkies on our stateside 33
MHz FM frequency, on loan to our Argentine operation.
After I installed the low-frequency aircraft homing beacon, the next
problem was getting reliable communications between our main bases in
Rio Grande and San Sebastian, almost 40 miles apart.
This is a land where hurricane force winds blow nearly every day. The
wind is so prevalent it is relied upon for aircraft operation.
Commercial air service to Rio Grande was cancelled when the wind was
calm as the runways were too short for take-offs and landings in the
calm. Commercial flights used DC-6`s (Aerolineas) and C-46`s (Austral).
Our company flew a DC-3 back and forth between Buenos Aires and Tierra
del Fuego. We had Beechcrafts on the Island.
For line-of-sight we needed elevation for 33 MHz antennas. We had
welders, line pipe, steel handbooks and I had a slide rule. So, I went
to work and produced guyed towers at both ends of the path. They were a
little over 100 feet and I put vertical homemade resonant vertical
centerfed dipoles for 33 MHz at their tops. These were connected by RG-8
with the 5-watt radios in the offices. Voila! We were in business with
full quieting. No bandswitching. No clarifiers. No melted coils in the
Motorolas.
We had a second office in Rio Grande but it didn`t need to talk to San
Sebastian. So I made a short self-supporting line pipe tower for that
office. It was about 30 feet. It too had a vertical dipole and a 5-watt
Motorola Handitalkie. When I made a call on that radio, an oil field
supply company in Midland Texas who shared the frequency came back to
me! We`re talking real DX and it was solid. As has been said, that`s the
way it is when the band is open.
Let me correct a statement in this thread that the loweer dipole element
is a radial. It is not. It is an axial and it radiates. Radials are
supposed to be balanced so that they do not radiate.
It is true that radials provide a 2nd antenna connection as does the 2nd
dipole element. That is where the similarity ends.
For the radio connection with Buenos Aires, the public correspondence
station, Radio Pacheco, seemed to work during telephone strikes. So we
went to work. It operated on certain HF channels. Al Hopson, our chief
pilot in Argentina was also a ham and he had had recognized a
Hallicrafters HT-20 that Glen McCarthy had left behind in the Chaco
Jungle of Bolivia when we bought his consession. Al put the HT-20 away
for safe keeping and asked me if I wanted it. He fetched it for me. I
went to Buenos Aires and visited Radio Boliche, "Barato y chi chi".
While there I bought a 2nd-hand Eddystone receiver that worked good and
was cheap (barato y chi chi just as their slogan said).
Back on the island again, it was out with the slide rule and my precious
copy of Ed Laport`s "Radio Antenna Engineering". When we gave Buenos
Aires a call with that HT-20 connected to that rhombic, we really
rattled their cans. The were very slow to believe that we were so far
away.
We ordered the FM stuff they needed to replace all the SSB crap, and
left them delighted with the improvements already made.
Yes. Centerfed vertical dipoles work fine without radials. They have
nulls at their tip ends which reduce mutual impedance with stuff in
those directions. Centerfed antennas are mostly independent of the earth
at close range when low-angle radiation is considered.
Best regards, Richard Harrison, KB5WZI
WOW!! That sounds like the makings of an Indiana Jones story line hehe..Thanks
for the entertaining and informative post..73's de kb8viv..
steve
Sorry I left out the word "vertical". Certainly a low horizontal dipole
is not independent of earth, but is closely coupled to the earth.
Just picture a 1/2 dipole horizontally -- then rotate it 90 degrees (or
stand it on end) -- same antenna -- different polarization.
So this picture should clarify why "No Radials".
The lower end is the other half of the dipole - it radiates in conjunction
with the upper half.
Others have advised you about the radiation from "Radials"
"Keyboard In The Wilderness" <KI...@zilch.net> wrote in message
news:PCzLb.55730$m83.35174@fed1read01...
I don't know where you get this from. I just ran EZNEC on a 10 m vertical
dipole with the following results:
A) Bottom at 8 feet, gain at 5 degrees is -2.44 dbi.
B) Bottom at 42 feet, gain at 5 degrees is +2.75 dbi.
If you want to work locals, the greater height will also help to get clear
local obstructions, like small hills
Tam/WB2TT
"Jim Leder" <k8...@hotmail.com> wrote in message
news:3ffed4df$0$88800$a04e...@nnrp.fuse.net...
He is better off with the sleeve antenna, because he can't run the feedline
off at a right angle.
Tam/WB2TT
Dale W4OP
A "dipole" is called a dipole because it has two "halves" or "poles" driven
in the center. If we want to get a little more descriptive what we are
talking about is a "center fed half wave antenna". This is the basic
antenna. (the gran-daddy of all antennas, but I won't entertain arguments
here about others including 1/4 wave ground planes) Just saying "dipole"
could also (much less common) mean something other than the half wave
length. When hams say "dipole" you can bet they're talkin' the half wave
version.
Turn it any way you want, it's the same thing, just keep the ends as far
from things as possible. Half wave antennas like this need no ground plane,
also called radials, when so constructed.
If you "run the coax feed-line out through the center of one of the halves"
you get the sleeve dipole. Since the coax runs right past the one end,
there's a lot of capacitance and you have to adjust the length of that half
a bit (shorter) for best SWR. In this case it may not be a true half wave
long in total, but that doesn't matter. Anything close to a half wave, when
it is getting power into it, will radiate like the full half wave. This can
take the form of the coax braid pulled back or a tube or pipe.
Going just a bit further, if you imagine spreading the very ends of the
braid out to form a big a cone, so it gets wider at the end away from the
center of the half wave - and keep going until it is all 90 degrees from the
coax (and the other half (1/4 wave)) then you have a 1/4 wave ground plane
antenna.
Then there's the "J antenna" now-a-days always called the "J-Pole" (except
by me)
P.S. The original "J-Pole" Was, and I believe still is, a commercial antenna
by, I believe Cushman, consisting of either 2 or, I believe 4, J's spaced
vertically on a supporting "POLE" for increased omni directional gain. Thus
the name "J-POLE".
Getting descriptive again, this is an "end fed half wave antenna". This is
the same radiating element, a half wave, just fed differently. To get the
high impedance feed required at the end of a half-wave, what looks a lot
like a shorted quarter wave stub matching section is used. This beast can
also be placed vertically or horizontally.
And by the way, there is no reason that the "1/4 wave" matching section has
to be in-line with the radiating half wave. It can be at 90 degrees if
necessary. Just keep in mind that the feed-point end is still at a high
impedance and should be kept away from other things to minimize the
capacitance and resulting de-tuning -- just like any end of any half wave
antenna. (I won't entertain arguments of whether the matching section is
really a true 1/4 wave or not)
Now the thing which triggered my comments, that of Richard , KB7QHC
>> You can follow my former instructions and simply make the upper
>> portion 5/8ths wavelength (if you have the room). If this makes it
>> too long (too close to ground) you will need radials.
I believe he was originally describing a sleeve dipole. The upper part
could indeed be 5/8 wave, HOWEVER that radiator has significant reactance,
capacitive. The 5/8 antenna is usually done as a ground plane and you will
see a "loading coil" at the bottom. It turns out that the 5/8 radiator has
a close to 50 ohm real radiation resistance, but with capacitance. In the
series representation it is 5 - Jx. I don't know the value (I have wondered
what it is for years-anybody model it & let me know??), but it can be tuned
out with a series inductor and nothing more. A good design, however, will
use a coil to ground with the coax feed tapped up the coil. This is to get
the radiating element grounded, but the net match is the same.
If I'm not mistaken, the original "Extended Zepp" or was it just called the
"Zep" or Zepplen",was a "di-pole" made up of two 5/8 elements on either
side of center. Should have had a rather thin radiating pattern broadside
and, yes some gain, like the 5/8 vertical does to obtain its "gain".
Should have perhaps had two smaller lobes something like 50 degrees of the
centerline (also like the 5/8)
Zat help any?
Whew!
Steve K;9;D:C:I
Hi Steve,
I have built such antennas for 10M and the reactance was quite
insignificant when swamped by the proximity of ground (being a foot
below the lower element which was actually a tower section insulated
above ground). The upper element was a 21 foot section of 1inch
tubing.
You could base feed it with a gamma loop type matching device. For
10m, the single turn loop would be appx 9-11 inches dia. The cap can
be an appx 16 inch length of coax. You need appx 50 pf or so to match
for 10m. I've built many of these for the higher HF bands. They work
well, and are easier to tune than it sounds. The cushcraft AR-10 is a
commercial version of this antenna, using that type matching device.
You can get a PDF manual on their website. Heck, you might just want
to buy a AR-10 and be done with it...:) They are fairly inexpensive.
They are also easy to butcher up and make other antennas from. I've
modified many of those for other uses. MK
Robin