Open Wire with G5RV =RFI
Ron
antenna. I want to operate 10-15-17 and 20 meters also. I took a look at
the G5RV and want to homebrew one. By using the open wire feed line am I
more susceptible to RF in the shack and RFI?
I live in a neighborhood where the homes are very close together and my
antenna would be only be 30 feet from my neighbor. I don't want to have
problems with the neighbors.
W5DXP
If you are careful how you run the open-wire line, you shouldn't have any
more feedline radiation (RFI) than you do with coax. My advice is just to
feed your 40m inv-V with 450 ohm window-line. That would be the easiest
way for you to get on 10m-30m. A 66' window-line fed dipole is described
on my web page. Covers all HF bands except 75m.
--
73, Cecil http://www.qsl.net/w5dxp
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 80,000 Newsgroups - 16 Different Servers! =-----
Richard Fry
the neighbors much be affected by the antenna feed line?
That is, if the natural impedance of the transmission line matches the
antenna input impedance, radiation from the line will be an insignificant
part of the total radiation from the antenna system.
RF
Visit http://rfry.org for FM broadcast RF system papers
_____________________________
"W5DXP" <Cecil....@ieee.org> wrote in message
news:3EF4ADAD...@ieee.org...
Roy Lewallen
radiate. A perfectly matched line can radiate badly, and a badly matched
line can radiate very little. In the sort of multiband antenna the
original poster described, the line will inevitably be very mismatched
on some, and probably all, bands. This doesn't at all cause line radiation.
What matters is the balance of current between the two feedline
conductors. If the currents are equal in magnitude and out of phase, the
line will radiate very little. The reason is that the field from one
conductor is equal to and opposite from the field from the other. If
they're not balanced like that, the line will radiate. Although a balun
can help, it's difficult to construct a balun which will work well on a
single non-resonant antenna, because of the large range of impedances
that might be encountered. The most reliable way to insure current
balance on that sort of setup is to use a balanced tuner. Although the
antenna is symmetrical, non-symmetrical feeding systems can cause line
imbalance.
A measure of the imbalance is the common mode current. One definition of
common mode current(*) is Icm = 1/2(I1 + I2) where I1 and I2 are the
(complex) values of currents in the two conductors, with positive being
defined in the same direction for both. If I1 and I2 are perfectly
balanced, I2 = -I1 and the common mode current will be zero. If common
mode current is present, the feedline will radiate just as though there
were Icm flowing on each wire of the feedline in phase, or 2*Icm flowing
on a single wire.
(*) Common mode current is sometimes defined without the 1/2 factor, in
which case line radiation is the same as it would be with Icm (by this
definition) flowing on a single wire.
Roy Lewallen, W7EL
W5DXP
> That is, if the natural impedance of the transmission line matches the
> antenna input impedance, radiation from the line will be an insignificant
> part of the total radiation from the antenna system.
A balanced antenna fed with unbalanced feedline is technically
not "matched". :-)
Richard Fry
assuming that the source Z matches the line Z?
RF
_________
Roy Lewallen" <w7...@eznec.com> wrote in message
news:vf9irf8...@corp.supernews.com...
Richard Fry
ohm, unbalanced, coaxial transmission line to drive vertically collinear
arrays of dipole radiators.
RF
_____________
"W5DXP" <Cecil....@ieee.org> wrote in message
news:3EF4D18F...@ieee.org...
> Richard Fry wrote:
> > That is, if the natural impedance of the transmission line matches the
> > antenna input impedance, radiation from the line will be an
insignificant
> > part of the total radiation from the antenna system.
>
> A balanced antenna fed with unbalanced feedline is technically
> not "matched". :-)
> --
> 73, Cecil http://www.qsl.net/w5dxpServers!
Roy Lewallen
matching network at the base, will present a bad impedance match to the
transmission line. The line will have a high SWR. The line won't radiate.
A resonant dipole with a 75 ohm input impedance fed with 75 ohm coax
will be well matched. If the distance along the outside of the feedline
and other conductive paths to ground (e.g., a rig's connection to
ground) is approximately a half wavelength, the line will radiate
considerably.
The source match has no effect on either the transmission line's SWR or
its current balance (that is, how much it will radiate). However, the
balance of the connection at the source does imact the current balance.
For example, if you feed a dipole with an open wire line, both the line
and the dipole are symmetrical. But if you plug one conductor of the
line into the center of your rig's coax output connector and connect the
other coductor to the rig's chassis, the line will sho' 'nuff radiate.
Neither the match at the load end nor the match at the source end have
anything to do with whether or not the transmission line radiates. It's
a myth that dies hard.
Roy Lewallen, W7EL
Roy Lewallen
done in a way that prevents current from flowing on the outside of the
coax. That is, the feed system results in balanced conductor currents.
Not everyone agrees with Cecil's interpretation of "matched", but I have
no intention of getting into an argument with him about definitions.
When unbalanced currents exist on a transmission line, the problem is
usually separated into two parts, an "odd mode" and an "even mode", also
known as "differential mode" and "common mode". (There are additional
modes if more than two conductors are involved.) The odd, differential,
or "transmission line" currents represent the purely differential
component of the currents. In the case of coax, these are inside the
shield. Normal transmission line analysis can be done with this "mode"
independently of the even mode conditions. And the line can be perfectly
or imperfectly matched with regard to this mode. The even mode, common
mode, or "antenna" currents are analyzed separately. They, and they
alone cause radiation, and can be analyzed as part of the antenna,
rather than part of the idealized transmission line. In the case of
coax, the common mode current is conveniently physically separate from
the differential mode current -- it's on the outside of the shield.
So the transmission line can be matched, by virtue of a proper match for
the differential mode currents, while still having a common mode current
which causes radiation.
Roy Lewallen, W7EL
Richard Fry
BUT, this is the EXACT configuration chosen by the designers of FM & TV
broadcast transmitters and antenna systems for the last 40+ years. Have
they been wrong all this time? Is this a just a perspective of a "ham" vs
commercial approach? Or does the same physics apply, regardless.... ?
RF
----- Original Message -----
From: "Roy Lewallen" <w7...@eznec.com>
> For example, if you feed a dipole with an open wire line, both the line
> and the dipole are symmetrical. But if you plug one conductor of the
> line into the center of your rig's coax output connector and connect the
> other coductor(sic) to the rig's chassis, the line will sho' 'nuff
radiate.
Richard Fry
Are not the currents on the inner and outer conductors of the coax feedlines
for these systems everywhere equal for FM/TV dipole arrays when the Z of the
feedline equals the Z of the dipole array?
This system configuration has been used by the designers of commercial FM
and TV RF system designers for 40+ years, and seems to have well-proven by
theory and practice. Do you have evidence to the contrary?
RF
Visit http://rfry.org for FM broadcast RF system papers.
__________________
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:vf9o0ij...@corp.supernews.com...
> If you'll take a good look at how these are fed, you'll find that it's
> done in a way that prevents current from flowing on the outside of the
> coax. That is, the feed system results in balanced conductor currents.
---major clippage---
>so the transmission line can be matched, by virtue of a proper match for
W5DXP
> Virtually ALL commercial FM broadcast antennas are designed for, and use 50
> ohm, unbalanced, coaxial transmission line to drive vertically collinear
> arrays of dipole radiators.
My Comet 2x4Max FM transmitting antenna is vertical collinear arrays of
dipole radiators but the feedpoint is not balanced. Feeding a balanced
antenna with an unbalanced feedline is a good way to cause common-mode
currents and feedline radiation.
W5DXP
> Not everyone agrees with Cecil's interpretation of "matched", ...
Since I put it in parentheses, it can mean almost anything. :-)
--
W5DXP
> This system configuration has been used by the designers of commercial FM
> and TV RF system designers for 40+ years, and seems to have well-proven by
> theory and practice. Do you have evidence to the contrary?
Do those antenna systems have a ground plane?
Roy Lewallen
definitions.
Roy Lewallen, W7EL
Richard Fry
Assume a single 1/2-wave, resonant dipole fed by 73 ohm coax. What amount
of the total radiated field is attributable to radiation from the coax?
RF
Visit http://rfry.org for FM broadcast RF system papers.
____________________
"W5DXP" <Cecil....@ieee.org> wrote in message
news:3EF4EFBD...@ieee.org...
> Richard Fry wrote:
> > Virtually ALL commercial FM broadcast antennas are designed for, and use
50
> > ohm, unbalanced, coaxial transmission line to drive vertically collinear
> > arrays of dipole radiators.
>
> My Comet 2x4Max FM transmitting antenna is vertical collinear arrays of
> dipole radiators but the feedpoint is not balanced. Feeding a balanced
> antenna with an unbalanced feedline is a good way to cause common-mode
> currents and feedline radiation.
> --
> 73, Cecil http://www.qsl.net/w5dxpServers
Roy Lewallen
> I do know how broadcast FM/TV arrays are fed, thanks.
>
> Are not the currents on the inner and outer conductors of the coax feedlines
> for these systems everywhere equal for FM/TV dipole arrays when the Z of the
> feedline equals the Z of the dipole array?
The current on the outside of the feedline has nothing to do with the
impedance match, as I said before. I suggest you take a look at "Baluns:
What They Do and How They Do It", in the ARRL Antenna Compendium, Vol.
1; "Some Aspects of the Balun Problem" (Maxwell), in QST, March 1983;
_Elecromagnetic Waves and Radiating Systems_ (Jordan and Balmain), p.
406; or _Antennas_ (Kraus), 2nd Ed., p. 741 if you're really interested
in learning what causes current to flow along the outside of a coax
feedline.
> This system configuration has been used by the designers of commercial FM
> and TV RF system designers for 40+ years, and seems to have well-proven by
> theory and practice. Do you have evidence to the contrary?
No, I don't. I don't in fact, have any evidence that TV and FM
broadcasters don't load up womens' corsets for antennas. But I'm hoping
that you, with your commercial broadcasting background, can direct us to
some pictures that show exactly how typical commercial antennas are in
fact fed.
Roy Lewallen, W7EL
W5DXP
> Assume a single 1/2-wave, resonant dipole fed by 73 ohm coax. What amount
> of the total radiated field is attributable to radiation from the coax?
It depends on a lot of factors including the length of the coax. If the
path back to ground down the outside braid of the coax is an even number
of electrical 1/2WL's, the coax braid can present a lower impedance
than the dipole to which it is connected. I would say 40% of the source
power radiated from the feedline would not be completely outside the bounds
of probability. If you are trying to radiate horizontally polarized RF,
that can be quite a worst case problem.
On this subject, I bow to the experts. In the ARRL Antenna Compendium,
Volume 1, is an article by some obscure author :-) that explains almost
everything you are asking about. The title is, "Baluns: What They Do
And How They Do It."
Richard Fry
I humbly suggest that you investigate this subject for yourself, to your
(high) satisfaction. Apparently there is a gulf between amateur radio, and
commercial broadcast antenna system applications.
Yes -- bedsprings, aluminum eavestroughs, and metal garage doors WILL
radiate. Even, as you post, women's corsets (if you refer to their metal
stays). But how efficiently???
I must confess to a bit of disappointment in your response, but maybe it was
hyperbole?
RF
Visit http://rfry.org for FM broadcast RF system papers.
____________
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:vf9tlln...@corp.supernews.com...
> > This system configuration has been used by FM and TV RF system
> > designers for 40+ years, and seems to have been well-proven by
Roy Lewallen
two things. One is the impedance of the path to ground along the outside
of the feedline compared to the impedance looking into half the dipole.
The current from the inside of the feedline shield splits between these
two paths in inverse proportion to the impedances. The second
determining factor is the orientation of the feedline relative to the
antenna. Mutual coupling between the feedline and antenna induces or
modifies the current on the feedline outside. This relationship is a
more complicated one.
To answer your question, I'll assume that the feedline is perfectly
symmetrical with respect to the antenna. And for simplicity, I assume
that the feedline goes directly down to ground from the antenna, where
the shield is connected to ground.
I modeled a 14 MHz dipole. The height was approximately 1/2 wavelength,
varied a little to illustrate the effect. The current on the outside of
the feedline is maximum when the impedance of the path to ground from
the antenna is minimum. This occurs when the distance is about a half
wavelength (or multiple), which is why it was chosen for the
illustration. The feedline becomes a half wave vertical radiator with an
approximately sinusoidal current distribution. Therefore, the ratio of
the field strength from the feedline to that from the dipole is
approximately equal to the ratio of currents in the two, since the two
radiators have about the same length and current distribution shape. (On
the feedline, the current is maximum at the ends and minimum at the
middle, the opposite of the distribution on the dipole.)
When the antenna is exactly a half wavelength high, the ratio of antenna
current to feedline outside current is about 1:0.68. So about 0.68/1.68
= 40% of the total radiated field is from the feedline. Remember,
though, that it's vertically polarized. Adjusting the antenna height,
and therefore the feedline length, just slightly -- by about a foot --
causes just about all the shield inside current to flow down the outside
of the feedline. When that happens, half the dipole -- the half
connected to the shield -- gets virtually no current. The current on the
feedline is about equal to the current in the active half of the dipole,
but it's flowing in twice the length of wire. So the field from the
feedline in that case is 2/3 of the total field. If the antenna
height/feedline length is changed by five feet in either direction, the
feedline current drops to about 17% of the dipole current, so the
contribution of the feedline radiation becomes about 15% of the total.
When the feedline length (that is, the total distance to ground along
the feedline and ground conductors) is an odd number of quarter
wavelengths, the impedance of the path along the feedline to ground is
high, so current on the outside of the feedline is negligible.
Note that even if the feedline is symmetrically placed with respect to
the antenna, some current will be induced by mutual coupling if there's
conducted current on the feedline. That's because conducted feedline
current will imbalance the currents on the dipole halves, so the
coupling from the two halves to the feedline will no longer cancel. The
model took this into account automatically.
Roy Lewallen, W7EL
Roy Lewallen
experience by helping us see some of these antennas you're describing.
Hopefully this discussion will give you an incentive to look up one or
more of the references I provided. After reading them, perhaps you can
resolve the apparent contradiction to your own satisfaction.
I'm not really interested in spending a lot of time trying to explain
something to someone who has no interest in learning. But I'm acutely
aware of all the "lurkers" who read this group, and it is worthwhile
going to the trouble for the benefit of those who really are interested
in gaining a better understanding of how things work.
Roy Lewallen, W7EL
xyz
W5DXP wrote in message <3EF4F032...@ieee.org>...
>Roy Lewallen wrote:
>> Not everyone agrees with Cecil's interpretation of "matched", ...
>
>Since I put it in parentheses, it can mean almost anything. :-)
>--
>73, Cecil http://www.qsl.net/w5dxp
>
>
I recall that there was a 'cat' that also suffered from the side effects of
the "...whatever I say it means..." attitude. The cat was a prime candidate
for la-la land!
Dean -- W4IHK
WB3FUP (Mike Hall)
with "window" line. What the antenna does not see
on top the feed line makes up as a "vertical'
component. I agree about the TVI problems,
particularly if the Coax fed antenna is used on a
frequency that it is not naturally resonate on.
(odd half wave multiples)
--
73 es cul
wb3fup
a Salty Bear
"W5DXP" <Cecil....@ieee.org> wrote in message
news:3EF4ADAD...@ieee.org...
W5DXP
> I recall that there was a 'cat' that also suffered from the side effects of
> the "...whatever I say it means..." attitude. The cat was a prime candidate
> for la-la land!
I always wondered where Schrödinger's cat came from. :-)
--
73, Cecil http://www.qsl.net/w5dxp
-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
Bob Miller
wrote:
>RE: Below
>
>I humbly suggest that you investigate this subject for yourself, to your
>(high) satisfaction. Apparently there is a gulf between amateur radio, and
>commercial broadcast antenna system applications.
Since commercial antennas are typically out in the country, on a big
piece of land, I doubt the station engineers are worried about TVI or
whatever as a ham might be on a tight city lot. Are we not talking
about apples and oranges, here?
I never really caught the point you were trying to make, if any.
Something about feedline radiation? Does it matter on a commercial
antenna?
What station do you work for?
Bob
k5qwg
Roy Lewallen
strict requirements regarding antenna pattern and the polarization of
the emitted wave. Feedline radiation would affect both. That's why I'm
sure that the feed system is designed in such a way as to reduce
feedline radiation to a very low value -- and that they don't just use a
center fed dipole directly connected to a coax feedline as Richard has
said. But without the ability to see one up close or a picture of one, I
don't know how the feed system is typically configured.
Roy Lewallen, W7EL
Richard Fry
The opening page of website http://www.eriinc.com/ has a photo of a
complete, two-bay c-pol FM transmitting antenna, face mounted on a tower.
The antenna is built on a 50 ohm rigid coaxial "spine," and the input
impedance of the array is 50 ohms, unbalanced. The stem feeding each bay of
the antenna contains a step transformer made by adjusting the diameter of
the inner conductor in the stem, and a tee allowing each bay to be connected
to the 50 ohm spine, to receive equal power from it, and to drive the two
dipoles in each bay. The two outputs of the stem tee are fed internally and
coaxially through the driven arms to connect to the outer arm assembly just
past the insulators. No balun is used.
The antenna input is driven here by 3" OD, 50 ohm Heliax(tm) coaxial cable,
the black jacket of which can be seen at the very bottom of the picture.
The outer conductor of the Heliax is bonded to the tower at the top and
bottom of the vertical run, and at least 200 ft intervals between.
This is the typical installation used in 95% + of all US FM broadcast
stations, the biggest variable being the number of bays. More bays for a
given vertical bay spacing = more gain and less antenna input power needed
to reach the station's licensed ERP (effective radiated power), and
generally less radiation at higher angles above and below the horizontal
plane.
Paper # 4 at http://rfry.org graphically shows the development of the
radiation patterns from this type of antenna, and the affect on those
patterns from mounting the same array on two, simple tower structures. It's
only three pages in length, with no equations.
Much attention is paid by the EPA and FCC to broadcast system radiation
(power density), as it may affect human health and safety. Broadcasters are
required as a condition of their continued licensing to show that they
comply with their limits for the site. For those wanting to read more, look
for OET Bulletin 65 at http://www.fcc.gov/oet/info/documents/bulletins/
Radiation from the main transmission line to the antenna would be a serious
issue, because its source could be much closer to human exposure,
physically, than the antenna is. But virtually all of the power density
measured around a typical FM broadcast site is from the antenna(s), and
adjacent mechanical environment -- not the main transmission line. This
issue was not even discussed in the EPA work that led to OET 65.
RF
Visit http://rfry.org for FM broadcast RF system papers.
________________
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:vfa778o...@corp.supernews.com...
N. Kirk Pearson
radiation due to the balanced line used in the construction of a G5RV?
Reg Edwards
> Feeding a balanced
> antenna with an unbalanced feedline is a
> good way to cause common-mode
> currents and feedline radiation.
=============================
Cec, How much does it matter ? Should we lose any sleep about it ?
Could you give us a roughly average value for the percent of Tx power which
is radiated from an unbalanced feedline.
Just a number will do.
---
Reg, G4FGQ
J. McLaughlin
about antennas. (Evidence has been provided on this news group that HF
broadcasters are likely to be an exception.) A story I have told
before: A church retained me to deal with their concerns about a
proposed FM broadcast station (to be built adjacent to the church
building). I found that the proposed antenna system had a major lobe
straight down the tower (while meeting regulatory requirements). I
re-did the station's antenna such that it lowered their costs! and did
not zap the church. A classic win-win situation.
Parenthetically: Roy is exactly on point about how impedance
matching (no matter which end) of a transmission line can have little to
do with radiation from the feed line. Too many believe that conjugate
matching has magical properties. That is far too simple an idea. 73
Mac N8TT
--
J. Mc Laughlin - Michigan USA
Home: J...@Power-Net.Net
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:vfa778o...@corp.supernews.com...
W5DXP
> Could you give us a roughly average value for the percent of Tx power which
> is radiated from an unbalanced feedline.
Reg, Since we are talking about one antenna, averages don't matter.
My individual life span is what matters to me, not the average life
span of one million people my age. :-)
Consider that the way one models the outside of the coax in EZNEC is
by running a wire from the antenna back to the source ground. When
I get time, I will model that configuration for worst case.
Ron
>
> The initial question re: RFI with G5RV, are you worried more about harmonic
> radiation due to the balanced line used in the construction of a G5RV?
====================
I am concerned about any radiation that may be a result of the open
wire. As of now I don't have any problems with my coax fed dipole.
Ron
W5DXP
> N. Kirk Pearson wrote:
>
>>The initial question re: RFI with G5RV, are you worried more about harmonic
>>radiation due to the balanced line used in the construction of a G5RV?
> I am concerned about any radiation that may be a result of the open
> wire. As of now I don't have any problems with my coax fed dipole.
Care needs to be taken in the routing of the open-wire line. A good
rule is at least 6 inches away from any metal and a twist every
couple of feet or so. You can tape coax to a metal support pole.
Don't do that with open-wire line. Also, don't coil open-wire line
upon itself.
Reg Edwards
> Care needs to be taken in the routing of the open-wire line. A good
> rule is at least 6 inches away from any metal and a twist every
> couple of feet or so.
Why metal? Nearly all metals are very low loss. Other materials can be much
lossier.
---
Reg
W5DXP
Support poles tend to be either metal or insulators like my fiberglas poles.
I was thinking of support poles when I wrote that. Metal can cause arcing.
Roy Lewallen
posted results for a worst case, and how much difference a few feet of
feedline length made.
And I posted similar data not too long ago in response to the very same
question you asked then.
Is your ISP dropping postings? If so, you should be able to find those
postings via google, and finally have the answer to your question.
Roy Lewallen, W7EL
Roy Lewallen
dipole. To determine how the designer avoided feedline radiation will
require a good look at the feed system. I did a brief web search for
this antenna, hoping to find a better picture, but haven't located one.
So far, I've determined that it doesn't seem to be made by Dielectric,
Jampro, and Shively. Can you direct us to the manufacturer? A model
number would also be helpful. Perhaps the manufacturer has a better picture.
Incidentally, if you don't believe in modeling or in explanations given
in either amateur or professional publications, you'll find actual
measurements of feedline current and dipole current imbalance in
"Baluns: What They Do and How They Do It" in the _ARRL Antenna
Compendium_, Vol. 1. And if even that isn't sufficient evidence, it's
not at all hard to do a similar experiment in your own back yard.
Roy Lewallen, W7EL
Richard Fry wrote:
> Roy/Bob (re your statements below):
>
> The opening page of website http://www.eriinc.com/ has a photo of a
> complete, two-bay c-pol FM transmitting antenna, face mounted on a tower. . .
Roy Lewallen
I've described two mechanisms by which current can be made to flow on
the outside of a coax feedline. I've given four references which support
this explanation (although admittedly I wrote one of them, so there are
three others rather than four). I've demonstrated it with modeling and
with actual measurements (reported in the Compendium article). You
reject all this evidence, and insist that high SWR causes current on the
outside. So can you please explain to us just how it is that high SWR
does cause current to appear on the outside of the feedline? Can you
demonstrate it with measurements or modeling, as I've done with my
explanation? Can you point us to some published references where the
mechanism is explained?
Thanks,
Roy Lewallen
Graham
If we're looking at the same picture, the manufacturer is ERI.
They have better close-ups of the individual elements at:
http://www.eriinc.com/fm_antenna.html
Cheers!
On Sun, 22 Jun 2003 13:18:55 -0700, Roy Lewallen <w7...@eznec.com>
wrote:
Mark Keith
Shouldn't really matter much which line you use as far as TVI or
radiation from the line. I've never really noticed much difference
between coax and ladder line, assuming no problems with feedline
lengths, balance, etc. Being your neighbors are 30 ft away, it's quite
likely any TVI problems would be about as likely to come from the
antenna itself, as the feedline, unless the antenna is real high. If
you keep the antenna balanced, and are careful about routing the
ladder line, and also keep away from lengths that may be a problem,
you should have little trouble. Both ladder lines and coax are equally
capable of radiating. Some of the worst cases I've seen involved coax.
Being the ladder line is balanced, radiation from the line is
normally canceled for the most part unless you have a balance problem
with the antenna or line. They don't radiate near as much as some
people might think. BTW...If you did decide to do this, there is
really no need to switch to a G5RV. Just feed your present dipole
straight with ladder line all the way to the tuner. If you don't like
how it works, it's easy to switch back to coax. MK
Dan Richardson
wrote:
>Roy/Bob (re your statements below):
>
>The opening page of website http://www.eriinc.com/ has a photo of a
>complete, two-bay c-pol FM transmitting antenna, face mounted on a tower.
>The antenna is built on a 50 ohm rigid coaxial "spine," and the input
>impedance of the array is 50 ohms, unbalanced. The stem feeding each bay of
>the antenna contains a step transformer made by adjusting the diameter of
>the inner conductor in the stem, and a tee allowing each bay to be connected
>to the 50 ohm spine, to receive equal power from it, and to drive the two
>dipoles in each bay. The two outputs of the stem tee are fed internally and
>coaxially through the driven arms to connect to the outer arm assembly just
>past the insulators. No balun is used.
Oh yea? No balun used? From the manufacture's web page:
"ERI has engineered broadband element models to be used as building
blocks in the construction of custom horizontal radiation patterns.
The 1080 Flat Panel and the 1080 utilize pressurized *balun* support
stems with crossed, dipole arms......"
Danny, K6MHE
N. Kirk Pearson
which is supposed to radiate on the G5RV. Am I missing something? Thanks.
Reg Edwards
> My experience suggests that "commercial broadcasters" do not have a clue
> about antennas.
==============================
You are probably correct.
From the standard of old-wives' radio engineering displayed on this
newsgroup it is probably worse.
Fortunately there are sufficiently very few proficient truly professional
engineers to keep the world going. But they just get on with their jobs
without wasting their valuable time trying to educate us ill-educated
bickering newsgroup readers.
On the other hand this newsgroup is dedicated to Amateur Radio. Not
salesmen. Let's see more callsigns. Long may the many various ways of
obtaining 'kicks' from the hobby continue.
---
Reg, G4FGQ.
Roy Lewallen
It sure doesn't look much like a typical coax-fed dipole to me. But
actually, in some ways it is. I'll explain.
Hopefully Richard can correct me if I'm wrong in my interpretation about
how the feed works. I'm assuming that the vertical and horizontal pipes
are the outside of a coaxial feed. The grayish rings part way out each
of the curved dipoles look like insulators, so I'm guessing that's where
the inner coaxial conductor connects. That is, the center conductor
connects to the outside 1/3 or so of each dipole, making the effective
feed points the gaps where the gray rings are. If I'm correct so far,
then all the current from the inside of the coax shield flows onto each
dipole at the inner edge of the gap (that is, the side of the gap
closest to the center of the dipole). This of course is the main dipole
current, responsible for the pattern. Starting with the inner side of
the gap, current flowing along the dipole has a choice when it reaches
the horizontal pipe extending between the dipoles -- it can flow on out
the other half of the dipole, or it can flow along the horizontal pipe
between the dipoles. The amount which goes each way depends on the
impedances of the two paths. This is exactly the situation that occurs
with a dipole fed at the center with a coax line. But let's see what
happens next.
Whatever current flows along the horizontal pipe between the dipoles,
each dipole contributes an equal amount. This is obvious from the
symmetry (if you assume the dipoles are fed in phase). So if one dipole
injects one amp into the horizontal arm, then the other dipole also
injects one amp into the other end of the horizontal arm. The equal
currents are flowing from opposite directions, so their fields will
cancel (completely, broadside to the arm, and less completely in other
directions) there'll be very little radiation from this horizontal arm.
It's the same situation as the currents on two opposed ground plane
radial wires. A very crude EZNEC model with straight rather than curved
elements, and guessing at some dimensions, shows about 0.14 amps flowing
along the horizontal arm for 1 amp at each dipole feedpoint. The small
amount of radiation from the horizontal arm could be compensated by
slight modification of the dipole elements to keep the pattern circular.
It might not be necessary, though.
But what keeps the current on the horizontal arm between the dipoles
from flowing along the horizontal mounting arm to the vertical
structure? I think the trick here is in the length of the mounting arm.
I'll bet it's very close to a quarter wavelength. That way, the
impedance seen by any current on the horizontal arm between the dipoles
will be high. The conducted current on the horizontal mounting arm and
therefore on the vertical structure will be very low -- for exactly the
same reason that a dipole fed with a quarter wavelength of coax, with
the bottom of the coax connected to ground, will have very little
current on the outside of the coax. The fact that the impedance is high
at the center of the horizontal arm between the dipoles means that not
much current will flow onto the arm from the dipoles in the first place.
I've seen quarter wave conductive standoffs used in microwave
applications. They act like insulators, since the impedance looking down
from the top is very high. I'll bet that both Richard Harrison and
Richard Fry have seen this technique used for antenna elements. And
that's what I believe is being done here. Conducted current on the
outside of the feedline is being inhibited by means of choosing the
correct length for the mounting arm. In exactly the same manner,
conducted feedline current can be inhibited on a coax fed dipole by
having the distance to ground along conductors from the antenna
feedpoint be appoximately an odd number of quarter wavelengths.
This antenna simply illustrates the principles I've been describing, and
was obviously designed by someone who understands the principles. In no
way does it provide evidence that matching an antenna inhibits current
on the outside of a coax feedline or that mismatching causes it.
Roy Lewallen, W7EL
Reg Edwards
> capable of radiating. Some of the worst
> cases I've seen involved coax.
=============================
What is your definition of "worst" ?
Amongst your worst cases could you tell us very approximately what percent
of transmitter output power was radiated from the feedlines.
Just a number will do.
----
Reg, G4FGQ
Richard Fry
Sorry to be blunt -- but the ERI "1080 Flat Panel" configuration you refer
to below really is not germane to my post.
ERI (and some other OEMs of FM broadcast arrays) do manufacture various
forms of panel antennas, but their physical and electrical design is much
different than the element design I referred to in my post, i.e., the
"rototiller."
But the rototiller is THE most common element of sidemount FM broadcast
arrays in use in the US. Similarly-performing designs are supplied by
Dielectric Communications (their DCR series), Jampro (their "Penetrator"
series), and several US domestic & offshore imitators of these basic
element configurations.
The picture I referred to, and my description of it apply ONLY to the ERI
rototiller element -- which is NOT a 1080 panel element. Therefore the
1080 description you referenced is irrelevant to that element configuration.
RF
Visit http://rfry.org for FM broadcast RF system papers.
____________________________________
W5DXP
> The only balanced line in a typical G5RV is the 31 ft. vertical component,
> which is supposed to radiate on the G5RV.
Nope, that 31 ft. is not supposed to radiate. The G5RV is a balanced
antenna and the 31 ft. is balanced line. Why would it radiate?
Richard Fry
Roy,
This IS the manufacturer's website Just "clicking around" on the various
choices there will lead you to their various product designs.
Rgds,
RF
Visit http://rfry.org for FM broadcast RF system papers.
________________________
----- Original Message -----
> (clipped) Can you direct us to the manufacturer?
> Roy Lewallen, W7EL
++++++++++
> Richard Fry wrote:
> > Roy (re your statements below):
Roy Lewallen
But I think I've done enough now. I've given an explanation, references,
modeling results, and measurements (in the Compendium article) to
illustrate the principles. It's up to you, now, to show why this is all
wrong, and how high SWR causes feedline radiation.
Roy Lewallen, W7EL
Roy Lewallen
radiate is an unbalanced tuner or source. Then, neither that symmetrical
antenna nor symmetrical feedline will be balanced, and the line will
radiate.
Roy Lewallen, W7EL
Richard Fry
Accepted, however I thought you would have discovered this on your own,
given your technical acumen.
Rgds,
RF
Visit http://rfry.org for FM broadcast RF system papers.
_________________________
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:vfchjpl...@corp.supernews.com...
> Thanks, I didn't realize that, and thought it was a consulting company.
+++++
Richard Clark
wrote:
>Roy (re your post below the line)...
>
>Accepted, however I thought you would have discovered this on your own,
>given your technical acumen.
>
>Rgds,
>
>RF
>
Are we to presume this is the sum total of your technical rebuttal?
Roy's credentials outstrip yours to this point by actual demonstration
of theory and application - much less frequent exposure to critical
examination in this group and within the greater technical community.
There are already a number of professional engineers here with scads
of commercial radio experience that would have supported your
statement if they perceived its merit - but they too recognize Roy's
expertise. As another poster pointed out, you haven't offered much
beyond flippant sales engineering jargon.
This may be an amateur venue, but you haven't even risen to that level
of standard. Given the nature of newsgroup activity, you can either
complain of the mild abrasion, or respond to the technical challenge.
73's
Richard Clark, KB7QHC
'Doc
Richard,
Sounds like a 'cop out' to me. Are you going to
explain how SWR causes feedline radiation, or not?
'Doc
Reg Edwards
> Posting does sometimes seem to be a waste of time.
===========================
My service provider does not issue a guarantee. In any case I do not
guarantee to have time to read all the reams of stuff appearing on these
overloaded newsgroups. I do have other unpaid responsibilities.
===========================
> Just yesterday I
> posted results for a worst case, and how
> much difference a few feet of
> feedline length made.
===========================
I am quite familiar wth how much difference a few feet of feedline makes. It
can even double 1% to something as catastrophic as 2%.
Did you post a numerical percent value. If not then your posting was little
better than waffle. Remember Kelvin.
===========================
> And I posted similar data not too long ago in response to the very same
> question you asked then.
===========================
Yes I vaguely recall the occasion. But it could not have been the very same
question. If I remember correctly it was you who could have learned
somthing.
You may recall, a longer time ago, as you didn't seem to know, I told you
in a newsgroup thread, how to model a dipole, centre-fed via coax using
Eznec. It is of course a the simplest of procedures which you later adopted.
And which on the later occasion you used without acknowledgement. Not that
I expected any acknowledgement for such an obvious task.
===========================
> Is your ISP dropping postings? If so, you should be able to find those
> postings via google, and finally have the answer to your question.
===========================
You have quite deliberately twisted my question. A familiar sort of tactic
of yours when you foresee the remote possibilty of losing an argument. You
take such matters too seriously. This is a hobby newsgroup not a shop
counter.
I had asked Cecil for an average, typical, expected, practical, usable,
sensible approximate value of the percentage of transmitter power radiated
from the feedline. I asked only for an approximate number which in the
absence of any other information could be of guidance to the original
questioner. A worst case value is meaningless unless referred to a
completely specified system.
Cecil didn't know*. He waffled away on the subject of his personal
longevity. Which, from Cecil, is quite acceptable, everybody understands,
indeed even expects in his witty humorous replies.
Cecil then said, when he can find the time, he would use Eznec to provide
the worst possible case value (of the percentage). The worst possible case
is of no practical interest or use for the simple reason that in a practical
dipole + feeder it so seldom occurs it would have to be specially contrived.
You have a habit of using worse case examples to over-emphasize your point
in an argument without saying so.
I did not reply to Cecil that Eznec is incapable of providing an answer to
the percentage question because I wished to avoid the inevitable silly
diverting argument which was sure to ensue. It would have dragged you in.
But since you have, as usual, dragged youself in, confident in your
verbosity and eminent position in the Establishment, I will now say that
Eznec cannot provide an answer to the percentage question because Eznec has
enough common sense to recognise a stupid question when fed with one.
Can we now expect another outburst from that devoted ally of yours, another
erstwhile member of the establishment - the dictionary swallower who treats
all questions he is incapable of inventing an answer to as trolling ?
* And it appears neither do you.
----
Yours, Reg. G4FGQ
Richard Clark
<g4fgq...@btinternet.com> wrote:
>
>Can we now expect another outburst from that devoted ally of yours, another
>erstwhile member of the establishment - the dictionary swallower who treats
>all questions he is incapable of inventing an answer to as trolling ?
>
Ah Punchinello!
You are clearly the greatest inventor of the group - playing both
sides of the question is easily your own special art form, but a Texas
cowboy is a rival to the bard of troll.
You demand knee-jerk response, I supply the knee (rhetorically of
course). ;-)
In all due respect to your infirmities, I will leave you the privilege
of the last comment (provided it is zipped up).
88's
Richard Clark, KB7QHC
Reg Edwards
> of the last comment
At least you recognise yourself.
Wes Stewart
wrote:
|Roy/Bob (re your statements below):
|
|The opening page of website http://www.eriinc.com/ has a photo of a
|complete, two-bay c-pol FM transmitting antenna, face mounted on a tower.
|The antenna is built on a 50 ohm rigid coaxial "spine," and the input
|impedance of the array is 50 ohms, unbalanced. The stem feeding each bay of
|the antenna contains a step transformer made by adjusting the diameter of
|the inner conductor in the stem, and a tee allowing each bay to be connected
|to the 50 ohm spine, to receive equal power from it, and to drive the two
|dipoles in each bay. The two outputs of the stem tee are fed internally and
|coaxially through the driven arms to connect to the outer arm assembly just
|past the insulators. No balun is used.
Sigh. I've been trying to bite my tongue and stay out of this hoping
that Roy's comments would give you a clue, but alas...
Yes a balun is used, it is just cleverly included inside the driven
elements. Looks something like a double bazooka to me.
The first references to antennas of this type that I'm aware of are:
"New Multi-Station Top Mounted FM Antenna", IEEE Transactions on
Broadcasting, Vol BC23, No 2, June 1977
and
"A New High Power Circularly Polarized FM Antenna", IEEE Transactions
on Broadcasting, Vol BC-25, No. 1, March, 1979.
In both of the above, the feeder was outside of the driven elements
and configured in something like a combination delta/gamma match. The
center of each of the curved elements is fixed to the grounded support
structure, as in the (lousy) picture. However the elements are fed on
one side similarly to a gamma match, except the feed rods form a delta
with respect to the support arm.
Mark Keith
I have no idea. I just know which lines are the hottest with common
mode currents in comparison to the others. "FS meter" It's not the
percentage of power radiated by the line that is the biggest problem
to me. It will still be radiated somewhere. It's the effects of the
rf on radios, keyers, mike circuits, etc that is the real problem.
I've seen a few coax fed dipoles, with "bad" lengths of coax, IE: near
1/2 wave, and no baluns, chokes, etc, that would radiate so much rf on
the line, my whole station would malfunction. Garbled audio, mike
bite, keyer lockup, etc.
I don't see that very often running ladder line. Not that it can't
happen, but being coax is unbalanced feeding a balanced antenna, I
view it as more prone to common mode problems. Of course, usually, I
don't run "bad" lengths, and I use a choke, so it's normally a non
issue. Besides, if you have a proper operating balun or choke, then
coax length becomes a non issue. My "worst case" examples that I saw
were with no baluns or decoupling used. MK
Reg Edwards
> percentage of power radiated by the line that is the biggest problem
> to me. It will still be radiated somewhere.
================================
Mark, almost exact. It is impossible to segregate feedline power from
antenna power. It is akin to saying only the middle section of a dipole
does the radiating.
There may be something undesirable happening but we mislead ourselves,
possibly jump to incorrect conclusions, by referring to it as radiation.
The severity of problem can be sensibly discussed only by attaching a number
to a quantity which, if only in principle, can be measured. It is important
to be able to predict in given circumstances whether anything needs to be
done about it and what should be done.
The years of entirely unquantified waffle on this newsgroup about "radiation
from the feedline" does nothing but magnify the size of what most of the
time is un-noticeable to a vital-to-attend-to catastrophy which invariably
occurs.
At HF, forget about watts radiated. Think in terms of the near field,
measurable amps times length, and mutual coupling with adjacent conductors
in the vicinity. I admit the subject loses its excitement but it is kept
within reasonable bounds.
And remember in the amateur environment, even on a supposed balanced
feedline the unbalance current is nearly always of appreciable magnitude if
only because it is within the powerful near-field of the main antenna as is
a director or reflector. And who loses much sleep about that? "If it works
don't fix it" always applies.
---
Reg G4FGQ
W5DXP
> All it takes for a symmetrical line feeding a symmetrical antenna to
> radiate is an unbalanced tuner or source. Then, neither that symmetrical
> antenna nor symmetrical feedline will be balanced, and the line will
> radiate.
My G5RV (store bought) came with a hefty 1:1 W2DU choke at
the coax-twinlead junction.
W5DXP
> I did not reply to Cecil that Eznec is incapable of providing an answer to
> the percentage question because I wished to avoid the inevitable silly
> diverting argument which was sure to ensue.
EZNEC will indicate the magnitude of the current in the wire used to
simulate the outside braid of the coax. It will also indicate the
magnitude of the current in the antenna wire. What you will see is
the simulated outside braid of the coax sucking up current that
should be going into the antenna.
Richard Harrison
"At HF forget about watts radiated."
That`s been my experience with transmission lines. Symmetry and close
spacing avoid radiation. Good coax doesn`t leak r-f through its shield,
but poor shielding at its ends lets r-f in or out. Unbalanced induction
or radiation can put signal on the coax exterior.
A balanced open-wire line avoids radiation due to equal opposing signals
at a distance from the wires. As in the case of coax, unbalanced
induction or radiation can put a common-mode (unbalanced signal) on its
wires.
King, Mimno, and Wing quantify, to some extent, radiation from
transmission lines. This begins on page 235 of "Transmission Lines,
Antennas and Wave Guides".
Best regards, Richard Harrison, KB5WZI
Wes Stewart
|
|The first references to antennas of this type that I'm aware of are:
|
|"New Multi-Station Top Mounted FM Antenna", IEEE Transactions on
|Broadcasting, Vol BC23, No 2, June 1977
|
|and
|
|"A New High Power Circularly Polarized FM Antenna", IEEE Transactions
|on Broadcasting, Vol BC-25, No. 1, March, 1979.
|
|In both of the above, the feeder was outside of the driven elements
|and configured in something like a combination delta/gamma match. The
|center of each of the curved elements is fixed to the grounded support
|structure, as in the (lousy) picture. However the elements are fed on
|one side similarly to a gamma match, except the feed rods form a delta
|with respect to the support arm.
A picture is worth a 1000 words. See:
http://www.eriinc.com/fm_antenna.html#Anchor-Roto-Filler-49575
and look at the "100 series circular" for what I describe above.
|
Ian White, G3SEK
>Roy Lewallen wrote:
>> All it takes for a symmetrical line feeding a symmetrical antenna to
>>radiate is an unbalanced tuner or source.
Or an unbalanced environment for the antenna - different layout of
houses/masts/trees at opposite ends, and/or sloping ground beneath...
Or an unbalanced installation, typically any installation where the
open-wire and/or coax doesn't run away from the main antenna at
right-angles for a long, long way.
>>Then, neither that symmetrical antenna nor symmetrical feedline will
>>be balanced, and the line will radiate.
>
...to some extent. The question is whether the line radiates - or
conducts RF currents back into the house and shack, or picks up noise on
receive - *enough* to cause problems. This depends on many other factors
besides the antenna system itself.
The balun/no-balun controversy is mostly divided between people who have
had problems and cured them with a balun, and the second group who
haven't had any problems... or not yet.
There is also a third group who think the right time to add the balun is
*after* you've had RFI complaints from family or neighbors. Anyone who
thinks it might be a good idea to take simple precautions *before* going
on the air is obviously a worrier and an old wife.
Cecil wrote:
>My G5RV (store bought) came with a hefty 1:1 W2DU choke at
>the coax-twinlead junction.
Unless you're in that third group, that seems a right good idea.
--
73 from Ian G3SEK 'In Practice' columnist for RadCom (RSGB)
Editor, 'The VHF/UHF DX Book'
http://www.ifwtech.co.uk/g3sek
Bill Ogden
Being a little dense, I simply buy the right length of coax to go from my rig
to the antenna, speaking of quads or dipoles that are reasonably symmetrical.
Your illustration with a half-wavelength transmission line certainly got my
attention. Feeding a 20, 17, 15, 12, 10 meter antenna through a common
feedline (where the distance involved is 100 - 200 feet) means there are lots
of magic transmission line lengths to avoid.
Is a current balun (as represented by, say, 8 turns of coax in a 1 foot
diameter coil), right at the antenna, a general solution for this exposure?
(Or, equally, about a foot of the right type of toroids at the same location?)
Is it reasonable to say that, no matter how good the antenna match (although
not relevant) and symmetry (relevant), it would be good practice to place a
current balun at the antenna end of a transmission line for any dipole, loop,
or other symmetric antenna?
Thanks, from a lurker
Bill
W2WO
Mark Keith
> "Mark Keith" wrote It's not the
> > percentage of power radiated by the line that is the biggest problem
> > to me. It will still be radiated somewhere.
>
> ================================
>
> Mark, almost exact. It is impossible to segregate feedline power from
> antenna power. It is akin to saying only the middle section of a dipole
> does the radiating.
>
> There may be something undesirable happening but we mislead ourselves,
> possibly jump to incorrect conclusions, by referring to it as radiation.
Maybe so, but the line does radiate in such a case. IE: dipole fed
with 1/2 wave line and no balun used as an example.
>
> The severity of problem can be sensibly discussed only by attaching a number
> to a quantity which, if only in principle, can be measured. It is important
> to be able to predict in given circumstances whether anything needs to be
> done about it and what should be done.
Well, I think Roy pretty much covered this in another post. Could be
up to about 40% in a worst case. But in my case, I usually don't know
the *exact* length of my line, so I can only guestimate.
>
> The years of entirely unquantified waffle on this newsgroup about "radiation
> from the feedline" does nothing but magnify the size of what most of the
> time is un-noticeable to a vital-to-attend-to catastrophy which invariably
> occurs.
Depends on the scenario. On HF using a dipole or other simple antenna,
I agree, it's usually not worth worrying about as long as there are no
adverse reactions to the gear in the shack from common mode currents.
But if you use a yagi on HF, the decoupling becomes more important if
you want a clean pattern. But not overly critical. It will still
"work" ok, just a bit skewed. On VHF/UHF, decoupling often becomes
very critical. So it varies from case to case.
>
> At HF, forget about watts radiated. Think in terms of the near field,
> measurable amps times length, and mutual coupling with adjacent conductors
> in the vicinity. I admit the subject loses its excitement but it is kept
> within reasonable bounds.
I've never really considered watts radiated. I only worry about the
effects to the shack, or effects to the pattern of the antenna if need
be. I have no easy accurate way of measuring the watts radiated by the
line, so there is not much use in worrying about the exact value. I
could guestimate it using theory I suppose, but I don't bother.
>
> And remember in the amateur environment, even on a supposed balanced
> feedline the unbalance current is nearly always of appreciable magnitude if
> only because it is within the powerful near-field of the main antenna as is
> a director or reflector. And who loses much sleep about that? "If it works
> don't fix it" always applies.
I agree with the last statement. If it isn't whacking out my radio and
keyer, etc, I usually don't worry about it on HF. But, I always
consider decoupling on VHF/UHF important even if the currents don't
bother the radio. It's a major performance issue there. MK
Roy Lewallen
> "Reg Edwards" <g4fgq...@btinternet.com> wrote in message news:<bd6m97$j71$1...@sparta.btinternet.com>...
>>The severity of problem can be sensibly discussed only by attaching a number
>>to a quantity which, if only in principle, can be measured. It is important
>>to be able to predict in given circumstances whether anything needs to be
>>done about it and what should be done.
>
>
> Well, I think Roy pretty much covered this in another post. Could be
> up to about 40% in a worst case. But in my case, I usually don't know
> the *exact* length of my line, so I can only guestimate.
The worst case in the example I gave was that about 2/3 of the total
field radiation was from the feedline. With the worst case feedline
length of about 1/2 wavelength to ground, half the dipole gets just
about no current at all -- the current it would normally get is diverted
to the outside of the feedline. The current on the feedline is equal to
the current on the dipole half which does get current, but the feedline
is twice as long as the radiating dipole half. Both have a sinusoidal
distribution, so the field from the feedline is twice the field from the
dipole.
The feedline-to-dipole radiation ratio can be made as large as desired
by choosing longer feedlines in multiples of a half wavelength. For
example, a feedline approximately 1-1/2 wavelengths long (shield
grounded at the bottom) can be adjusted to suck the current from half
the dipole just like the 1/2 wavelength line did. If that happens, the
dipole has only 1/4 wavelength of radiating wire, while the feedline has
1-1/2 wavelengths carrying the same amount of current. So in that case,
the feedline is radiating 1.5/1.75 = 86% of the total field.
These are obviously contrived cases, with the line length carefully
adjusted to maximize feedline radiation. It simply illustrates how
profound feedline radiation can be, not how bad it's likely to be with a
randomly chosen feedline length. You might get lucky and pick a length
where feedline radiation is low, or have a situation where it doesn't
matter or even helps with your communication needs. Or you might not.
Roy Lewallen, W7EL
Roy Lewallen
> Question from a lurker, for Roy:
Welcome to the newsgroup, Bill.
> Being a little dense, I simply buy the right length of coax to go from my rig
> to the antenna, speaking of quads or dipoles that are reasonably symmetrical.
For a large number of situations, that's adequate. But for some, it's not.
> Your illustration with a half-wavelength transmission line certainly got my
> attention. Feeding a 20, 17, 15, 12, 10 meter antenna through a common
> feedline (where the distance involved is 100 - 200 feet) means there are lots
> of magic transmission line lengths to avoid.
Yes, probably so. As it turns out, though, it doesn't take much of a
deviation from the magic length to knock the current down considerably.
> Is a current balun (as represented by, say, 8 turns of coax in a 1 foot
> diameter coil), right at the antenna, a general solution for this exposure?
> (Or, equally, about a foot of the right type of toroids at the same location?)
Yes, provided that the feedline is positioned at a right angle to the
antenna to minimize coupling. But for a coax coil over that frequency
range I'd use the dimensions for a 21 MHz balun given in the ARRL
Handbook - for RG-8 sized cable, 8 ft. of wire in a coil of 6-8 turns,
or for RG-58, 6 ft. of wire in a coil of 8 turns.
If you can't run the feedline at more or less a right angle, you might
also have trouble with feedline current due to coupling with the
antenna. (This also occurs with ground plane and J pole antennas where
the feedline is collinear with the antenna.) So if the single current
balun at the feedpoint doesn't reduce the current adequately, a second
one about 1/4 wavelength below the first will usually be effective.
Hopefully, you'd have trouble on at most one band, so could position the
second balun appropriately for that band.
> Is it reasonable to say that, no matter how good the antenna match (although
> not relevant) and symmetry (relevant), it would be good practice to place a
> current balun at the antenna end of a transmission line for any dipole, loop,
> or other symmetric antenna?
Yes.
>
> Thanks, from a lurker
>
> Bill
> W2WO
Thanks for joining the discussion.
Roy Lewallen, W7EL
Bill Ogden
> Yes, provided that the feedline is positioned at a right angle to the
> antenna to minimize coupling. But for a coax coil over that frequency
> range I'd use the dimensions for a 21 MHz balun given in the ARRL
> Handbook - for RG-8 sized cable, 8 ft. of wire in a coil of 6-8 turns,
> or for RG-58, 6 ft. of wire in a coil of 8 turns.
>
I'll check the Handbook. 8 feet into 6-8 turns is a fairly tight coax coil, but I
assume it will not hurt the coax (LM400 type).
Getting away at right angles to a quad is a little tricky. My feeds from the 5 quad
elements go "up" to the boom and "over" to a coax switch. I plan to put the coax
choke on the feedline to the coax switch. I realize this is not ideal, but it is
probably better than nothing. I'll keep the plane of the coil at right angles to the
plane of the quad elements.
Regards,
Bill
W2WO
Roy Lewallen
shield at both ends of a piece of coax. Wind the coax into a coil you
think might do the job. Connect one end of the coax coil to the center
conductor of the antenna analyzer, and the other to the analyzer
connector shell. Adjust the coil to make the impedance as high as
possible. Generally, 500 - 1000 ohms is adequate for good balun performance.
Roy Lewallen, W7EL
Reg Edwards
ownership over the equipment's intended working life.
So DON'T bother painting the mast if it's needed only for field day.
And DO fit a choke balun only if the equipment won't work properly without
it.
Reg Edwards
> EZNEC will indicate the magnitude of the current in the wire used to
> simulate the outside braid of the coax. It will also indicate the
> magnitude of the current in the antenna wire. What you will see is
> the simulated outside braid of the coax sucking up current that
> should be going into the antenna.
=============================
Yes, Cec, it was I who originally described on this newsgroup how to model
it. It is the same simple model as used in program FEEDPOWR (dated 25th
March) which does calculate percentages. But only because "radiated power"
is the (misleading) popular description of what is going on.
However the 'numbers' do allow one antenna+feedline to be compared almost
instantaneously with another by sweeping any input parameter. Also, if you
are a nit-picker, it will tell you the extremely few line lengths to avoid.
If you search for the worst possible frequencies you will find you wouldn't
use those particular combinations of frequency and dimensions for altogether
different reasons.
Anybody who asks about accuracy in the midst of dozens of neglected sources
of uncertainty can be considered to be a Joker.
As you say Cec, EZNEC will crudely estimate actual unbalance amps in the
feedline. Current along a given length of conductor is what really matters
and is much the best way of tackling a particular problem. Furthermore,
feedline unbalance current is easily measured whereas radiated power is
unmeasurable and is therefore a meaningless quantity in the present context.
It's akin to saying most radiation is from the middle portion of a dipole.
And it must not be forgotten the current at the antenna end of a feedline
may have a much smaller effect than that indoors at the transmitter end even
if the top-end current is greater. It is further away from what is being
interfered with.
What practical use is made of the value of the unbalance current depends
entirely on the environment through which the feedline runs and the effects
it has on other equipment or people in the vicinity. This defies exact
analysis and comes under the heading of RF Compatibility. Of course, in the
most simple of cases the coupling-coefficients with nearby conductors can be
crudely estimated. Its near-field stuff.
But 'Suck-it-and-see' is a better and fastest way of doing things. Plus or
minus a couple of decibels is not a "Life or Death" matter.
To gauge the magnitude of what the argument has all been about, use FEEDPOWR
to model a 40m dipole fed via 50-ohm coax. Assume a ground connection
resistance of 20 ohms at the Tx end of the coax Vary feeder length from 1 to
100 metres (300 feet). The maximum power 'radiated from the feedline' occurs
at a line length of 1/2-wavelength and is 10.5 percent of total radiated
power. The AVERAGE power over the whole length is about 2.5 percent. And I'm
not going to lose much sleep over THAT!
----
=======================
Regards from Reg, G4FGQ
For Free Radio Design Software
go to http://www.g4fgq.com
=======================
W5DXP
> But 'Suck-it-and-see' is a better and fastest way of doing things. Plus or
> minus a couple of decibels is not a "Life or Death" matter.
Unless your feedline is running close to your wife's TV antenna. :-)
> The AVERAGE power over the whole length is about 2.5 percent. And I'm
> not going to lose much sleep over THAT!
Worst case is more than a magnitude worse than that and Murphey
never sleeps. :-)
Reg Edwards
> EZNEC will indicate the magnitude of the current in the wire used to
> simulate the outside braid of the coax. It will also indicate the
> magnitude of the current in the antenna wire. What you will see is
> the simulated outside braid of the coax sucking up current that
> should be going into the antenna.
The current entering a radiating element (the braid) gives no indication of
the power being radiated unless the radiation resistance of the braid is
known. EZNEC is unable to tell you what braid radiation resistance is. And
it's not the sort of thing which can be guessed at - specially when it's
fed from the top.
But that's only half the story. The current leaving the other end of the
braid is just as important. And so is the input impedance of the bottom end
of the braid plus the impedance of whatever it is connected to. Again, EZNEC
can't help you.
So forget about radiation from feedlines although I admit the description is
in the right sort of ballpark. Just be satisfied with amps at a particular
place over a particular length of braid and its geometrical relationship
with what is being interfered with.
If my TV receiver is suffering the first thing I do is shift it a few feet.
If that doesn't work I try a little ferrite choke in its power lead. The
last thing I do is buy another 20 or 30 feet of coax, wind it in the scruffy
form of a hank and suspend it 40 feet above ground from the centre of the
dipole as described in the ARRL handbooks. ;o)
----
Yours, Reg, G4FGQ.
W5DXP
> The current entering a radiating element (the braid) gives no indication of
> the power being radiated unless the radiation resistance of the braid is
> known. EZNEC is unable to tell you what braid radiation resistance is. And
> it's not the sort of thing which can be guessed at - specially when it's
> fed from the top.
On the contrary, Reg, I'm surprised that you haven't written a program to
do just that. It is similar to the last half of a folded monopole.