Multiband doublet feedline length
dale.j.
a multiband doublet using 450 ohm open wire.
tnx
Dale
--
Email: dal...@mac.com
Unknown
wrote:
>What is the correct way to determine the non-resonant feedline length of
>a multiband doublet using 450 ohm open wire.
>
> tnx
>Dale
A multi-band doublet may in itself not be exactly resonate so
transmission line's length resonate length may not be a factor.
Normally - in typical ham style - use the length of line necessary to
make from you antenna to your transmitter (or antenna tuner). If you
encounter difficulties tuning and a few feet to your transmission
line.
Danny, K6MHE
In my many years I have come to a conclusion that one
useless man is a shame, two is a law firm, and three
or more is a congress. - John Adams
email: k6mhe<at>arrl<dot>net
http://www.k6mhe.com/
Reg Edwards
frequencies of both antenna and line and take the 450-ohm line (or
preferably 600-ohm) from the antenna all the way to the shack with a
little bit extra.
You will need a choke balun and tuner at the shack end. And you will
never have a more efficient arrangement.
----
Reg.
Cecil Moore
> With a multi-band, center-fed, doublet, forget all about resonant
> frequencies of both antenna and line and take the 450-ohm line (or
> preferably 600-ohm) from the antenna all the way to the shack with a
> little bit extra.
I think what the original poster is trying to avoid is a
voltage maximum, current minimum point at the transmitter.
For instance, that 25 ohm antenna fed with 600 ohm open-
wire line would present an impedance of 15K ohms at the
transmitter when the feedline is an odd multiple of 1/4
WL long.
--
73, Cecil http://www.qsl.net/w5dxp
Cecil Moore
> What is the correct way to determine the non-resonant feedline length of
> a multiband doublet using 450 ohm open wire.
Assuming that you are trying to avoid feeding the antenna
system at a current minimum point, the free demo version
of EZNEC will do that for you.
There's a DOS program on my web page at:
http://www.qsl.net/w5dxp/imax.exe
that will give the approximate feedline length for
feeding the antenna system at a current maximum
point.
Reg Edwards
"Cecil Moore" wrote
But that might occur at only one frequency in the range 1.8 to 30 MHz
and the probability of it falling in an amateur band is remote.
If it should fall into an amateur band and its on his favourite
frequency then all he has to do is change the feeder length by a few
feet.
Statistically he can't lose.
----
Reg.
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
So I would cut my feedline to an odd multiple of a quarter wavelength of
the lowest band I intend to use?
In my old 1960 ARRL antenna handbook (Ninth edition) they tell me to add
one half of the dipole plus the length of the feedline then go to a
chart, figure 3-50 p. 107, which shows various lengths that would be ok.
This stuff gets confusing when there are so many different ways that it
is explained.
--
Email: dal...@mac.com
Cecil Moore
> Statistically he can't lose.
How about Murphy's Law? :-)
Unknown
wrote:
>So I would cut my feedline to an odd multiple of a quarter wavelength of
>the lowest band I intend to use?
>
>In my old 1960 ARRL antenna handbook (Ninth edition) they tell me to add
>one half of the dipole plus the length of the feedline then go to a
>chart, figure 3-50 p. 107, which shows various lengths that would be ok.
>
>This stuff gets confusing when there are so many different ways that it
>is explained.
The information in the arrl antenna handbook you are quoting above is
for center feeding resonate (or near resonate) 1/2-wave dipole with
coax transmission line.
You are describing a multi-band doublet. That antenna is seldom
resonate on all bands (if any) and you are using open ladder
transmission line to feed it. Resonance is not an issue in this case.
Just put up the anenna, feed it with enough transmission line to make
to you shack. Should you encounter a problem matching with you tuner
then just add a few feet to the transmission line and be done with it.
Cecil Moore
> So I would cut my feedline to an odd multiple of a quarter wavelength of
> the lowest band I intend to use?
No, your antenna is probably not 25 ohms as in the example.
> In my old 1960 ARRL antenna handbook (Ninth edition) they tell me to add
> one half of the dipole plus the length of the feedline then go to a
> chart, figure 3-50 p. 107, which shows various lengths that would be ok.
There's a DOS program on my web page which will give the
approximate lengths for optimum feedline lengths:
http://www.qsl.net/w5dxp/imax.exe
For instance, that program says 58 feet and 89 feet are
good lengths for feeding a 102 foot dipole on 14.2 MHz
with 450 ohm ladder-line. Halfway between those two lengths,
i.e. anything around 74 feet, should be avoided for 20m.
More information on feedline length for a 130 foot dipole
is at http://www.qsl.net/w5dxp/notuner.htm
The free demo version of EZNEC will also yield this kind
of information.
Reg Edwards
> frequency then all he has to do is change the feeder length by a
>few feet.
On the other hand, it may be easier just to change the length of the
doublet.
----
Reg.
Dave
fed with 600 ohm open wire tuned feeders. Feeder length is just enough
to get from the tuner to the antenna <G>, about 180 feet. The
tuner/balun are bonded together and connected to electrical neutral.
No RF in the shack on 80 through 10. Well that's not really true. I'm
sitting in the near field; but, I don't have RF hotspots on anything in
the station at 1KW.
Unknown
<Relaxand@melthe/rose.comm> wrote:
>In article <NeadnThuMpMqk_PZ...@comcast.com>,
> Dave <W1...@comcast.net> wrote:
>
>> I use a "multi-band center fed doublet" for 80 through 10 meters. It is
>> fed with 600 ohm open wire tuned feeders. Feeder length is just enough
>> to get from the tuner to the antenna <G>, about 180 feet. The
>> tuner/balun are bonded together and connected to electrical neutral.
>>
>> No RF in the shack on 80 through 10. Well that's not really true. I'm
>> sitting in the near field; but, I don't have RF hotspots on anything in
>> the station at 1KW.
>>
>
>Thanks Dave. I just reconfigured my feedline and shortened it up also
>re-doing the physical layout. I have used this setup, open wire feed
>with 126 ft center fed wire for years and have never had a RF feedback
>problem in the shack. I asked the question because in the process of
>reconfiguration I consulted several sources which illuminated
>inconsistencies regarding length. From the responses I received here
>there does not seem to be any agreement on length and my question was
>not answered, at least not to my satisfaction.
>
>Anyway so far I'm ok with my current setup.
>
>73
>Dale, K9VUJ
>
It seems to me that we need to remember that radiation coming from the
feed line is cause by common current. If the feed line is balanced (no
common mode current) there will be no radiation from the feed line
regardless of the line's impedance or the feed point impedance.
Owen Duffy
<Relaxand@melthe/rose.comm> wrote:
>inconsistencies regarding length. From the responses I received here
>there does not seem to be any agreement on length and my question was
>not answered, at least not to my satisfaction.
That statement reflects your ability to comprehend what was written by
many, rather than what was written.
If you expected a specific length as implied by your question "What is
the correct way to determine the non-resonant feedline length of a
multiband doublet using 450 ohm open wire." then you have been
dissapointed.
The consistent answer (even if not always well expressed) was that
length is not particularly critical, though there may be some length /
frequency combinations (that are very dependent on your particular
implementation and equipment) that may challenge your ATU.
I go along with the advice to use a convenient length of feedline, and
if it is challenging to match up with the ATU, then lengthen the
feedline a little.
Your dissapointment is perhaps caused by your expectation of a
specific length.
Owen
--
dale.j.
Dan Richardson <> wrote:
I think you have brought up a good point as to the balance of the
parallel wires which is self canceling.
Maybe I will be able to sleep once again tonight...
73
Dale
--
Email: dal...@mac.com
Dave Platt
The always Benevolent dbu. <Relaxand@melthe/rose.comm> wrote:
>Thanks Dave. I just reconfigured my feedline and shortened it up also
>re-doing the physical layout. I have used this setup, open wire feed
>with 126 ft center fed wire for years and have never had a RF feedback
>problem in the shack. I asked the question because in the process of
>reconfiguration I consulted several sources which illuminated
>inconsistencies regarding length. From the responses I received here
>there does not seem to be any agreement on length and my question was
>not answered, at least not to my satisfaction.
Well, let me take a crack at it, if I may.
First-off: I don't think there's necessarily any single "right" length
of feedline, even for a single individual installation. There are
certainly a number of "wrong" lengths of feedline:
- The classic one - length is too short to reach the rig/transmatch
- Too long - feedline drapes on the ground, is looped around near
metallic objects, etc. - and there are excessive losses or shock
hazards
- End-of-feedline impedance on one or more bands is outside the range
which can be matched properly.
The first two sets of conditions are fairly easy to pin down. :-)
The third is, I think, the place where the interesting variables come
in. Transmatches aren't all equal, by any means, in the ranges of
impedances that they can match (and the matchable impedances don't
necessarily form a nice SWR-value circle on the Smith chart!), or the
losses in the tuner, or the amount of power that the tuner can handle
before it starts to suffer from arcing or balun overheating/saturation.
It's likely that in any given installation, there will be multiple
lengths of feedline which will allow a match on a particular set of
bands. The matches will probably vary in efficiency and Q. Which one
will be "best" is, to some extent, a value judgement by the installer
and will depend on how the station is going to be operated. For
example, if you spend most of your time operating on one or two bands,
you might choose a feedline length which allows efficient and fairly
low-Q transmatch settings, so that you don't have to adjust the match
as often (or as critically) when QSY'ing within a band.
Now... how to evaluate the possibilities. Here's one approach which
might be worth trying - it's a bunch of work but should let you end up
with some useful data.
[1] Grab a whole bunch of Smith chart blanks (or get some good
Smith chart software, or some other software which lets you handle
impedance transformations conveniently).
[2] Figure out the minimum and maximum lengths of feedline which would
be physically acceptable in your installation... from the shortest
which will reach the rig, to the longest which can conveniently be
zig-zagged back and forth between feedpoint and rig. Convert
these lengths to electrical length (i.e. divide by the feedline's
velocity factor, if it's much less than 1.0).
[3] For each band, convert the minimum feedline length into
wavelengths, modulo 1/2 wavelength (180 degrees). Also, calculate
the difference between minimum and maximum feedline lengths in
wavelengths. If it's more than 180 degrees, just call it 180.
[4] Determine the range of impedances that your transmatch can handle,
on each band that you're interested in working. You _may_ be
able to use whatever blanket marketing/spec statement the tuner
manufacturer provides (e.g. matches any SWR < 10:1, "all
impedances from 25 to 1000 ohms", etc.) but it wouldn't hurt to run
some tests yourself. You may wish to rule out some impedances
even if they're theoretically matchable - e.g. a T-match tuner can
often produce a "match" to a near-short-circuit, but the losses
in the tuner are so high that it's not worth the bother.
[5] For each band, plot the range of matchable impedances on a Smith
chart which is normalized to the nominal impedance of your
balanced feedline. Use one Smith chart per band to avoid
confusion.
[6] Choose the length of your doublet, based on space available and/or
desired pattern on one or more bands and/or lore and/or any other
criteria you desire.
[7] For each band, use appropriate tables to determine the feedpoint
impedance of the doublet on the band in question. You may wish to
calculate several values when the band is fairly wide (e.g. 80
meters).
[8] Plot the doublet's feedpoint impedance for each band on that band's
Smith chart.
[9] Using a compass centered at the Smith chart's origin, rotate the
feedpoint impedance "towards generator" by the number of degrees
corresponding to the shortest-possible feedline length, as
calculated in step 3. Plot this point.
[10] Use the compass (centered at the origin) to draw an arc from this
point, in the "towards generator" direction, for the number of
degrees corresponding to the difference between minimum and maximum
feedline lengths.
What this should leave you with (we hope) is a chart for each band,
which shows where the antenna's feedpoint impedance, transformed by
the feedline, falls inside the range of impedances that the transmatch
can handle. If the impedance arc comes near the centroid of one of
the "We can match this" circle/blob, then the closest-to-centroid
point on the arc would correspond to a feedline length which might be
a very good one for that band.
If the arc doesn't enter a band's matchable circle/blob at all, or
just barely enters it, then you may wish to go back to Step 6, pick a
different doublet length, and try again.
Now, what you can do is take a piece of paper and draw a straight
line. Put the minimum feedline length at one end and the maximum
at the other.
For each band, look at the impedance arc, record the sections at which
it's outside of the matchable-impedances blob (measure the number of
electrical degrees of rotation from the load), and figure out the
feedline lengths that these unmatchable sections correspond do. Use a
colored marker to denote these lengths on the straight "minimum to
maximum feedline" graph. Borrow color markers or crayons from your
kids so that you have one color per band :-)
Once you've finished doing this for all of the bands you're interested
in, take a look at the straightline graph of feedline length. If
you're unlucky, *every* point on that graph will be colored as "not
matchable" on one or more of the bands that you care about. If so,
you can go back to Step 6 and try a different doublet length, or you
can resign yourself to having to switch between two different lengths
of feedline when QSYing.
If you're not unlucky, there will be at least one range of feedline
lengths which can be matched on all bands. If so, I'd suggest choosing
a length which causes the impedance arcs you plotted to fall near the
centroids of the matchable-impedance areas, on the bands that you use
most heavily.
The above is very much a manual procedure, requiring at least an hour
of calculating and plotting and cussing. There are almost certainly
some software tools which can be used to automate and simplify this
process.
--
Dave Platt <dpl...@radagast.org> AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!
dale.j.
Owen Duffy <ow...@no.where> wrote:
I keep going back to my very old ARRL antenna handbood which shows a
chart of favorable lengths of feedline. Is it wrong?
I also would like to point out W8JI website and the lengths of feedline
that he recommends, which seems to be consistent with my ARRL handbook,
1960 edition, for this type antenna:
http://www.w8ji.com/short_dipoles_and_problems.htm
Arrow down.
--
Email: dal...@mac.com
Owen Duffy
wrote:
>I keep going back to my very old ARRL antenna handbood which shows a
>chart of favorable lengths of feedline. Is it wrong?
My 2000 ARRL Handbook describes a 135' multiband dipole antenna but
does not make any recommendations on preferred feedline lengths.
Perhaps they have changed their mind since your edition was published.
Given sufficiently good transmission line, lengths to avoid are only
lengths to avoid because they present a load that the tuner cannot
efficiently transform, or the voltage exceeds the capability of the
tuner.
Highest voltages usually occur at frequencies much lower than the
first parallel resonance... so avoiding resonance doesn't address the
most significant region of high voltage.
>I also would like to point out W8JI website and the lengths of feedline
>that he recommends, which seems to be consistent with my ARRL handbook,
>1960 edition, for this type antenna:
>
>http://www.w8ji.com/short_dipoles_and_problems.htm
>
He does more than recommend a feedline length in isolation of other
parameters.
Tom's article is relevant and interesting.
You will see that he links to articles by Cebik and myself on the
G5RV.
Also relevant to the multiband dipole application is the analysis in
my article at http://www.vk1od.net/LOLL/index.htm and its reference
documents.
To get off the 1960 tram, factors that are relevant are:
- they probably did not foresee the WARC bands and design to include
those bands;
- the regions where efficient transformation is a problem are usually
very narrow, but quite dependent on the exact implementation
(including proximity of other conducting objects, ground / weather
conditions), and locating them outside amateur bands may require
in-situ tuning.
Owen
--
Owen Duffy
wrote:
>I think you have brought up a good point as to the balance of the
>parallel wires which is self canceling.
Balanced current may be self cancelling, but currents are not self
balancing.
Owen
--
dale.j.
Owen Duffy <ow...@no.where> wrote:
Thanks for all the links which I am reading. My dipole is about 126
feet long which should be about right for 3.7 Mhz. I mostly use it on
75 M. The new feedline I have installed is a bit shorter than what I
was using, I think it's now around 47 feet, the old stuff was 50 plus
feet. I notice my Johnson matchbox now finds it somewhat difficult to
tune up this antenna on 40 M so I may have to add on a few feet. A
couple of years ago I had two loading coils on each leg of the tuner
output and the whole thing was a perfect match on 75 and 40, all I had
to do was change the bandswitch on the matchbox, nothing else.
I guess I'll just not worry about length of feedline only to make it
tune nicely and fit the physical space from the antenna.
73
dale
--
Email: dal...@mac.com
Cecil Moore
> Thanks for all the links which I am reading. My dipole is about 126
> feet long which should be about right for 3.7 Mhz. I mostly use it on
> 75 M. The new feedline I have installed is a bit shorter than what I
> was using, I think it's now around 47 feet, the old stuff was 50 plus
> feet. I notice my Johnson matchbox now finds it somewhat difficult to
> tune up this antenna on 40 M so I may have to add on a few feet.
Did you take a look at the information on my web page at:
http://www.qsl.net/w5dxp/notuner.htm
It tells you everything you need to know about that antenna
including why 47 feet is not the best feedline length on 40m.
In particular, the chart at:
http://www.qsl.net/w5dxp/pnts130.gif
shows the best lengths for 40m to be 30 feet or 92 feet.
A good compromise single length for all-HF-band operation
is 100 feet.
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
Yes I did Cecil. I went out and remeasured and found that I have about
67 total feet of feed line now. I may shorten that up again. I'll
re-check the links you provided, thanks.
--
Email: dal...@mac.com
dale.j.
BTW Cecil, I could never get 100 feet of open wire in my back yard
unless I coil it up or string it all over the place and that would not
be good. Isn't there a shorter length that would work for 75 & 40?
73
Dale j.
--
Email: dal...@mac.com
Unknown
wrote:
>BTW Cecil, I could never get 100 feet of open wire in my back yard
>unless I coil it up or string it all over the place and that would not
>be good. Isn't there a shorter length that would work for 75 & 40?
Look, if the dog gone tuner will tune the thing okay why mess around
adjusting the feed line length?
Cecil Moore
> Cecil Moore <myc...@hotmail.com> wrote:
>>http://www.qsl.net/w5dxp/pnts130.gif
>>
>>shows the best lengths for 40m to be 30 feet or 92 feet.
>
> Yes I did Cecil. I went out and remeasured and found that I have about
> 67 total feet of feed line now.
(30+92)/2 = 61 feet, worst case for feeding that dipole
on 40m. For all-HF-band operation, I would highly
recommend a feedline length of 100 feet. You can use
a 60 foot run of small non-conductive rope with the
ladder-line spiraled around it and tie-wrapped to it
to obtain the better case 100 feet.
Cecil Moore
> BTW Cecil, I could never get 100 feet of open wire in my back yard
> unless I coil it up or string it all over the place and that would not
> be good. Isn't there a shorter length that would work for 75 & 40?
As you can see from: http://www.qsl.net/w5dxp/pnts130.gif
the 3.8 MHz feedline needs to be about 110 feet. Worst case
for 3.8 MHz would be about 55 feet. Why can't you use a non-
conductive messenger rope and spiral the 100 feet of tie
wrapped ladder-line along its 60 ft. length? That has worked
well for me. I predict that your tuner will have no problem
tuning 100 ft. of feedline at least on the lower bands. Since
your antenna is 3 feet shorter than mine, it may exhibit
different characteristics on the higher bands.
Richard Clark
>Your dissapointment is perhaps caused by your expectation of a
>specific length.
Hi Owen,
There is one way to obtain a simple answer, you make the transmission
line an odd 8th wave long which will offer reasonable tuning. This
insight comes courtesy of Walt Maxwell's techniques offered for
measuring steep loads found in his classic "Reflections."
73's
Richard Clark, KB7QHC
Cecil Moore
> There is one way to obtain a simple answer, you make the transmission
> line an odd 8th wave long which will offer reasonable tuning. This
> insight comes courtesy of Walt Maxwell's techniques offered for
> measuring steep loads found in his classic "Reflections."
Where does Walt say that?
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
Well I'll give it some thought. All I need is to find room somehow for
33 more feet of ladder. Maybe I will try it jury rigged first to see
how it tunes, but not today, it rains. Doesn't the matchbox do the
electrical tuning of the feedline instead of using fixed lengths as you
are doing? As you said the length I have now does not tune very well on
40, I can get it to work but it's like opening a safe.
I just dug out my Drake MN-2700 user manual and it says to use 135 foot
long horizontal fed in the middle with 67 feet of ladder. Weird, this
stuff is worse than making a pizza.
Unknown
wrote:
>Richard Clark wrote:
>> There is one way to obtain a simple answer, you make the transmission
>> line an odd 8th wave long which will offer reasonable tuning. This
>> insight comes courtesy of Walt Maxwell's techniques offered for
>> measuring steep loads found in his classic "Reflections."
>
>Where does Walt say that?
In chapter 15.
Owen Duffy
wrote:
Richard,
The original question was regarding a multiband double of unspecified
length.
Aren't there lengths of the dipole that in concert with an eight wave
feedline would challenge the ATU?
Have you made assumptions about the doublet length in making this
recommendation?
Owen
--
Cecil Moore
> Doesn't the matchbox do the
> electrical tuning of the feedline instead of using fixed lengths as you
> are doing?
If it's a Johnson Matchbox, it has a limited tuning range.
But for dipoles 1/2WL at the lowest frequency of operation,
feeding in the rough vicinity of the current maximum point
will work. For 80m and 40m, you might get by with 80 feet
of feedline.
> I just dug out my Drake MN-2700 user manual and it says to use 135 foot
> long horizontal fed in the middle with 67 feet of ladder. Weird, this
> stuff is worse than making a pizza.
If that is for 80m or 40m, the Drake must like very high
impedances.
Richard Harrison
"I just dug out my Drake MN-2700 user manual and it says to use 135 fiit
long horizontal fed in the middle with 67 feet of ladder."
Drakes idea is probably a versatile antenna similar to the one on page
22.11 of the 2006 ARRL Handbook.
At 80 meters it is about 1/2-wavelength fed with about 1/4-wavelength of
feedline. A grounded radio presents a high impedance to comon-mode
travel where the line meets the antenna at the resonant frequency of the
antenna. Your radio should not bite you on 80 meters and a versatile
tuner should make the antenna useable through 10 meters. The pattern
goes through a lot of change in that range of frequencies. This is
predicted by a modeling program and some of the transition is shown in
the handbook.
Best regards, Richard Harrison, KB5WZI
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
I'll give the 100 ft a go and let you know.
When you say wind the ladder line around a tag rope, I assume you mean
to use ties around one wire of the ladder and make the loop maybe a foot
in diameter or maybe 6 inches? Sort of a helix affair. Would that
cause a problem with the ladder being so close to the turns? I've read
where I should keep the ladder from sharp angles and away from metal
objects so as not to unbalance the line and destroy the cancel
properties of ladder line.
The more I read about this the more it looks like black magic.
Dalej.
--
Email: dal...@mac.com
CW
"dale.j. " <nos....@nospam.com> wrote in message
news:nos.pamz-7FA88A...@news-rdr-03.rdc-kc.rr.com...
Richard Clark
>The original question was regarding a multiband double of unspecified
>length.
>
>Aren't there lengths of the dipole that in concert with an eight wave
>feedline would challenge the ATU?
Probability certainly suggests so. Russian roulette's popularity is
based, in part, with the notion "I'm feeling luckier than you." We
can both agree that perhaps the same mind-set is at work dreaming in
Technicolor and surround-sound for this doublet.
>Have you made assumptions about the doublet length in making this
>recommendation?
Hi Owen,
That was the point of my (Walt's) choice of an odd 8th wave length
transmission line, at least for a fire breather load, it pulls it into
a relatively placid range for matching (measuring).
If the original poster needs a simple rule, the odd 8th wave is good
as any. However, if this is in an effort to cover every band,
series/parallel resistors will do the job every time.
Owen Duffy
wrote:
47', now 67'... slowly more information emerges, even if changing.
The following is a one band (40m) perspective of your multiband
antenna.
If you have a 126' dipole (as you said), it has a feedpoint impedance
that is quite high around 7MHz because it is close to a full wave
dipole which has a feedpoint impedance around 4200+j0 (depending on
height, ground etc). A 126' dipole 15m above average ground is
probably around 3600+j1400 at 7.2MHz.
If you were using 67' of Wireman 551, you would expect the load seen
by the tuner to be around 660-j1400 ohms (assuming an effective 1:1
balun)... which is not extreme, but it is fairly high and might
challenge some ATUs. Lengthening the feedline will bring the load
impedance down fairly quickly, just another 10' (ie 77' overall) gives
90-j400. Shortening the feedline a little from 67' will make it worse,
62.5' will give you the maximum load impedance of 3500+ ohms.
Standing back and taking a system view, it is questionable whether
making the dipole resonant or very close to resonance at 3.7MHz as you
have done, makes harmonic operation easier. It certainly presents the
worst case feedline VSWR on 7MHz, the highest feedline voltage (>2500V
peak at 1kW), though the transmission line loss for 90' of 551 is only
0.6dB in the above scenario.
Owen
PS: I know conventional wisdom is to make a multiband dipole half
wavelength at the lowest operating frequency, and the ARRL shows such
an antenna with 130' dipole length.
It turns out that if a dipole with practical length of open wire
feeder to an ATU at the tx is less than about 35% of a wavelength,
system losses increase dramatically with reduced length, so the dipole
should not be less than about 35% of wavelength at the lowest
frequency. You can see the effect in Fig 2 of my article at
http://www.vk1od.net/LOLL/index.htm . In that case, the system losses
with a 66' dipole are acceptable down to about 5.5MHz, for 3.5MHz the
dipole needs to be greater than 5.5/3.5*66' or 103'.
--
Bob Miller
wrote:
>What is the correct way to determine the non-resonant feedline length of
>a multiband doublet using 450 ohm open wire.
>
> tnx
>Dale
I have a 130 foot dipole, about 25' high in the middle, 20' at the
ends, fed with 47' of the Wireman's ladderline, going directly to the
4-1 balun of an old mfj989c tuner. From everything I've read in this
thread, it shouldn't tune worth a darn on 40 meters, but I can get a
1-1 swr match on 40, as well as every other band. I guess all you can
do is put it up and see if it works :-)
One of these days I might try Cecil's idea, a 100 foot feed of
ladderline, but I'd have to string about 50' of that back and forth to
make it work.
bob
k5qwg
Cecil Moore
> I'll give the 100 ft a go and let you know.
>
> When you say wind the ladder line around a tag rope, I assume you mean
> to use ties around one wire of the ladder and make the loop maybe a foot
> in diameter or maybe 6 inches? Sort of a helix affair. Would that
> cause a problem with the ladder being so close to the turns?
You are forming 100 ft. of ladder-line into a 50 ft. long helix.
The spacing on the turns will not be a problem as it will be
a very loosely wound helix. Tie a knot in the messenger string
every 10 inches. Anchor one point of the ladder-line to a knot
in the messenger string using black tie-wraps. Take 20 inches of
ladder-line and anchor it at the next knot down from the first knot.
That will give you one turn of the helix. Repeat until you have
a 100 foot feedline. If you didn't want to use tie-wraps, you
could tie the 10 inch spaced knots through the ladder-line every
20 inches of ladder-line.
Cecil Moore
> Look at his site. Cecil does not use a tuner.
>
> "dale.j. " <nos....@nospam.com> wrote:
>>Doesn't the matchbox do the electrical tuning of the
>> feedline instead of using fixed lengths as you
>>are doing?
When I was using an ICOM-706, I didn't use a tuner. Now
that I have an ICOM-756PRO, I choose a feedline length
that resonates the antenna system in the center of a
band and then use the autotuner for excursions toward
band edges.
Making the feedline a fixed length of 100 ft. and then
using a tuner is somewhat similar to what I am doing now.
Cecil Moore
> It turns out that if a dipole with practical length of open wire
> feeder to an ATU at the tx is less than about 35% of a wavelength,
> system losses increase dramatically with reduced length, so the dipole
> should not be less than about 35% of wavelength at the lowest
> frequency.
Agrees pretty closely with Walter Maxwell's 3/8 wavelength
minimum dipole advice.
> You can see the effect in Fig 2 of my article at
> http://www.vk1od.net/LOLL/index.htm . In that case, the system losses
> with a 66' dipole are acceptable down to about 5.5MHz, for 3.5MHz the
> dipole needs to be greater than 5.5/3.5*66' or 103'.
Agrees pretty closely with the length of a G5RV.
Cecil Moore
> I have a 130 foot dipole, about 25' high in the middle, 20' at the
> ends, fed with 47' of the Wireman's ladderline, going directly to the
> 4-1 balun of an old mfj989c tuner. From everything I've read in this
> thread, it shouldn't tune worth a darn on 40 meters, but I can get a
> 1-1 swr match on 40, as well as every other band.
That "1-1 swr" is on the coax between the tuner and the
transmitter. The SWR on the ladder-line is probably around 12:1.
> One of these days I might try Cecil's idea, a 100 foot feed of
> ladderline, but I'd have to string about 50' of that back and forth to
> make it work.
Please see my other posting. Using a messenger string, tie
wrap 20 inches of ladder-line to every 10 inches of string.
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
I just tacked on additional ladder line for a total of 91 feet, it's
draped all over the place. The matchbox tunes more difficult on 75 M
but tuned right up on 40. The matchbox controls are skewed quite a ways
off the center point. tuning control is at 26 and match is about 15 so
it does not seem to like this length any better. I could tack on
another 15 feet but then I'd have to lay it on the ground which is not
good. I could shorten up the original 67 feet to perhaps 50 plus but
that would make it worse?
The Drake MN2700 uses a 4 to 1 balun for matching ladder line. I use my
Johnson KW matchbox for the ladder line.
--
Email: dal...@mac.com
Bob Miller
wrote:
>Bob Miller wrote:
If I were to use W7FG's 600-ohm line, would 100' still be a good
length?
bob
k5qwg
Cecil Moore
> If I were to use W7FG's 600-ohm line, would 100' still be a good
> length?
I don't know. If you have EZNEC, it can tell you.
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
I just dug out my Practical Wire Antennas book by John Heys, G3BDQ,
RSGB 1989, P. 16 & 17 "centre-fed antennas using tuned feedlines", and
it collaborates the information in my old ARRL 1960 handbook which says
to calculate one half of the horizontal wire plus the feedline length
and that total number should not equal certain critical lengths one of
which is 160 feet on 3.6 Mhz, there are more. I hate to beat this to
death, but frankly I don't think any of us has a clear understanding of
this antenna.
Dalej.
--
Email: dal...@mac.com
Owen Duffy
>... It certainly presents the
>worst case feedline VSWR on 7MHz, the highest feedline voltage (>2500V
>peak at 1kW), though the transmission line loss for 90' of 551 is only
>0.6dB in the above scenario.
When I read that in the clarity of a new day, I can see that is not
correct as written, I should have said:
It certainly presents the worst case feedpoint impedance at 7MHz, the
highest feedpoint voltage (>2500V peak at 1kW), though the
transmission line loss for 90' of 551 is only 0.6dB in the above
scenario. (A shorter dipole may result in higher voltage along the
feedline, and higher feedline loss, depending on the lengths of each
of dipole and feedline.)
Owen
--
Owen Duffy
wrote:
>In article <z8_bg.13144$fb2....@newssvr27.news.prodigy.net>,
> Cecil Moore <myc...@hotmail.com> wrote:
>
>
>I just dug out my Practical Wire Antennas book by John Heys, G3BDQ,
>RSGB 1989, P. 16 & 17 "centre-fed antennas using tuned feedlines", and
>it collaborates the information in my old ARRL 1960 handbook which says
>to calculate one half of the horizontal wire plus the feedline length
>and that total number should not equal certain critical lengths one of
As has been noted in several posts, this is not a new approach.
The approach seeks to avoid length combinations that result in the
impedance maximums at desired operating frequencies, whether or not
those impedance maximums are ok for the ATU.
It is an approximation in several ways, including that it does not
consider the velocity factor of the transmission line, it does not
consider the effects of the environment on the impedance at the feed
point of the radiator, it assumes that higher order modes are exactly
harmonically related.
So, even then, the lengths indicated as undesirable may have
significant error incorporated.
Let's look at an example:
Consider a model of a 40m dipole of length 20.12m at 10m above average
ground, and I want to use it at 28MHz with Wireman 551 feedline of 20m
(or greater).
The wavelength at 28MHz is 10.7m, so your method would suggest that
3*10.7 or 32.1m is a "total" length to avoid, meaning that a feedline
of 32.1-20.12/2m or 22.04m is the feed length to avoid.
At this "undesirable length", the model yields a load impedance of
82+j58 which should not be particularly troublesome to most tuners at
28MHz.
In fact the corresponding modelled voltage maximum occurs with a
feedline of 19.4m (not at 22.04m), so the method is 2.64m or 12% in
error in this example. The next voltage maximum occurs at feedline
length of 24.2m, so the forecast undesirable length is roughly in the
middle of these two lengths, and about as far as you can get (ie the
most error) from the modelled voltage maxima.
>which is 160 feet on 3.6 Mhz, there are more. I hate to beat this to
>death, but frankly I don't think any of us has a clear understanding of
>this antenna.
You don't understand it, so no one else does? I have got the picture.
Owen
--
dale.j.
Owen Duffy <ow...@no.where> wrote:
With all the J factoring ect I tend to get a bit intimidated. Then I go
back to the books from years ago. I don't have EZnec.
Dalej.
--
Email: dal...@mac.com
Cecil Moore
> I just dug out my Practical Wire Antennas book by John Heys, G3BDQ,
> RSGB 1989, P. 16 & 17 "centre-fed antennas using tuned feedlines", and
> it collaborates the information in my old ARRL 1960 handbook which says
> to calculate one half of the horizontal wire plus the feedline length
> and that total number should not equal certain critical lengths one of
> which is 160 feet on 3.6 Mhz, there are more. I hate to beat this to
> death, but frankly I don't think any of us has a clear understanding of
> this antenna.
EZNEC can solve the problem. EZNEC was not available to the
1960 hams. Let's take a 130 foot dipole on 3.6 MHz with
95 feet of 450 ohm ladder-line with a velocity factor of
0.9. 65+95=160 The impedance seen by the tuner is 100-j400.
Virtually every tuner in the world should be able to match
that impedance. The antenna system is resonant around 4 MHz.
The free demo version of EZNEC is available at:
http://www.eznec.com
How does the 1960 ARRL handbook handle velocity factor?
Cecil Moore
> With all the J factoring ect I tend to get a bit intimidated. Then I go
> back to the books from years ago. I don't have EZnec.
The free demo version will do what you need done.
It is available at: http://www.eznec.com
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
> dale.j. wrote:
> > With all the J factoring ect I tend to get a bit intimidated. Then I go
> > back to the books from years ago. I don't have EZnec.
>
> The free demo version will do what you need done.
> It is available at: http://www.eznec.com
OSX don't do executables and I'm not switching to windows. I'll just
struggle along with my musty old books.
Dalej.
--
Email: dal...@mac.com
Unknown
wrote:
>OSX don't do executables and I'm not switching to windows. I'll just
>struggle along with my musty old books.
You can lead a hourse to water but you can't ...............
dale.j.
Dan Richardson <> wrote:
> On Mon, 22 May 2006 01:23:08 GMT, "dale.j. " <nos....@nospam.com>
> wrote:
>
> >OSX don't do executables and I'm not switching to windows. I'll just
> >struggle along with my musty old books.
>
> You can lead a hourse to water but you can't ...............
What did you do before EZnec?
--
Email: dal...@mac.com
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
In Heys book circa 1989, he says "The velocity factor is only of
significance in certain antenna types, for example those requiring
quarter-wave 'stubs'. Such stubs are needed for the four -element
collinear antenna illustrated in Fig 21 and described later. Open wire
lines have a velocity factor of about 0.975 which means that a
quarter-wavelength stub at 7 Mhz will be 10in. (25cm) shorter than a
basic electrical quarter-wave-length."
In my old dusty musty ARRL 1960 and 1983 antenna handbooks I did not
find any reference to velocity factor in using this type of antenna.
Dalej.
--
Email: dal...@mac.com
Unknown
wrote:
Made a lot of mistakes that I don't do now.
dale.j.
Dan Richardson <> wrote:
> On Mon, 22 May 2006 09:11:21 GMT, "dale.j. " <nos....@nospam.com>
> wrote:
>
> >In article <l35272dj1m4f9p8o5...@4ax.com>,
> > Dan Richardson <> wrote:
> >
> >> On Mon, 22 May 2006 01:23:08 GMT, "dale.j. " <nos....@nospam.com>
> >> wrote:
> >>
> >> >OSX don't do executables and I'm not switching to windows. I'll just
> >> >struggle along with my musty old books.
> >>
> >> You can lead a hourse to water but you can't ...............
> >
> >
> >What did you do before EZnec?
>
> Made a lot of mistakes that I don't do now.
>
>
>
> In my many years I have come to a conclusion that one
> useless man is a shame, two is a law firm, and three
> or more is a congress. - John Adams
>
> email: k6mhe<at>arrl<dot>net
> http://www.k6mhe.com/
Well, I downloaded it on my shop machine and when I get around to it
I'll take a look as long as I don't have to spend a lot of time learning
it's operation. I'm going jogging now.
73
Dale
PS Nice web site Dan.
--
Email: dal...@mac.com
Cecil Moore
> What did you do before EZnec?
I remember my "Elmer" in the 50's adding extra
lengths of open-wire line on certain bands for
his 80m dipole. He said he did it to make his
Harvey Wells Band-Master Deluxe happy. I don't
recall him owning an SWR meter but he did have
a 6C4 homebrew GDO and a sensing loop for locating
current maximum points on the open-wire line. The
transformed impedance is lowest at the current
maximum point.
When I was a Novice, every time I changed bands
from 40m to 80m, I would let my coax-fed dipole
down and jumper across two insulators. That was
the extent of my knowledge half a century ago.
Cecil Moore
> Well, I downloaded it on my shop machine and when I get around to it
> I'll take a look as long as I don't have to spend a lot of time learning
> it's operation. I'm going jogging now.
I'm sending you the EZNEC file for a 127' dipole with
100' of 450 ohm ladder-line used on 3.6 MHz. It obeys
the 20 segment limit for the EZNEC demo.
Unknown
wrote:
[snip]
>
>Well, I downloaded it on my shop machine and when I get around to it
>I'll take a look as long as I don't have to spend a lot of time learning
>it's operation. I'm going jogging now.
Before you run the program, I suggest you printout Roy's "Getting
Started" portion of his manual, then one evening set down with a cup
of coffee and read it over. Roy has provided an excellent section on
getting started. It is not difficult and it will have you up and
running in a short order.
>
>73
>Dale
>
>PS Nice web site Dan.
Thank you and good luck.
73,
Danny, K6MHE
Al - VA3KAI
I read the postings regularly for new ideas on antennas and feedlines.
This recent posting with your ladder-line physical shortening idea is just
the solution I was searching for to reduce my ~65' of 300 ohm ladder-line to
stand-off my 40' tower and keep it out of my house. Now I can add a coax RF
choke loop at the base of my ladder-line feeder (still outside) and feed the
rest of the coax into the basement to my shack. I'll likely use knotted
black dacron line for the stringer/messenger support.
The antenna is an inverted 204' G5RV - any suggestions on how many RG-8X
turns and what diameter choke loop to use for mostly 160, 80 and 40 metre
band operations with this antenna - I was thinking about 8-9 turns on a 9"
loop? I plan to have it about 37' up with the ends dropping down to 20'. I
know it will be mostly a cloud-burner at this height, but it is the best I
can do given my tower height and tree availability and it will get me on
160M.
Thanks....... Al, va3kai
"Cecil Moore" <myc...@hotmail.com> wrote in message
news:z8_bg.13144$fb2....@newssvr27.news.prodigy.net...
Cecil Moore
> The antenna is an inverted 204' G5RV - any suggestions on how many RG-8X
> turns and what diameter choke loop to use for mostly 160, 80 and 40 metre
> band operations with this antenna - I was thinking about 8-9 turns on a 9"
> loop?
IMO, turns of coax, like 1/2WL of feedline, doesn't make much
sense on 160m because of the number of turns required. Turns
of coax on a large toroid makes more sense at the low frequencies.
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
Thanks Cecil, I downloaded it on my G5 and saved it to my AMD XP machine
in the workshop via my network. I did run it, but I'm going to have to
spend some time with it. I'll take Dannys advice and download Roys
getting started.
73
Dale
--
Email: dal...@mac.com
dale.j.
Cecil Moore <myc...@hotmail.com> wrote:
Seems like the Harvey Zmatch was more narrow tuning than the Johnson.
--
Email: dal...@mac.com
Buck
>Did you take a look at the information on my web page at:
>
>http://www.qsl.net/w5dxp/notuner.htm
Can you make that work using coax instead of open wire?
--
73 for now
Buck
N4PGW
Richard Clark
>
>>Did you take a look at the information on my web page at:
>>
>>http://www.qsl.net/w5dxp/notuner.htm
>
>
>Can you make that work using coax instead of open wire?
Hi Buck,
I built one some 10 years ago for what I call my Binary Variable
Transmission Line. I had obtained some precision transmission line,
that is solid copper shield (no vinyl jacket), but RG-58 size and I
coiled it in 6 inch loops of 1', 2', 4', 8' and 16'. I then used 5
DPDT switches to add the sections in, or to jump them. I mounted it
in a box and it works fine.
73's
Richard Clark, KB7QHC
Cecil Moore
>>Did you take a look at the information on my web page at:
>>http://www.qsl.net/w5dxp/notuner.htm
>
> Can you make that work using coax instead of open wire?
You cannot use 50 ohm coax to transform the impedance to
50 ohms since 50 ohms is the center of the SWR circle.
You could use side-by-side runs of 75 ohm coax to achieve
a balanced Z0 of 150 ohms.
But the reason that 450 ohm ladder-line works so well is
that 450/50 = 9:1 and 4050/450 = 9:1. 450 ohm ladder-line
can handle most any and all of the antenna impedances with
an SWR less than 18:1 on the ladder-line.
Cecil Moore
> But the reason that 450 ohm ladder-line works so well is
> that 450/50 = 9:1 and 4050/450 = 9:1. 450 ohm ladder-line
> can handle most any and all of the antenna impedances with
> an SWR less than 18:1 on the ladder-line.
Forgot to say the above applies to dipoles that are 1/2
wavelength on the lowest frequency of operation. It
doesn't apply to dipoles that are much shorter than 1/2
wavelength.
Owen Duffy
wrote:
>Buck wrote:
Since Cecil's scheme operates the transmission line at very high VSWR
on some frequencies (no matter what the line Zo), lossy lines such as
typical coax may (probably will) have unacceptable losses on some
frequencies.
Owen
--
Buck
>I just tacked on additional ladder line for a total of 91 feet, it's
>draped all over the place. The matchbox tunes more difficult on 75 M
>but tuned right up on 40. The matchbox controls are skewed quite a ways
>off the center point. tuning control is at 26 and match is about 15 so
>it does not seem to like this length any better. I could tack on
>another 15 feet but then I'd have to lay it on the ground which is not
>good. I could shorten up the original 67 feet to perhaps 50 plus but
>that would make it worse?
Suppose you trim the feedline a few inches at the time until the tuner
works on 80 and the other bands as well?
dale.j.
Buck <n...@this.time> wrote:
It's actually working pretty good right now and I have the ladder line
all symmetrical in the way it's routed from the antenna. I'm using my
original 126 ft long horizontal wire fed in the middle with 450 ohm
ladder which is 67 feet long going to my Johnson KW matchbox. It tunes
it on 75 and 40. 40 is a little tricky but I can get there ok. I also
use this same antenna as an inverted L on 160 by re-routing the ladder
via a knife switch to my Drake MN-2700 tuner. It tunes very nice on
160. When I get time from painting the house and other chores I will
try to sit down and learn EZnec, then tweak the feedline and maybe the
antenna too. I wish I had more room, I love to experiment with wire
type antennas.
73
Dale, K9VUJ
--
Email: dal...@mac.com