Linear decoupling traps
JN
The idea is to build the antenna of zip-cord type line 140ohm vf 0.73.
Trap is made using 1/4 wave shorted stub, decoupling occurs at the open end.
My question:
Is there any difference if the shorted end is pointing to the center or to
the tip of dipole.
Only the stub portion has two wires in parallell, the rest is only one wire.
----------------------------------------xxcenterxxx----
I_________
OR
----------------------------------------xxxxcenterxxx------
________I
73 Jouko OH5RM
cark...@juno.com
> I am planning a two band antenna for 30 and 80m using linear traps.
Jouko.......you might try to GOOGLE:
"lattin antenna"
That may lead you to useful information.
Lee KA0FPJ
JN
Jouko.......you might try to GOOGLE:
"lattin antenna"
That may lead you to useful information.
Lee KA0FPJ
Yes Lee, Google found quite a lot of hits, but they all refer to one 5band
antenna design with no much
real information. So my question still remains open. Could some modelling
program give the answer?
I think those two alternatives differ at least in how much inductive loadind
they are causing to the lower frequency
and so shortening the total length of antenna.
Jouko OH5RM
Richard Clark
>
>
>Jouko.......you might try to GOOGLE:
>
>"lattin antenna"
>
>That may lead you to useful information.
>
>Yes Lee, Google found quite a lot of hits, but they all refer to one 5band
>antenna design with no much
>real information. So my question still remains open. Could some modelling
>program give the answer?
>I think those two alternatives differ at least in how much inductive loadind
>they are causing to the lower frequency
>and so shortening the total length of antenna.
Hi Jouko OH5RM
Yes modeling can give an answer:
NO
It does not work.
73's
Richard Clark, KB7QHC
JN
>
> Yes modeling can give an answer:
> NO
>
> It does not work.
>
> 73's
> Richard Clark, KB7QHC
Hi Richard,
Sorry I didnt quite understand your short answer.
What does not work? The whole principe of decoupling stubs?
Unfortenately I myself have no modelling program.
73 Jouko OH5RM
Richard Clark
Hi Jouko,
Basically, what you describe was discussed here last week as the W9INN
dipole, and recently as the Lattin dipole. Being the same thing as
your twin line (parallel line, or folded stub, or whatever); the
premise is these elements resonate and thus trap an antenna for
multiband operation. Those stubs are not oriented correctly.
Let's take this by degrees. Any dipole is a multiband antenna. Those
bands might be useful, and they might not. The point is that being
multiband is nothing remarkable in itself. What is remarkable is if
that antenna is useful in every band you want to use it in. This is
the "Holy Grail." Nearly 60 years ago, a Ham invented the Lattin
antenna. We cannot say it was designed because it never performed
according to claims (and I do mean NEVER). It was even patented.
Designs do work, inventions rarely do. We get inventors here every
week, some hang around for years. The bottom line is that if these
inventions worked, we would be using them (and that is 2% of the goal
of these inventors, the other 98% is seeking validation as being
eminent thinkers). The Lattin antenna's balance sheet shows 1PPM
usage, and no pursuit of validation (the inventor is dead, but some
still keep the vigil and change the flowers at Internet memorial
sites).
You can try your hand at modeling by visiting:
http://www.eznec.com/
The Lattin design is easily constructed by a model, I've done several
dozen variations. You can also model stubs that are oriented
correctly and that will work.
JN
I am not talking abt Lattin or W9INN.
These are things I hear in this thread for the first time.
I am talking abt using 1/4 wave stub as decoupling element
to make a (reduced size?) two band antenna.
This principle is used in commercial antennas like HY-GAIN AT-18
tower vertical and many others, stubs in line with element.
Orr in his antenna book is using orientation, open to center.but in line
with the element
So I am sure the principle works ok, and I was asking abt the different
orientation
alternatives. I understand that you would suggest 90 degree?
But does it really matter?
Designs do work, inventions rarely do. Yes I agree.
Therefore I am not going to build any complex mess where everything is
hanging of everything, just simple one time decoupling the element for
higher fequency.
73 Jouko OH5RM
Cecil Moore
> The Lattin design is easily constructed by a model, I've done several
> dozen variations. You can also model stubs that are oriented
> correctly and that will work.
Modeling real-world lossy stubs seems to violate
EZNEC's guidelines. How does one do it?
--
73, Cecil http://www.w5dxp.com
Cecil Moore
> So I am sure the principle works ok, and I was asking abt the different
> orientation
> alternatives. I understand that you would suggest 90 degree?
> But does it really matter?
The series stub can be coiled in a circle
as a lot of commercial antennas do.
Richard Clark
>I understand that you would suggest 90 degree?
>But does it really matter?
Hi Jouko,
At right angles, yes. Does it matter? That depends on proximity to
other conductors, including itself if you "coil" it around the
radiator, or along the radiator.
When I looked at your first posting, I was confused by the single
wire, and the text graphics should have been done in fixed font.
Keep the line conductors at least 3, preferably more, wire diameters
from the radiator or themselves. Use the largest diameter for the
multiplier of 3 (or more).
Richard Clark
wrote:
>Modeling real-world lossy stubs seems to violate
>EZNEC's guidelines. How does one do it?
One does not get hyperbolic about trivialities.
JN
>
> ----------------------------------------xxcenterxxx----
> I_________
>
> OR
>
> ----------------------------------------xxxxcenterxxx------
> ________I
>
>
> 73 Jouko OH5RM
>
Different text sizes seem to destroy the graphics, I used the medium size in
Outlook express.
Anyway the upper is OK , the lower should look like this
______________________________________/center/_____
___________I
OH5RM
Buck
from OP
>My question:
>Is there any difference if the shorted end is pointing to the center or to
>the tip of dipole.
>Only the stub portion has two wires in parallell, the rest is only one wire.
Jouko, I noticed this thread took a detour, but I never saw anyone
actually try to answer your question. I too, am looking to see the
answer on this.
It is amazing, all this talk about a Lattin antenna, that appears to
have lots of sites talking about how it works, but no one actually
having built one or used one, and a simple "yes" or "no" would answer
your question.
I don't have the answers or the means of testing one at this time. If
you get the answer direct, please post it here.
Thanks
Buck
N4PGW
--
73 for now
Buck, N4PGW
"Small - broadband - efficient: pick any two."
JN
> actually try to answer your question. I too, am looking to see the
> answer on this.
>
Yes Buck it is often difficult to get an answer just to the question you are
asking.
Anyway very soon all snow has melted away here and then its possible to
testbuild it.
73 Jouko OH5RM
Jerry
"JN" <jn...@vfemail.net> wrote in message
news:480601b7$0$23850$9b53...@news.fv.fi...
Hi Jouko
I thought Richard's statement that modeling would tell you about how the
design might perform, was good advice. If you use EZNEC, I'd be willing to
help you if you dont already know how to use it to model your antenna.
Jerry KD6JDJ
Richard Clark
Repeated message seeing that you both missed it:
Cecil Moore
> Cecil Moore <nos...@w5dxp.com> wrote:
>> Modeling real-world lossy stubs seems to violate
>> EZNEC's guidelines. How does one do it?
>
> One does not get hyperbolic about trivialities.
Is 24 dBi omnidirectional gain from a vertical
antenna enough to "get hyperbolic about
trivialities"?
http://www.w5dxp.com/SUPRGAIN.EZ
z...@eircom.net
Theory says that whatever condition exists at far end of 1/4 wave
stub will
appear at fed end. ie. if far end is left open a high impedance will
appear at fed end, at resonant freq. So it would appear that the
shorted end should be nearest
to feed point of antenna. The centre of dipole or bottom of a
vertical.
Hope that makes sense!................................Rod EI3CZ
JN
> design might perform, was good advice. If you use EZNEC, I'd be willing
> to help you if you dont already know how to use it to model your antenna.
>
> Jerry KD6JDJ
>
Jerry, thanks for the offert.
I dont know how good these modelling programs are in this situation.
But if it is easy, so could somebody do the following:
Take 20m of wire, that is half of a dipole.
Put a 1/4 electric wavelength shorted stub for 10,1MHz so that the open end
is 7,25m from center and the shorted end pointing to the tip of antenna. One
side of stub is the 20m wire itself, like in the upper picture. Ignore all
velocity factors to make it sinple.
Now tell me what are the resonance frequencies of that kind of element and
if possible feed point impedande (80m and 30m)
I know I could do it myself but as you know learning to use a new program
reliably takes lot of time.
73 Jouko OH5RM
Cecil Moore
> Theory says that whatever condition exists at far end of 1/4 wave
> stub will appear at fed end.
That's true for an ideal 1/2WL stub but for an
ideal 1/4WL stub, an open at the far end will
appear as a short at the feed end and a short
at the far end will appear as an open at the
feed end.
Jerry
Hi Jouko
I am not sure I am reading your text properly. But, I did model a 40
meter wire with a trap on each side. You are welcome to E-mail me if you
want to see what EZNEC says about *my* model. Maybe you can help me model
your concept.
I have a friend in vasa Finland who can get you any information about
contacting me. Google Patrik Tast.
Jerry KD6JDJ
Jim, K7JEB
> Take 20m of wire, that is half of a dipole.
> Put a 1/4 electric wavelength shorted stub for 10,1MHz so that the open end
> is 7,25m from center and the shorted end pointing to the tip of antenna. One
> side of stub is the 20m wire itself, like in the upper picture.
I did just that with EZNEC, but with a free-space dipole. Without the
stubs, the antenna resonated at 3.6 MHz. With the stubs, that dropped
to 2.75 MHz and additional low-impedance points were noted at 6.5 and
11 MHz. The patterns at 2.75 and 6.5 had the desired dipole shape,
but the 11 MHz pattern had multiple lobes.
Jim Bromley, K7JEB
Glendale, AZ, USA
JN
>
> Richard Clark, KB7QHC
Hi Richard,
No everything red carefully.
I fully understand what you are suggesting.
Cutting the element and insertin stub at right angles
Sure works OK but that is no answer to my original question, HI
73 Jouko OH5RM
Dale Parfitt
<z...@eircom.net> wrote in message
news:fba46514-1b58-4d9b...@k13g2000hse.googlegroups.com...
You have it just backwards. A 1/4 wave stub is an impedance converter. i.e.
a short at one end appears as an open at the other end.
Dale W4OP
Richard Clark
wrote:
>Is 24 dBi omnidirectional gain from a vertical
>antenna enough to "get hyperbolic about
>trivialities"?
Hysterical AND hyperbolic apparently.
Richard Clark
Your original question:
>>>My question:
>>>Is there any difference if the shorted end is pointing to the center or to
>>>the tip of dipole.
My specific response:
Keep the line conductors at least 3, preferably more, wire diameters
from the radiator or themselves.
Certainly more was said, but you have shied from terse responses in
the past.
You have a tuned line, that is illustrated several times by you. You
have a radiator, that is illustrated several times by you. You have
asked about the line in proximity to the radiator several times, that
is illustrated several times by you (you really need to use fixed font
to do this, it is the protocol of allowing everyone to see what you
mean). I have responded about the proximity of the line to the
radiator several times with a very simple rule.
The line going radially away from the radiator (at right angle)
absolutely satisfies that rule. Keeping the line away from the
radiator (at a 0 degree angle, either up or down towards a tip) can
also satisfy that rule - but only if you observe the rule. If you do
not observe the rule you will have problems. You may have problems
even if you do - but your question lacks your requirements.
So, ostensibly, yes there is a difference - this is just the nature of
life. The real question is would it matter to you? Some folks worry
over half a dB, others can live with -10dB. This has all been
discussed to some length by others, in related topics about hairpin
matches. A general reading of the postings helps put specific
problems into context and relationship. This means browse the topics
that are active - not just your own.
Now, do you have an original question that has gone without an answer?
JN
http://farm3.static.flickr.com/2208/2419428284_1c7fd3ba40_o.jpg
Measures taken from hat, but if one could say the resonance points and
impedances I would be grateful
Originally I asked the difference if you turn the stub 180 degrees.
73 Jouko OH5RM
Cecil Moore
Great minds think alike. :-) I was doing the same thing
at different frequencies. I noticed that a 1/2WL dipole
with 1/4WL stubs on each end tends to resonate on frequencies
in a ratio of about 2.57:1. 18.14/2.57=7.06 so such an
antenna should be resonant on both 17m and 40m. EZNEC
agrees. Here are the approximate dimensions for the dualband
40m/17m antenna. Cutting and trying will be necessary for
fine tuning to resonance. Note that the antenna is about 13
feet shorter than a 1/2WL dipole on 40m. It has a dipole
pattern on both 40m and 17m and a 50 ohm SWR less than 2:1
on both bands. It appears to be linear-loaded on 40m and
stub-matched on 17m.
26.8' 26.8'
+-------------------------FP-------------------------+
| |
+------------ ------------+
13.3' 13.3'
Since 10.125/2.57=3.9 MHz, this antenna should be scalable
to become a dualbander on 75m and 30m.
Richard Clark
Hi Jouko,
If we are looking at four wire segments to the left of the center of
the dipole, then this is a lattin design element. All my comments
about that stand.
As for speculating where it resonates, there is not enough
dimensioning to be exact (while there is little enough to make a poor
guess).
I would only ask, do you want it to act as a trap such that the
radiating portion of the antenna is confined to the right (as
illustrated in your link) of this construction (like a conventional
trap for a dual band operation)? Or do you want the element to act as
linear loading to make the entire structure resonate at a shorter
physical length?
If you turn the element, yes, it will make a difference. There is
still a long road to travel to make it "work" and it probably won't
look like your link's illustration and it won't resonate where you
expect it to.
You still have to observe the separation of the two lines by at least
3 times the largest diameter wire/rod.
Jerry
Hi Jouko
I suspect that you either understimate your ability to learn or you arent
aware of how EZNEC works. I submit to you that a modeling program is the
best method of establishing the antenna design you are describing. You can
probably learn everything you need to know about your antenna within a few
hours of thinking on your own. You dont need to attend a class room course
on use of the EZNEC program.
I considered computer modeling to be beyond my ability till Richard Clark
encouraged me to 'just try it'. I tried it, and I consider the program to
be the best tool a HAM could have for developing antennas. I am sure you
will figure out how to model your "trap" antenna and learn how well it will
work for you within a very short time. You even have access to some EZNEC
experts from this rraa group. Try it, you'll like it.
Jerry KD6JDJ
Jim, K7JEB
exchange of a single, crucial, 1000-word, picture,
I think we've bottomed-out on this. I'm seeing two,
distinct, dipole-like responses, one at 3.6 MHz
and one at 10.2 MHz. The EZNEC outputs and
source file are at:
http://www.arizona-am.net/test/Antenna_Stubs_SWR.pdf
http://www.arizona-am.net/test/Antenna_Stubs_picture.pdf
http://www.arizona-am.net/test/Antenna_Stubs_10MHz_SWR.pdf
http://www.arizona-am.net/test/Antenna_Stubs_3MHz_SWR.pdf
http://www.arizona-am.net/test/Antenna_Stubs_wires.pdf
http://www.arizona-am.net/test/Antenna_Stubs_Patt.pdf
http://www.arizona-am.net/test/Ant_End_Stubs.EZ
Jim, K7JEB
Cecil Moore
> After a bit of e-mail correspondence and the
> exchange of a single, crucial, 1000-word, picture,
> I think we've bottomed-out on this. I'm seeing two,
> distinct, dipole-like responses, one at 3.6 MHz
> and one at 10.2 MHz.
10.2/3.6=2.8 It appears that frequency ratios
between about 2.4 and 2.8 can be easily achieved.
That's 75m+30m, 40m+17m, 30m+12m.
Unfortunately, a 2:1 ratio seems difficult to
achieve.
JN
In real life, at least in my case the stub will be made of twin-lead type
material which shortens it abt 30%.
If the shortening effect at lower frequency due to the inductive loading of
stub is less, then it would be possible.
Going one step further you could easily add another band to the design as
parallell dipole for 17m using unused parts of
douple-line. That should have litle interaction to the original double band
antenna.
http://farm4.static.flickr.com/3016/2420028687_e9fd3b1e6c_o.jpg
As Jim K7JEB verified, the total half length is not 20m but 15,5m due to the
loading effect of the stub.
The real stub length is 1/4 WL at higher frequency respecting the velocity
factor of material used.
73 Jouko OH5RM
Ken Fowler
On 16-Apr-2008, "Jim, K7JEB" <k7j...@yahoo.com> wrote:
> After a bit of e-mail correspondence and the
> exchange of a single, crucial, 1000-word, picture,
> I think we've bottomed-out on this. I'm seeing two,
> distinct, dipole-like responses, one at 3.6 MHz
> and one at 10.2 MHz.
I also played around with this yesterday on EZNEC. Started with a 75 M
center fed half wave at 30 feet high, added 20 M quarter wave stubs one
quarter wave (20 M) oout from the center. Got two low SWR points at ~3.5
MHz and ~14 MHz. By changing the distances to the stubs, the length of the
stubs, and the length beyond the stubs, I got the low SWR points to 3.9MHz
(2.8:1) and 14.2 MHz (1.2:1). Azimuth pattern on 14.2 MHz was sort of
omni-directional with major lobes at 45, 135, 225, and 315 degrees.
Pattern on 75 M was omnidirectional at high angles and max gain straight
up.
Conclusions: Might be useful as a way to add 20 Meters to a shortened 75 M
dipole. High SWR on 75 would still require a tuner. Losses were not
determined. More height might help.
Ken Fowler, KO6NO
Cecil Moore
> I also played around with this yesterday on EZNEC. Started with a 75 M
> center fed half wave at 30 feet high, added 20 M quarter wave stubs one
> quarter wave (20 M) oout from the center.
That's a conventional trapped dipole with stub traps
instead of LC traps. Here's a graphic from my web page:
http://www.w5dxp.com/eznec.gif
Buck
wrote:
I only have EZNEC 3 and the EZNEC 4 demo.
I don't remember the dimensions, but I remember that in the
instructions, EZNEC could not make accurate measurements with parallel
wires within a certain distance like 8-10 inches.
(someone correct me here.)
holding that assumption...
I have not modeled that antenna or made one, but I had an experience
with an 80 meter dipole that may relate. I cut a wire way too long
for 80 meters, I think it was something like 140 feet or so. I use
insulated stranded copper wire so I tried folding back the ends until
I shortened the antenna to 75 meters. What I discovered was, that the
change in frequency of the antenna did not match the reduced length of
the wire. After trying for quite some time, even after measuring the
antenna, I found it physically shorter than the calculated length, but
the center frequency, which changed a little, hadn't changed
significantly as expected. My conclusion was that I basically made a
linear-loaded dipole and the total electrical length of the antenna
was basically the wire length minus a small amount for interaction
between the folded back wire and the original leg. It wasn't until I
trimmed the wire itself that I raised the frequency of the antenna.
The wire I used had the ends loosely wrapped around the main wire of
the dipole, not tightly wrapped like the turns on a hangman's noose.
It was insulated THHN stranded copper.
Let's look at the 40/80 meter antenna cut for the CW portions (3.5 & 7
MHz). Assume, for the sake of argument that the 468/f = length in
feet accounts for the velocity factor and that it is the same for the
40 meter portion as the whole wire...
In theory, the Lattin antenna should be a total of 133.7 feet long,
with 66.85 (1/2) of the total length being the 40 meter wire and 1/2
on each end being the 40 meter decoupling trap. Overall there is a
total length of 200 feet of wire which theoretically would give you 40
and 80 meters. UNLESS the RF reacts according to my experience above
which would produce an antenna which would be tuned for 7.0 and 2.3
MHz (give or take reaction)
If this is true, and even with the reaction, the 40 meter dipole would
work, but you couldn't make the Lattin antenna work with the those two
bands. I might assume that 30 and 80 meters would work as the total
length of the 30 meter portion with trap would be less than 133 feet
and the 80 meter dipole would extend past the trap as an additional
extension of wire. Presumeably, that antenna would be resonant on 30
meters, 80 meters and somewhere else, maybe close to 6 or 7 mhz (Total
length of the wire from feed to end of trap.)
-------------------------------------------------------- O -------...
3.5 MHz | 10 Mhz
__________
trap plus ? Mhz
I don't know how the diagram shows, I hope you get an understanding of
what I am saying.
Comments?
=====================
PS, after writing all that, I re-read your post more carefully. I
think your EZNEC model reflected what i said all along. The
difference being that the continuation of the 80 meter dipole element
seems to be overlooked by EZNEC.
(ok, done this time :)
--
73 for now
Buck, N4PGW
"Small - broadband - efficient: pick any two."
JN
>I am planning a two band antenna for 30 and 80m using linear traps.
> The idea is to build the antenna of zip-cord type line 140ohm vf 0.73.
> Trap is made using 1/4 wave shorted stub, decoupling occurs at the open
> end.
Thanks to everybody for valuable comments.
Special thanks to Jim K7JEB, who made the simulation and verified that the
antenna
works as expected.
After much confusion with the text graphics, I learned the lesson to never
use them again
So here is how the antenna works
http://farm4.static.flickr.com/3078/2423002784_8c7c03acc4_o.jpg
I am going to use this kind of material, which is very commonly used by
telephone companies here.
http://www.saunalahti.fi/hohtola/ham/killu/killu.html
And here is how it will be made:
http://farm3.static.flickr.com/2072/2422189823_285653e9b9_o.jpg
Why not normal traps? Easier to build
Why not parallell dipoles? Much shorter
You get three bands if needed very easily
73 Jouko OH5RM
Roy Lewallen
> . . .
> I don't remember the dimensions, but I remember that in the
> instructions, EZNEC could not make accurate measurements with parallel
> wires within a certain distance like 8-10 inches.
>
> (someone correct me here.)
That's not correct. However, it is essential to align the segment
junctions so they're directly across from each other when modeling
closely spaced parallel wires.
Roy Lewallen, W7EL
JN
"Roy Lewallen" <w7...@eznec.com> wrote in message
news:VfidnZbNEK-reZXVnZ2dnUVZ_tHinZ2d@easystreetonline...
Roy,
One question:
Down the list is my posting FINAL PLAN.
Is it possible with EZNEC to simulate it?
It is made of transmission line (partly)
Yes or No is enough.
I have the DEMO version.
73 Jouko OH5RM
Roy Lewallen
>
> Roy,
>
> One question:
> Down the list is my posting FINAL PLAN.
> Is it possible with EZNEC to simulate it?
> It is made of transmission line (partly)
>
> Yes or No is enough.
> I have the DEMO version.
>
> 73 Jouko OH5RM
Yes, a simulation should be reasonably accurate. The segmentation
limitation of the demo version will probably reduce the accuracy some,
but you'll be able to get a very good idea of how it will work.
Roy Lewallen, W7EL
Buck
wrote:
Is this true for 3 as behind a version or two.
Thanks
Buck
Roy Lewallen
> On Fri, 18 Apr 2008 11:11:05 -0700, Roy Lewallen <w7...@eznec.com>
> wrote:
>
>> Buck wrote:
>>> . . .
>>> I don't remember the dimensions, but I remember that in the
>>> instructions, EZNEC could not make accurate measurements with parallel
>>> wires within a certain distance like 8-10 inches.
>>>
>>> (someone correct me here.)
>>> . . .
>> That's not correct. However, it is essential to align the segment
>> junctions so they're directly across from each other when modeling
>> closely spaced parallel wires.
>>
>> Roy Lewallen, W7EL
>
>
> Is this true for 3 as behind a version or two.
This is true for every NEC-based program, which includes all versions of
EZNEC. See "Parallel Wires" in the EZNEC manual index.
Roy Lewallen, W7EL
Buck
wrote:
You wrote the program, so I am not arguing with you, I will check it
out again. I sure feel I remember that being a restriction. Funny
thing is, that I haven't set learning the program to a higher priority
since I didn't think it would model what I want to experiment with,
which is parallel wires/poles.
I'll have to look into that more.
Thanks for the heads up.
BTW what is the upgrade cost for 4.0 and 5.0 from 3.0?
Thanks
Buck
n4PGW
Roy Lewallen
>
> You wrote the program, so I am not arguing with you, I will check it
> out again. I sure feel I remember that being a restriction. Funny
> thing is, that I haven't set learning the program to a higher priority
> since I didn't think it would model what I want to experiment with,
> which is parallel wires/poles.
>
> I'll have to look into that more.
>
Perhaps you're remembering a restriction with MININEC-based programs,
which had problems with close spaced parallel wires.
> Thanks for the heads up.
>
> BTW what is the upgrade cost for 4.0 and 5.0 from 3.0?
Sorry, the upgrade price is available only to people who have purchased
EZNEC. You can find information on purchasing a new program at
http://eznec.com.
> Thanks
> Buck
> n4PGW
Roy Lewallen, W7EL
Buck
wrote:
>Buck wrote:
My wife purchased it at a hamfest about three years ago before we
married.
JN
Yes indeed, the simulation verified that my FINAL PLAN is working OK
Two very clean resonances at 3.6MHz and 10.1MHz SWR abt 1.3 on both bands
Antenna length only 26m instead of normal 41m for 80m dipole.
For the stub section I used 0.003m=3mm spacing and 1.5mm insulation on 1mm
wire DielC 2.6.
Is there any possibility in EZNEC to calculate how long is quarter WL on
10.1MHz
with these stub variables?
By using transmmission line objects it could have been possible to use the
VF of line 0.73.
I did run out of segments to simulate tri-band version with additional
parallel dipole.
All simulation was made in free space and probably needs some minor tuning
in real life.
73 Jouko OH5RM
Buck
>Yes indeed, the simulation verified that my FINAL PLAN is working OK
>Two very clean resonances at 3.6MHz and 10.1MHz SWR abt 1.3 on both bands
>Antenna length only 26m instead of normal 41m for 80m dipole.
>For the stub section I used 0.003m=3mm spacing and 1.5mm insulation on 1mm
>wire DielC 2.6.
>
>Is there any possibility in EZNEC to calculate how long is quarter WL on
>10.1MHz
> with these stub variables?
>
>By using transmmission line objects it could have been possible to use the
>VF of line 0.73.
>
>I did run out of segments to simulate tri-band version with additional
>parallel dipole.
>
>All simulation was made in free space and probably needs some minor tuning
>in real life.
>
>73 Jouko OH5RM
>
>
Do you have the ARRL version of the demo, or just the one downloaded
from EZNEC?
If you have the ARRL version, you can modify a sample design and get
all the segments you need; you just can't save the antenna.
JN
> Do you have the ARRL version of the demo, or just the one downloaded
> from EZNEC?
>
> If you have the ARRL version, you can modify a sample design and get
> all the segments you need; you just can't save the antenna.
>
>
> --
> 73 for now
> Buck, N4PGW
>
> www.lumpuckeroo.com
>
> "Small - broadband - efficient: pick any two."
Hi Buck,
I downlowded the Demo and it is limited to 20 segments.
My ARRL antenna book is so old that it had no EZNEC.
73 Jouko OH5RM
Buck
>>
>>Perhaps you're remembering a restriction with MININEC-based programs,
>>which had problems with close spaced parallel wires.
I apologize. I think I mentioned that your program had limitations
that would not allow me to model what I wanted, which in my case is
close parallel elements. I was pretty sure that was why I quit using
it. Maybe there was another reason.
Whatever the case, I have, and still do, appreciate hearing your
knowledge and information.
It isn't my desire to design a new earth-shattering antenna or to
super-fine tune a yagi, but to use the rules of thumb in antennas to
build antennas with specific purposes in mind such as for portability,
maximum number of HF bands with minimal losses, and maximum gain with
minimum cost.
I gave up on an all-band HF antenna. there are just too many bands
these days. I built one from 80 - 6 meters in parallel dipole format.
It became such a bear to handle that it wasn't worth the effort trying
to keep it untangled, not to mention the weight of it.
I guess I digress.
73 for now
N4PGW
Buck
Roy Lewallen
Buck, send me an email so we can get this sorted out off-line.
Roy Lewallen, W7EL
Roy Lewallen
>
> Yes indeed, the simulation verified that my FINAL PLAN is working OK
> Two very clean resonances at 3.6MHz and 10.1MHz SWR abt 1.3 on both bands
> Antenna length only 26m instead of normal 41m for 80m dipole.
> For the stub section I used 0.003m=3mm spacing and 1.5mm insulation on 1mm
> wire DielC 2.6.
>
> Is there any possibility in EZNEC to calculate how long is quarter WL on
> 10.1MHz
> with these stub variables?
Not directly.
> By using transmmission line objects it could have been possible to use the
> VF of line 0.73.
>
> I did run out of segments to simulate tri-band version with additional
> parallel dipole.
>
> All simulation was made in free space and probably needs some minor tuning
> in real life.
EZNEC isn't able to model some radiating lines. If you use the
transmission line model, the line doesn't radiate. If you model it with
wires to allow radiation, you can use the wire insulation feature to
simulate a thin layer of insulation on the wires. This will have a small
(typically 2 - 3%) effect on the effective length of the wires as far as
common mode currents go, but very little effect on the differential
(transmission line) operation. But there's no way to model insulation
between wires, so when it has a significant effect on operation, the
analysis won't be accurate. Sometimes it's possible to separately model
the common and differential mode characteristics. For example, a folded
dipole can be modeled as a conventional dipole (for the common mode)
with transmission line model stubs across the feedpoint (representing
the differential, or transmission line mode) and a transformer for the
impedance transformation. Or a coax can be modeled as described in the
EZNEC manual. But in many cases, building such a model is difficult at
best. However, you can often get a good idea of how a system will work,
even if the actual antenna will require some adjustment compared to the
model.
Roy Lewallen, W7EL