-
You can now help support WorldwideDX when you shop on Amazon at no additional cost to you! Simply follow this Shop on Amazon link first and a portion of any purchase is sent to WorldwideDX to help with site costs.
Homemade Sigma 4 11 meter base antenna
- Thread starter Lil'Yeshua
- Start date
B
BOOTY MONSTER
Guest
i think i recall a comment about the 4 element basket having more gain than the 3 element basket in the original design ..... i don't recall if i or anyone asked if there was a potential for even more gain with more basket elements .. so i'll ask now/again ..... if 8 basket elements would give more gain it might also remove the "minor distortion in the pattern" caused by just removing the ring . it would also make playing with basket element lengths easier since the X insulator for those elements wouldn't have to be moved up and down with those changes to keep the basket elements from bowing in or out .... and it would also make it much easier to play with basket element angles .... 
i'm guessing it would require a LOT more construction work for a minimal amount of added gain ...... but i don't know that .
anyhow ......... for the folks here that may not have seen this post by freecell (i so wish he would/could come back ..... i really miss his postings) he describes how a vertical/omni antenna changes as the vertical is lengthened , the ground element angle is changed and how different matching systems are used for different designs ..... ENJOY
WorldwideDX Amateur Radio Forums - Antenna Group
"i have posted this some time ago in The Members Portal, i've decided to
release it here first in the Antenna Group. let me know what you think........
Where the story begins........
when the ground wave field strength of vertical antennas is examined, it is found that if a 1/4
wave antenna has 100 mV. field strength at 1 mile, a 1/2 wave vertical will have about 125 mV.,
a 5/8 wave will have about 140 mV., but a 3/4 wave will have only about 80 mV..
the conditions under which these tests were conducted assume a fixed amount of transmitter power
with all antennas tested matched to the feedline so that Z=50 or R=50, X=0 at the target test
frequency and that all antennas are tested at the same exact feedpoint height.
as the antenna design with regard to wavelength^ is extended from 5/8 wave to 3/4 something not
unexpected occurs. additional lobes appear in the upper elevation plane reducing the amount of
field strength in the major lobe at the lower takeoff angle and the measured ground wave signal
strength is reduced.
(^ electrical, not physical)
and so, conventional antenna theory is confirmed and vindicated not only by the test results
above but given further credibility and affirmation by the many college level engineering texts
that have been written on the subject.
whether or not this was the starting place for the engineers at Avanti when they contemplated
the design of the controversial Avanti Sigma IV is not known but it makes sense that this would
be as good a point as any. herein was the challenge presented to not only defy conventional
theory and design but also in the process to come up with a design that eliminated the formation
of these additional and wasteful high angle lobes while increasing the ground wave gain and at
the same time reducing the sensitivity of the design to objects in the near field which tend to
wreak havoc with feedpoint matching and distort the radiation pattern.
first let's re-establish a few facts so that it's easier to follow along with what lies ahead.
beginning with a basic 1/4 wave ground plane with radials at a 90 degree right angle to the
active radiating element, typical input impedance is roughly 36 ohms at a predetermined height
above earth ground. as the radial elements are lowered towards earth ground the input impedance
rises. conversely, as the radial elements are raised the input impedance is lowered.
as the radials are raised towards the radiating element something else interesting occurs. the
radiating element becomes less sensitive to the influence of surrounding objects and terrain
in the near field, allowing the antenna to be mounted in locations and at heights above ground
that would be extremely deleterious not only to the feedpoint match but also to the radiated
pattern emanating from the radiating element.
as the radials are swept upwards towards the radiating element capacitance between the radials
and the radiating element increases, effectively raising the resonant frequency of the system
or causing the electrical length of both the radials and the radiating element to be shortened.
so, the closer the radials come to the radiating element the more the coupling capacitance
increases, the higher the resonant frequency climbs and the lower the feedpoint impedance drops,
approaching single digits and 0 ohms.
also remember that the gamma match is used to match 50 ohm feedline to loads representing less
than 50 ohms. let's assume at this point that we have raised the radials to the point that the
angle represented at the apex (bottom of the antenna) is such that it approximates a value of
15 degrees. at this point we have gross mismatch, (single digit impedance) a 3/4 wave vertical
element that started out resonant in the middle of the cb band (or whatever) and is now resonant
much further up the band (1/4 radials included) but we have eliminated the sensitivity of the
vertical element to surrounding objects and terrain which provides a distinct operational
advantage, mentioned previously. now we have to match the feedpoint to the feedline and provide
some compensating influence to counter the rise in the resonant frequency of the radiating
element.
as we can see from the above the gamma match is the logical choice for impedance matching and we
can re-resonate both the vertical element and the radials by increasing the physical lengths
until resonance is re-established at the intended operating frequency. we can lengthen the
radials and simultaneously improve the frequency vs. bandwidth attributes AND provide additional
physical support for the radials by simply connecting an aluminum loop that will intersect the
radial ends while extending the physical lengths of each from 89.5" to 107".
with that done we can now direct out attention to re-resonating the vertical element for the
intended operating frequency by providing additional tubing length (inductance) beyond what is
dictated by the standard 3/4 wavelength formula. with that accomplished all that is left is to
implement the gamma match and then adjust these last two variables to match the feedline while
maintaining resonance at the target frequency. this can be the tricky part as the gamma strap,
referred to by SigmaIV enthusiasts as the "Dogbone" also comes into play with these other two
adjustments.
after all of the hard work what you end up with is a full-size modified 3/4 wave vertical that
easily produces twice as much signal as a conventional 3/4 wave @ 80 mV..
as to all of the questions posed by the pundits from the J-Pole camp i have only this to say.
you have no clue what you're talking about. even the most uneducated person can look at these
two designs based on nothing but geometry and determine that they can't be the same antenna
design based on the dissimiliar symmetry alone. what seemingly escapes most of you is the way
that the feedline smoothly transitions into the SigmaIV while closely maintaining the same
basic concentricity (inner conductor to outer conductor diameters) as that of the feedline with
no exceptions as to the size and type of feedline used. this efficiency factor coupled with the
full-size 3/4 wave radiator and revolutionary upswept radial design is why the SigmaIV was ahead
of its time when it was first introduced to the market and lends itself to the design of even
longer antennas (in terms of wavelengths) to produce even higher gain figures without the
restrictions presented by the development of multiple lobes as was the case in the conventional
designs that we were previously limited to. "
i can kind of visualize a 1/4wgp growing like a young sprout into a fully developed mature tree .
i'm guessing it would require a LOT more construction work for a minimal amount of added gain ...... but i don't know that .
anyhow ......... for the folks here that may not have seen this post by freecell (i so wish he would/could come back ..... i really miss his postings) he describes how a vertical/omni antenna changes as the vertical is lengthened , the ground element angle is changed and how different matching systems are used for different designs ..... ENJOY
WorldwideDX Amateur Radio Forums - Antenna Group
"i have posted this some time ago in The Members Portal, i've decided to
release it here first in the Antenna Group. let me know what you think........
Where the story begins........
when the ground wave field strength of vertical antennas is examined, it is found that if a 1/4
wave antenna has 100 mV. field strength at 1 mile, a 1/2 wave vertical will have about 125 mV.,
a 5/8 wave will have about 140 mV., but a 3/4 wave will have only about 80 mV..
the conditions under which these tests were conducted assume a fixed amount of transmitter power
with all antennas tested matched to the feedline so that Z=50 or R=50, X=0 at the target test
frequency and that all antennas are tested at the same exact feedpoint height.
as the antenna design with regard to wavelength^ is extended from 5/8 wave to 3/4 something not
unexpected occurs. additional lobes appear in the upper elevation plane reducing the amount of
field strength in the major lobe at the lower takeoff angle and the measured ground wave signal
strength is reduced.
(^ electrical, not physical)
and so, conventional antenna theory is confirmed and vindicated not only by the test results
above but given further credibility and affirmation by the many college level engineering texts
that have been written on the subject.
whether or not this was the starting place for the engineers at Avanti when they contemplated
the design of the controversial Avanti Sigma IV is not known but it makes sense that this would
be as good a point as any. herein was the challenge presented to not only defy conventional
theory and design but also in the process to come up with a design that eliminated the formation
of these additional and wasteful high angle lobes while increasing the ground wave gain and at
the same time reducing the sensitivity of the design to objects in the near field which tend to
wreak havoc with feedpoint matching and distort the radiation pattern.
first let's re-establish a few facts so that it's easier to follow along with what lies ahead.
beginning with a basic 1/4 wave ground plane with radials at a 90 degree right angle to the
active radiating element, typical input impedance is roughly 36 ohms at a predetermined height
above earth ground. as the radial elements are lowered towards earth ground the input impedance
rises. conversely, as the radial elements are raised the input impedance is lowered.
as the radials are raised towards the radiating element something else interesting occurs. the
radiating element becomes less sensitive to the influence of surrounding objects and terrain
in the near field, allowing the antenna to be mounted in locations and at heights above ground
that would be extremely deleterious not only to the feedpoint match but also to the radiated
pattern emanating from the radiating element.
as the radials are swept upwards towards the radiating element capacitance between the radials
and the radiating element increases, effectively raising the resonant frequency of the system
or causing the electrical length of both the radials and the radiating element to be shortened.
so, the closer the radials come to the radiating element the more the coupling capacitance
increases, the higher the resonant frequency climbs and the lower the feedpoint impedance drops,
approaching single digits and 0 ohms.
also remember that the gamma match is used to match 50 ohm feedline to loads representing less
than 50 ohms. let's assume at this point that we have raised the radials to the point that the
angle represented at the apex (bottom of the antenna) is such that it approximates a value of
15 degrees. at this point we have gross mismatch, (single digit impedance) a 3/4 wave vertical
element that started out resonant in the middle of the cb band (or whatever) and is now resonant
much further up the band (1/4 radials included) but we have eliminated the sensitivity of the
vertical element to surrounding objects and terrain which provides a distinct operational
advantage, mentioned previously. now we have to match the feedpoint to the feedline and provide
some compensating influence to counter the rise in the resonant frequency of the radiating
element.
as we can see from the above the gamma match is the logical choice for impedance matching and we
can re-resonate both the vertical element and the radials by increasing the physical lengths
until resonance is re-established at the intended operating frequency. we can lengthen the
radials and simultaneously improve the frequency vs. bandwidth attributes AND provide additional
physical support for the radials by simply connecting an aluminum loop that will intersect the
radial ends while extending the physical lengths of each from 89.5" to 107".
with that done we can now direct out attention to re-resonating the vertical element for the
intended operating frequency by providing additional tubing length (inductance) beyond what is
dictated by the standard 3/4 wavelength formula. with that accomplished all that is left is to
implement the gamma match and then adjust these last two variables to match the feedline while
maintaining resonance at the target frequency. this can be the tricky part as the gamma strap,
referred to by SigmaIV enthusiasts as the "Dogbone" also comes into play with these other two
adjustments.
after all of the hard work what you end up with is a full-size modified 3/4 wave vertical that
easily produces twice as much signal as a conventional 3/4 wave @ 80 mV..
as to all of the questions posed by the pundits from the J-Pole camp i have only this to say.
you have no clue what you're talking about. even the most uneducated person can look at these
two designs based on nothing but geometry and determine that they can't be the same antenna
design based on the dissimiliar symmetry alone. what seemingly escapes most of you is the way
that the feedline smoothly transitions into the SigmaIV while closely maintaining the same
basic concentricity (inner conductor to outer conductor diameters) as that of the feedline with
no exceptions as to the size and type of feedline used. this efficiency factor coupled with the
full-size 3/4 wave radiator and revolutionary upswept radial design is why the SigmaIV was ahead
of its time when it was first introduced to the market and lends itself to the design of even
longer antennas (in terms of wavelengths) to produce even higher gain figures without the
restrictions presented by the development of multiple lobes as was the case in the conventional
designs that we were previously limited to. "
i can kind of visualize a 1/4wgp growing like a young sprout into a fully developed mature tree .
Bob, can you give us a link to this article?
BTW, I did a J-Pole model and slanted the stub element out to about 20* degrees, about the same as the Sigma4. I did show more skewing of the pattern, but I don't know if it could be detected just using a radio, and I don't know if the receiver's meter could detect a difference either. I'll see if I saved the patterns and post then here. After I did this I decided to do the W8JI model so I could add some credibility in the effort, but as you will see below, I found errors in his work too. I hope you can find Cebiks link. Maybe he has a model on the J-Pole with enough details to duplicate...that I can test for sure.
BTW, I did some models of an article by W8JI on a 2 meter J-Pole for your and Shockwave and maybe DB at: http://www.worldwidedx.com/cb-antennas/150476-w8jis-2m-j-pole-article.html
Last edited:
My J-Pole ideas
Bob and Shockwave, here are two of my models, one is a regular J-Pole with 8" spacing as apposed to W8JI's model with 1.2" spacing, where I slanted the stub out to about 20* degrees...to see how much difference it made to the skewing of the pattern.
BTW, these models do not show the mast connected to the Earth, because I have the pattern set to Free Space.
It's skewed, but it's not much. Is this about what you have seen?
View attachment Shockwave's slanted J-Pole idea.pdf
Bob and Shockwave, here are two of my models, one is a regular J-Pole with 8" spacing as apposed to W8JI's model with 1.2" spacing, where I slanted the stub out to about 20* degrees...to see how much difference it made to the skewing of the pattern.
BTW, these models do not show the mast connected to the Earth, because I have the pattern set to Free Space.
It's skewed, but it's not much. Is this about what you have seen?
View attachment Shockwave's slanted J-Pole idea.pdf
Interesting results, I wish I'd saved my models in the same way, if I get chance over the weekend I'll attempt similar
Yep they have a right to require payment, but I have a right not to pay too.
They suggest that a lot of stuff is still free, but I can't seem to access it, because I'm not registered. I tried to register free, but they request payment again, so I haven't figured out their madness yet.
Sorta reminds me of how our Government works or don't. <can not stand it>
Interesting results, I wish I'd saved my models in the same way, if I get chance over the weekend I'll attempt similar
35, if you are curious about the currents issues using Eznec that Shockwave and I have discussed, the issue is that the Antenna view may handle phase properly, but the tabular Currents Log does not always agree unless you use a trick like W8JI showed me on the 2 meter J-Pole thread I just posted. I am hopeful that I can get familiar and fully understand what is the proper way to model, and this procedure (trick) may be the solution concerning currents.
If we do not considered currents, we can still make models, but the currents are important...even though I don't see any difference yet in other results like gain, angle, or pattern development.
I'm a bit of a cheapskate and I'm using MMANA GAL Basic at the moment, a lot more facitities than the free version of Eznec, but at the same price but the cheapskate in me is fighting a losing battle with the Ham inside me that says spend a little and pay for Eznec, time will tell, I've got some photography equipment to sell thsi weekend, if it goes I'll pay for Eznec and relearn what I'd unlearnt :blushing:
but the cheapskate in me is fighting a losing battle with the Ham inside me that says spend a little and pay for Eznec, time will tell, I've got some photography equipment to sell thsi weekend, if it goes I'll pay for Eznec and relearn what I'd unlearnt :blushing:
Thanks Homer, this doesn't have any modeling info that I can use, but I did see the free space pattern of Cebik's J-Pole, and it looks just like mine. The skewed difference is so small as to make little operational difference, just as Cebik suggested and as I claimed earlier...the reason for my models in this case.
Again we see a case where it's all about very small differences...differences that are of little consequence, maybe even to real world experiences.
Not to be mean, but if I could model a feed line and a choke, I would bet I would likely find similar results.
If I add a MFJ-915 RF Isolator at the feed point of a Sirio Vector 4000,would that affect the way I set up the feed line coax,it being in half wavelengths? Would that detract from the SWR adjustment?