Modified Vector 4000

[bob85]

i see what you mean eddie, i have seen currents represented both ways, antenna mode currents in the radials are out of phase with currents in the lower 1/4wave of mopole but in phase with currents above the radials, like a j or open sleeve monopole, im not sure about transmissionline mode currents with angled radials.

 

[Marconi]

i see what you mean eddie, i have seen currents represented both ways, antenna mode currents in the radials are out of phase with currents in the lower 1/4wave of mopole but in phase with currents above the radials, like a j or open sleeve monopole, im not sure about transmissionline mode currents with angled radials.

Bob, at one point I was thinking I understood a little bit about transmission line mode and antenna mode currents. I thought it all had to do with the difference in equal and opposite phase relationships of parallel elements, roughly meaning TLM currents do not radiate much, similar to coax, while the ATM currents do most of the radiating.

BTW, I also notice that the currents flowing on the radials in Dxer's first image are inside of the radials while they're on the outside in the second image. I don't get this either.

I have probably seen currents represented several different ways also, but I would think there was some standard that modeling uses for understanding that avoids any confusion in this regard, that is until I noticed the differences above. In one of the links below is a warning for using the antenna view feature in Eznec, to make sure that phase is on, and that the wires are all properly connected or the phase relationships will be in error.

Jpole vs. Coaxial Dipole EZNEC Shootout | Ham Radio Help Desk

1/2 Wave vs. 5/4 Wave VHF EZNEC Shootout | Ham Radio Help Desk

 

[Shockwave]

Marconi, I'll try and respond but that was a lot at once.

It is NEC that suggests moderate changes in radial length and angle do not show a noticeable change in angle of radiation. I kept the original model at about one wavelength above ground and changed lengths and angle with little effect on the 8 degree angle. I could change radiator, loop, and radial lengths plus or minus 5 inches or more with no change in sidelobe angle. The gain did change. The angle will change a lot with height. It continues to lower until it is several wavelengths above ground where it levels off at about 1 degree.

When I built the FM antenna It was done entirely with field testing and prior to having access to things like the Avanti patent and EZNEC. This provided the most accurate results since they are actual tests but that takes the longest time. The 11 meter antenna did not receive as much care. When I worked on this I was interested in testing the ideas in the Avanti patent that suggested a wider angle radial would produce more gain. This required a shorter radial length with the loop being longer.

I don't think my testing failed me, it just took longer to really get a feel for the performance of my last modified 11 meter Vector. When I work on the FM model I constantly compare it to my last model by quickly changing the antennas on the same mast. I didn't have the ability to do that on 11 meters. It was only after I was done and replaced the stock Vector on the roof with the new one that I could see.

Problem there was too much time elapsed. In an hour conditions over 50 miles varied and at first made me think the signal was stronger on 11 meters. The next day it was less, that night it was more. Using it for several months showed and average of no gain increase and I believe what appeared as a gain increase on 10 was the result of a better match there. The stock Vector at that time was 31 feet, mine was shorter.

Tuning this antenna for maximum gain takes more then meets the eye. Nearly every change you make can be offset by changing another aspect of the antenna. Worse yet they all interact with each other. Start with radials that are too short and the antenna tunes better with a bigger loop. Start with a radiator that's too long and your gamma will have to be tuned with more capacitive reactance.

Tweaking a UHF Quagi for maximum gain is easier to do then this HF omni. In the interest of preventing others from wasting time coping my measurements on the "Modified Vector" I've made it clear I believe I made a mistake in it's tuning by not consistently comparing it against a reference antenna (stock version in this case) while I was tuning it. Something I do after each change on the FM version within minuets of the change.

I think it's no coincidence that the intensively time consuming tests that Bob has done on 11 meters line up with my testing on FM showing radial lengths just over 1/4 wave puts more signal where it counts. Further confirmation of this can be found in Sirio's latest upgrades to the Vector.

I have modeled a conventional 3/4 wave vertical with four 90 degree radials and when the radials are moved this far away from the radiator it does indeed have a major impact on radiation angle. Maximum sidelobe gain goes from below 8 degrees to over 45 degrees with no power at all close to the horizon.

I'm not sure where the exact change happens but it's clear that NEC shows small changes in radial angle from the stock Vector do not alter radiation angle. Big changes that approach horizontal radials have a very negative affect on the radiation angle in this design. Keep in mind the loop must be removed when the radials are spaced out too far but that in itself doesn't alter the antennas performance much once the radial length is compensated.

DXers post does show NEC is reporting currents in the bottom of the radiator to be out of phase with the top when the phase option is on. That would suggest the antenna once again is no better then a half wave. I wouldn't bet the farm on this. Remember Dr. Cebik who wrote a book on NEC modeling commented that this design would be difficult to accurately model with NEC.

I see NEC as a tool that helps you apply basic concepts to the antenna design. Where you can make simple changes and see how they interact with other variables in the design. If you think you can use it and a tape measure to make a perfect working model or measure and perfect actual gain, I doubt you'll get it right with the Sigma.

 

[bob85]

you may have seen this one eddie,

5/8 Wave J-Pole vs. 1/2 Wave J-Pole EZNEC Shootout | Ham Radio Help Desk

 

[Marconi]

you may have seen this one eddie,

5/8 Wave J-Pole vs. 1/2 Wave J-Pole EZNEC Shootout | Ham Radio Help Desk

Yes Bob, I saw that one some time back when someone post the original link to Ham Help. I thought that was the link I just posted and you're right.

SW, my posts always seem to cover a lot and for sure that is when I'm trying to understand. This is why I break my return responses down and include specific quotes from you guys, so I don't get off track.

Thanks!

 

[Henry HPSD]

Hi guys,


It took a while, me moving..hollidays etc..

The good news...
I have found G0KSC Justin and DK7ZB (both famous for enzec designs yagi's (try a google search
and Siegfried Jackstien (forgot his call) availible to answer questions.


The door is open now with them for futher contacts.
The first impression was that it just is a J pole with a quater wave matching section..
However Siegfried said there migth be some "add" in gain due to the currents in the radial section.

Oke...the next step...

I have thougths of my own but dont want to mistaking..
Wich question need to be answerd?

Kind regards,

Henry

 

[Marconi]

Marconi, I'll try and respond but that was a lot at once.

It is NEC that suggests moderate changes in radial length and angle do not show a noticeable change in angle of radiation. I kept the original model at about one wavelength above ground and changed lengths and angle with little effect on the 8 degree angle. I could change radiator, loop, and radial lengths plus or minus 5 inches or more with no change in sidelobe angle. The gain did change. The angle will change a lot with height. It continues to lower until it is several wavelengths above ground where it levels off at about 1 degree.

I can understand the change with height affecting the angle of radiation, but in light of what Bob says about his own work and what he says Sirio claims about the reason for their recent change in the Vector, did you question your results when using Eznec to compare the modifications to the model you adjusted? Both Bob and Siro see the opposite of what you claim above.

When I built the FM antenna It was done entirely with field testing and prior to having access to things like the Avanti patent and EZNEC. This provided the most accurate results since they are actual tests but that takes the longest time. The 11 meter antenna did not receive as much care. When I worked on this I was interested in testing the ideas in the Avanti patent that suggested a wider angle radial would produce more gain. This required a shorter radial length with the loop being longer.

I understand that you tested using field test procedures only with your FM model and the 11 meter Modified Vector, that is why I was surprised when you changed your mind---that it worked better than a stock Vector. It seems to me that you changed your thinking when you received the Eznec model from Henry and not a new revelation in studying your field test results. Am I wrong?

If you remember, how did the change in the sleeve (shortening and widening the angel) affect the resonance of the antenna initially, before you re-tuned? Did you try to see where the new resonance was before re-tuning?

I ask this because I’m interested in understanding how such changes affect resonance. I recall the first three knockoff Sigma4’s that I bought for some buddies were from CTE, and they clearly stated on the box that the antenna was 31’> and was tunable from 10 thru 12 meters. Two of the antennas owner’s installed them as long as possible as there was very little installation information. Later, with all the subsequent bitching I finally got the story that they did not work worth a damn, having no ears is the way they were described. I was told later on by someone else at one of the installs that the owner insisted the antenna be made a long as possible, and that the antenna had some fluted (ribbed) tubing in the base of the antenna and he had never seen such before. Does that Vector sound familiar to you SW or anybody?

I have never experimented with my Sigma, but if I’m right about the bottom quarter wave section’s pattern being compressed by the contributing affects of the upper 1/2 wave section, then I believe adjusting the radial length must have a very strong affect on resonance as well, and making the radials longer will require you to shorten the radiator---just like Sirio did with their New Vector. Both you and Bob, on the other hand, claim to have made your radiators up to 3’ feet longer and fixing the resonance by adjusting the gamma.


You both may be right on the mark here, but it seems like a real stretch to make a tune for the gamma matcher to me. I can understand a few inches of adjustments, but several feet of tip and who knows what adjustment to the gamma leaves a big question in my mind. I have a gamma off of a LW 150, that I sold and it is shorter overall than my AS Sigma4, but the LW 150, base is longer @ 13" vs. 9" inches for the S4. It may be enough to tune for you guys, but that doesn't leave much room for adjustment. I’m waiting for Homer to make his adjustments to see what he finds. Hopefully he will explore these resonance affects.

I don't think my testing failed me, it just took longer to really get a feel for the performance of my last modified 11 meter Vector. When I work on the FM model I constantly compare it to my last model by quickly changing the antennas on the same mast. I didn't have the ability to do that on 11 meters. It was only after I was done and replaced the stock Vector on the roof with the new one that I could see.

I understand this and it’s plausible. After I did my work last summer, I was too burned out to really hash-out my notes, and after doing so, months later, I had revelations that I did not see at the time of testing.

Problem there was too much time elapsed. In an hour conditions over 50 miles varied and at first made me think the signal was stronger on 11 meters. The next day it was less, that night it was more. Using it for several months showed and average of no gain increase and I believe what appeared as a gain increase on 10 was the result of a better match there. The stock Vector at that time was 31 feet, mine was shorter.

That is about the same time as my testing here, and I was seeing the same strong variables in signals. Bob will tell you how I was complaining last summer, and it was not due to DX interference as a rule. But here again, you were making the antenna as long as possible and yet you kept exploring the effects higher in frequency.

Tuning this antenna for maximum gain takes more then meets the eye. Nearly every change you make can be offset by changing another aspect of the antenna. Worse yet they all interact with each other. Start with radials that are too short and the antenna tunes better with a bigger loop. Start with a radiator that's too long and your gamma will have to be tuned with more capacitive reactance.

That is the nature of using a gamma match and it has been my experience too. In the past I found it much easier to set my beams a bit lower in height than desired---just so I could tune at installed height. That is the only way I was able to overcome some of this interaction. With a horizontal yagi, if you can get close to the boom you’re tuning is made less difficult. But, with the Vector you have to really be able to get up high, well above the feed point, in order to fix the gamma. Then you have to let it down to fix the radiator length. With this in mind, I never could figure out how Bob did all of those tuning iterations to his Vector up on an 80’ foot scaffold pole---unless maybe he had a full scaffold setup.

Tweaking a UHF Quagi for maximum gain is easier to do then this HF omni. In the interest of preventing others from wasting time coping my measurements on the "Modified Vector" I've made it clear I believe I made a mistake in it's tuning by not consistently comparing it against a reference antenna (stock version in this case) while I was tuning it. Something I do after each change on the FM version within minuets of the change.

Noted.

I think it's no coincidence that the intensively time consuming tests that Bob has done on 11 meters line up with my testing on FM showing radial lengths just over 1/4 wave puts more signal where it counts. Further confirmation of this can be found in Sirio's latest upgrades to the Vector.

Agreed! I think this whole modeling idea is what drove this thread to try and find out if Bob’s idea had merit and was provable in some way. When you fixed Henry’s file that you sent to Bob, which antenna did you take the measurements from, and what did you change?

I have modeled a conventional 3/4 wave vertical with four 90 degree radials and when the radials are moved this far away from the radiator it does indeed have a major impact on radiation angle. Maximum sidelobe gain goes from below 8 degrees to over 45 degrees with no power at all close to the horizon.

Could you email me that 3/4 wave GP file output that is similar to the Vector output you sent to Bob? The one he posted: edromans@comcast.net

I'm not sure where the exact change happens but it's clear that NEC shows small changes in radial angle from the stock Vector do not alter radiation angle. Big changes that approach horizontal radials have a very negative affect on the radiation angle in this design. Keep in mind the loop must be removed when the radials are spaced out too far but that in itself doesn't alter the antennas performance much once the radial length is compensated.

Well, I’m not sure I remember my exact question about where the change in phasing happens either, but I just made a post trying to raise the issue for how in-phase vs. out of phase was noted on Henry’s and your subsequent antenna view files. I also noted how phasing was shown on Dxer’s files and that was when I noticed the differences. I’m waiting to see if anyone can give us an explanation in that regard. SW, do you have a comment about this with your models?

DXers post does show NEC is reporting currents in the bottom of the radiator to be out of phase with the top when the phase option is on. That would suggest the antenna once again is no better then a half wave. I wouldn't bet the farm on this. Remember Dr. Cebik who wrote a book on NEC modeling commented that this design would be difficult to accurately model with NEC.

Again, what does your file(s) show in this regard, in phase or out of phase?

The issue of the bottom radiating or not has been on the table since Bob started talking about the Sigma4. I assume if the models show the antenna out of phase, then it is not radiating from the bottom 1/4 wave monopole or the radials. If that is so, then according to these models, the Sigma4 represented is simply a modified form of a J-Pole using a skeleton sleeve. If what you tongue-in-cheek suggest is true, “the antenna once again is no better than a half wave,” or a standard J-Pole, then the improvements must be due to the improved symmetry of the skeleton sleeve setup. If this is true then it will not support my idea for how it works either.

I have read the ARRL article on “The Open Sleeve Antenna,” and the other link “The Skeleton Sleeve Fed Monopole” that Bob refers us to. I find a lot of discussion on matching, impedance, use of different element to length ratios for different patterns, and symmetry, but I find little regarding possible collinear affects, or changing of the angle of radiation by steering. I also note that gain is discussed briefly at the very end of the article, but that comment “…for gains up to 3db over a 1/4 wave monopole” was supposing that the antennas were over a perfectly conducting ground plane. That old argument about 5/8 wave showing a 3db better gain than a 1/4 wave comes to mind and has been around for years. I agree it is true in theory, but only in theory are we able to realize an infinite conducting ground plane, Mother Earth has a way of making the 1/4, 1/2, and 5/8 wave all balance out a lot closer than a 2-3db difference.

I see NEC as a tool that helps you apply basic concepts to the antenna design. Where you can make simple changes and see how they interact with other variables in the design. If you think you can use it and a tape measure to make a perfect working model or measure and perfect actual gain, I doubt you'll get it right with the Sigma.

That may be SW, but I’ll wait until the Fat Lady dances before I make up my mind on modeling working or not. The ARRL AHB is not a Bible on antennas, but if you read the next to last paragraph in the article on open sleeves, you’ll see ARRL claims the it’s easy to model using Nec and Mininec. So, maybe Eznec is not a good modeling tool for the Sigma4.

 

[bob85]

eddie, i did my tuning on a 36ft pole in a field then raised it to 73ft, at this location i have tuned up at 2ft 12ft and 30+ft,
you would be surprised how a vector will tune with wildly differeing monople lengths, its not like tuning a groundplane,

you are correct about groundplanes needing a massive groundplane to realise the low takeoff angles if we are talking about ground mounted 1/4waves, the benefit of open sleeve is it can give low angles without the need for an extensive radial system extending many wavelengths from the antenna,

i don't have a definitive answer for the sigma style, im gleaning snippets of info from many sources none deal specifically with sigmas,

avanti acknowlege that the sigma radiates most of its signal in the upper 1/2wave which suggests there is some radiation from the lower tapered 1/4wave sleeve that contributes in a constructive way with radiation from the portion of monopole above the sleeve,
i would be surprised if changing the relative phase does not change radiation angle, ez-nec is clearly wrong when it reports a big drop in gain as you move from uniform diameter wire to tapered tubes,
problems with modelling is likely the reason cebik warned me about accurate models and as he put it, ( the pages of meaningless argument that style antenna can cause ),
if it was a simple as putting measurements in nec there would be little argument imho

 

[Marconi]

Bob, I'm hanging on for some positive results too. Maybe modeling will not or cannot answer these questions directly, but you gotta' know that much of what I say is in part just playing a devil's advocate.

 

[Shockwave]

I can understand the change with height affecting the angle of radiation, but in light of what Bob says about his own work and what he says Sirio claims about the reason for their recent change in the Vector, did you question your results when using Eznec to compare the modifications to the model you adjusted? Both Bob and Siro see the opposite of what you claim above.

Bob, Sirio And my FM version all agree more gain is apparent with longer then 1/4 wave radials. It was only when I tried to expand the angle on the 11 meter version that I messed up. In the field testing shows more signal on the horizon at a distance with longer radials. I agree with this and that my last modified Vector was wrong.



I understand that you tested using field test procedures only with your FM model and the 11 meter Modified Vector, that is why I was surprised when you changed your mind---that it worked better than a stock Vector. It seems to me that you changed your thinking when you received the Eznec model from Henry and not a new revelation in studying your field test results. Am I wrong?

Yes, you are wrong. NEC did not change my signal reports, time and propagation variables did. That forced me to take an average signal over a longer period of time too compare it to the old antenna.


If you remember, how did the change in the sleeve (shortening and widening the angel) affect the resonance of the antenna initially, before you re-tuned? Did you try to see where the new resonance was before re-tuning?

As with any antenna increasing the electrical wavelenge of the radiator or counterpoise tends to lower the wavelength frequency. What I've tried to explain unsuccessfully is that the lengths of the elements used in the Sigma design should be adjusted primarily for maximum far field signal and let the gamma fix resonance.

I ask this because I’m interested in understanding how such changes affect resonance. I recall the first three knockoff Sigma4’s that I bought for some buddies were from CTE, and they clearly stated on the box that the antenna was 31’> and was tunable from 10 thru 12 meters. Two of the antennas owner’s installed them as long as possible as there was very little installation information. Later, with all the subsequent bitching I finally got the story that they did not work worth a damn, having no ears is the way they were described. I was told later on by someone else at one of the installs that the owner insisted the antenna be made a long as possible, and that the antenna had some fluted (ribbed) tubing in the base of the antenna and he had never seen such before. Does that Vector sound familiar to you SW or anybody?

There have been lots of Sigma clones. I've seen the one with the cheap ribbed tubing and owned the gold one with no variable tap on the gamma. While some manufacturers may instruct you to set the radiator to 31 feet for every frequency between 10 and 12 meters, common sense tells us maximum gain can not be had with one radiator length across that spectrum.

I have never experimented with my Sigma, but if I’m right about the bottom quarter wave section’s pattern being compressed by the contributing affects of the upper 1/2 wave section, then I believe adjusting the radial length must have a very strong affect on resonance as well, and making the radials longer will require you to shorten the radiator---just like Sirio did with their New Vector. Both you and Bob, on the other hand, claim to have made your radiators up to 3’ feet longer and fixing the resonance by adjusting the gamma.


You both may be right on the mark here, but it seems like a real stretch to make a tune for the gamma matcher to me. I can understand a few inches of adjustments, but several feet of tip and who knows what adjustment to the gamma leaves a big question in my mind. I have a gamma off of a LW 150, that I sold and it is shorter overall than my AS Sigma4, but the LW 150, base is longer @ 13" vs. 9" inches for the S4. It may be enough to tune for you guys, but that doesn't leave much room for adjustment. I’m waiting for Homer to make his adjustments to see what he finds. Hopefully he will explore these resonance affects.

Again, with this antenna you have to pay less attention to resonance and more to gain when adjusting lengths. The gamma will compensate for resonance very easily. You really need to experiment to see how the antenna is different then something like a center fed gamma match half wave element. This antenna works best when you tune all elements for maximum far field signal while consistently readjusting the gamma to maintain resonance.



I understand this and it’s plausible. After I did my work last summer, I was too burned out to really hash-out my notes, and after doing so, months later, I had revelations that I did not see at the time of testing.



That is about the same time as my testing here, and I was seeing the same strong variables in signals. Bob will tell you how I was complaining last summer, and it was not due to DX interference as a rule. But here again, you were making the antenna as long as possible and yet you kept exploring the effects higher in frequency.

I didn't make the antenna longer. In fact, I said it was shorter then the stock 31 foot Vector and that I made the loop larger to experiment with wider radial angles.



That is the nature of using a gamma match and it has been my experience too. In the past I found it much easier to set my beams a bit lower in height than desired---just so I could tune at installed height. That is the only way I was able to overcome some of this interaction. With a horizontal yagi, if you can get close to the boom you’re tuning is made less difficult. But, with the Vector you have to really be able to get up high, well above the feed point, in order to fix the gamma. Then you have to let it down to fix the radiator length. With this in mind, I never could figure out how Bob did all of those tuning iterations to his Vector up on an 80’ foot scaffold pole---unless maybe he had a full scaffold setup.



Noted.



Agreed! I think this whole modeling idea is what drove this thread to try and find out if Bob’s idea had merit and was provable in some way. When you fixed Henry’s file that you sent to Bob, which antenna did you take the measurements from, and what did you change?

I just added a few feet to the radiator on Henry's file because it was obviously shorter then all of the real antennas. I've mentioned before that NEC is forcing us to model the antenna shorter then real world because it's not compensating for the gamma match. That makes a huge difference in what you can do with this antenna.



Could you email me that 3/4 wave GP file output that is similar to the Vector output you sent to Bob? The one he posted: edromans@comcast.net

I'll look for the file but I'm not sure if I saved it. I think I just increased the length of an existing 1/4 wave antenna in the program files. The pattern is just as I described. No power near the horizon, tons around 45 degrees.



Well, I’m not sure I remember my exact question about where the change in phasing happens either, but I just made a post trying to raise the issue for how in-phase vs. out of phase was noted on Henry’s and your subsequent antenna view files. I also noted how phasing was shown on Dxer’s files and that was when I noticed the differences. I’m waiting to see if anyone can give us an explanation in that regard. SW, do you have a comment about this with your models?

My comment would be in agreement with Dr. Cebik in that accurate modeling of this design in NEC would be difficult. I don't see myself gaining more knowledge with NEC then he possessed anytime soon. The more I play with the design in NEC I confirm it's very difficult to get the model to perform like the real antenna. Less modeling and more hands on.



Again, what does your file(s) show in this regard, in phase or out of phase?

The issue of the bottom radiating or not has been on the table since Bob started talking about the Sigma4. I assume if the models show the antenna out of phase, then it is not radiating from the bottom 1/4 wave monopole or the radials. If that is so, then according to these models, the Sigma4 represented is simply a modified form of a J-Pole using a skeleton sleeve. If what you tongue-in-cheek suggest is true, “the antenna once again is no better than a half wave,” or a standard J-Pole, then the improvements must be due to the improved symmetry of the skeleton sleeve setup. If this is true then it will not support my idea for how it works either.

I have read the ARRL article on “The Open Sleeve Antenna,” and the other link “The Skeleton Sleeve Fed Monopole” that Bob refers us to. I find a lot of discussion on matching, impedance, use of different element to length ratios for different patterns, and symmetry, but I find little regarding possible collinear affects, or changing of the angle of radiation by steering. I also note that gain is discussed briefly at the very end of the article, but that comment “…for gains up to 3db over a 1/4 wave monopole” was supposing that the antennas were over a perfectly conducting ground plane. That old argument about 5/8 wave showing a 3db better gain than a 1/4 wave comes to mind and has been around for years. I agree it is true in theory, but only in theory are we able to realize an infinite conducting ground plane, Mother Earth has a way of making the 1/4, 1/2, and 5/8 wave all balance out a lot closer than a 2-3db difference.

My files show the same as everyone else. When the phase option is on it indicates the lower and upper portions of the radiator are not in phase. That means that NEC sees the antenna as having similar gain to a half wave. It also supports the age old thought that no vertical beyond a 5/8 wave will put gain on the horizon. The only problem here is that all the books and models that support this are completely wrong when you apply them to the Sigma design.

None of the mentioned antennas are like the Sigma because no other antenna is like it. The key difference is the feed system on the Sigma is not a simple transmission line with parallel conductors that don't radiate. The tapered angle of the radials prevents complete cancellation of the radiation. Folding the radials upwards at an angle to the radiator absolutely puts more gain on the horizon.



That may be SW, but I’ll wait until the Fat Lady dances before I make up my mind on modeling working or not. The ARRL AHB is not a Bible on antennas, but if you read the next to last paragraph in the article on open sleeves, you’ll see ARRL claims the it’s easy to model using Nec and Mininec. So, maybe Eznec is not a good modeling tool for the Sigma4.

It's not a particular version of NEC that has difficulty modeling this antenna, it's NEC in general. Parallel sleeve antennas are easy to model with predictable gains that line up with field testing. The tapered base on the Sigma has affects that no one seems to explain well. It beats the 5/8 wave by 1 db and is not nearly as dependant on good ground conductivity to provide gain.

 

[Marconi]

Thanks Shockwave for all your patience. I think you're right about Eznec. I wish I understood more.

 

[Shockwave]

No problem Marconi. I was also hoping NEC would help us to understand how this design worked in more detail. That seems to be in question now just as experts have said in the past. Fully explaining this antennas theory of operation seems like an impossible task. Fortunately confirming it's ability to outperform every single section vertical known to exist can be proven in the field much easier. This I would bet the farm on.

 

[Marconi]

Shockwave and Bob did you notice this in Dxer's Models?

Shockwave suggested that Dxer fix the Vector so the current minimum between the 1/4 wave and the 1/2 wave elements be located at the end of the radials @ the hoop. Below are the images that show this change.

If we look closely we will see that the effort did not benefit the Vector in gain or angle. In fact the gain went down. Admittedly the antenna was made about 1.5' foot shorter to accomplish the change. Maybe if some other adjustment had been made, may the affect would have kept the current null in place without lowering the gain.

Shockwave, Dxer offered his Vector file to you. If you got that file maybe you could make the adjustment to the radial length instead of the radiator, keeping the radiator at 26.6' feet, and see if the gain would return or get better. I grant the amount of change resulted is a very small change, so maybe fixing is not worth the effort.

Shockwave if this is the way it happened, what was the point of the fix to make the current null located closer to the ends of the radials and how did you know that could be important?

It is also interesting to compare the two Vectors against the Sigma4 that Dxer showed with his composite view of the RF patterns in the last image below.

Vector 25.277' notice the current minimum is right at the hoop.


Vector 25.0277'


Vector 26.6' notice the current minimum is well above the hoop.


Vector 26.6'


Two color coded Vectors and a Sigma4 pattern in black overlaid for comparison.


It should be noted that the two Vectors show a nice gain reduction in the high elevation lobes compared to the original Sigma4 that Dxer modeled. Overall it looks like the longest Vector shows a slight advantage in maximum gain. I will note however, that these Vectors are not based on known dimensions if it matters. IMO, the diameters for the Vectors are larger than the physical antenna. Note that Dxer indicates that this caused his Vector models to have a top element that was .625" in diameter instead of the .50" diameter tip that is on the stock antenna. So the rest of the antenna's diameter is larger than the stock is physically, all the way down to the base element.

 

[Henry HPSD]

Hello Guys,

I have been in touch with the designer of Eznec (Roy lewallen)
With the adjustments (feedpoint/segmentation) there is no reason why eznec wouldnt predict accurate results.

The antenna "specialist" seem to confirm it concernce a skelton monopole as Bob earlier stated.

With that in mind i can only think of 2 reasons why it would primarily outperform or work less than any other type of antenna inthe lines of:

1-) Heigth..the antenna is always physical higher and can see futher on the horizon..
2-) Different t.o.a. in comparisment to others.. this can be a plus or downside.


I do not know how the "field" gain test were done..But there are many variable factors
As when i "measure" gain i alway put a second receiving antenna on the ground a couple wl away from the transmitting antenna to "catch" the reflected waves.
Without one i have had different readings. now i am not saying this is the way to go..
I just want to point out "real" test are very difficult as there are so many things to bear in mind.

I do understand that some may have "optimal" results as of years of "testing" etc..
Although this is ofcourse very valuable for that person, I however question if this would be the case at other locations as "everything" changes.
And that brings me back towards modeling..with that information you can "predict" what would happen in most cases.


PS for those wondering what happend to the mail i send to solarcom (ground radials on the imax...) i got nothing returned...

73 Henry

 

[Marconi]

Hello Guys,

I have been in touch with the designer of Eznec (Roy lewallen)
With the adjustments (feedpoint/segmentation) there is no reason why eznec wouldnt predict accurate results.

The antenna "specialist" seem to confirm it concernce a skelton monopole as Bob earlier stated.

With that in mind i can only think of 2 reasons why it would primarily outperform or work less than any other type of antenna inthe lines of:

1-) Heigth..the antenna is always physical higher and can see futher on the horizon..
2-) Different t.o.a. in comparisment to others.. this can be a plus or downside.


I do not know how the "field" gain test were done..But there are many variable factors
As when i "measure" gain i alway put a second receiving antenna on the ground a couple wl away from the transmitting antenna to "catch" the reflected waves.
Without one i have had different readings. now i am not saying this is the way to go..
I just want to point out "real" test are very difficult as there are so many things to bear in mind.

I do understand that some may have "optimal" results as of years of "testing" etc..
Although this is ofcourse very valuable for that person, I however question if this would be the case at other locations as "everything" changes.
And that brings me back towards modeling..with that information you can "predict" what would happen in most cases.


PS for those wondering what happend to the mail i send to solarcom (ground radials on the imax...) i got nothing returned...

73 Henry

Hello Henry. I wouldn't expect Mr. Lewallen to be anything but subjective in his response about his own software, unless he had actually modeled and made comparisons for the Vector/Sigma4 that was modified.

We’re all aware of how subjective field testing can be. Even so, we do the best we can while realizing what we determine really only applies to the location for the test being made, and at best that is no where near perfect.

You told us that you modeled using thin wires out of convenience. I have no problem with that, if the use of wires vs. correct tubing diameter and taper---does not ill-affect the results too much either way. It would be nice if you could model using thin wires and not have to mess with thicker tubing and taper. Am I assuming correctly, that when you used the thin wires rather than tubing, that the element lengths all ended up being short, unless they were adjusted for the same resonance as the tubing and tapered elements?

I don’t know which Vector you used to approximate your measurements, but you told me the radiator was about 24’ feet long using thin wire. That is considerably shorter than my Sigma 4 and for sure shorter still compared to the original Vector 4000, yet your model shows resonance low in frequency to 26.950 mhz. I don’t get that, because in that situation the resonance should have gone up using shorter and thinner wires. On the other hand you indicate the radials were 108” long which is much longer than my Sigma4 @ 90.5” and maybe longer than the older version of the Vector that Shockwave may have used. So I ask, in your opinion is it possible that frequency is primarily affected by a change in resonant length of the radials and less so when adjusting the radiator length? I think Bob tells a story about his Vector, where the tip collapsed down into the next section and it didn’t seem to make a difference.

I also read that SW indicates he made changes which are noted to have raised the resonance to 27.200 mhz with his adjustments. If I measured SW’s wires chart, that Bob posted, correctly, he made the radiator longer to 25.26’ feet using the same wire diameters. This should have lowered the frequency and that is not what is indicated.

SW’s chart also indicates he made the radials 3” inches shorter @ 105” compared to your radials @ 108”. Here again, we see this tending to lower the frequency, as it should, even while making the radiator longer. Does this also suggest that the resonance for this antenna design is primarily affected by the length of the radials and not the top element? Homer questioned all this a while back, so maybe he can shed some light on this idea if he got to do his modifications to his homemade Vector. Homer, how say you?

Henry you also told us that using a square radial hoop vs. a circle made no difference, but have you checked to see if using thin wires that were or were not adjusted for resonance vs. tapered tubing---makes any difference in the model results that are worth noting?

How did you get your model to resonance taking measurements from an antenna with thick tubing without converting for thinner wires, were you just lucky?

In the text of your Vector file it notes the impedance being easy to match with a gamma match, so I assume you did not included a matcher at the bottom of the model. Your bandwidth chart indicates ZO impedance at 34 ohms with a very low SWR. If that is the case then the antenna is very close to a resonant 1/4 wave with a suitable ground plane and IMP that is what we might expect for a 3/4 wave radiator. I agree this makes the antenna perfectly suitable for a gamma device. What would you think happened to the value of Zo after SW changed the length of the radials and the radiator?


Can you describe where the source is that shows this Zo impedance on your original model? I assume it has to be at some point above the hub on the first element where we might expect to see the gamma tap point. I see a small bump a little above the hub. Is that the feed point, and how does that get there? I don't see it noted in the wires chart.

Thanks for your work.