Loading ...
Sorry, an error occurred while loading the content.

I want a hex beam

Expand Messages
  • kt6gradio
    Hi fellow amateur radio operators, I am convinced that the hex beam will replace my dipole. So this summer, when I making some money again, I plan on
    Message 1 of 27 , Jan 31, 2010
    • 0 Attachment
      Hi fellow amateur radio operators,

      I am convinced that the hex beam will replace my dipole. So this summer, when I making some money again, I plan on purchasing a multi band hex beam antenna. I plan on mounting it on a 54 foot crank up tower and have the room and no HOA or CCR to contend with.

      I do have a few questions,

      1. Has anyone ever mounted two hex beams parallel, on say a 30 foot beam and run them in phase? would that even work? Would out of phase work better?

      2. With the spacing between elements being restricted by the length of the spreaders, would designing a spacer that was stiff (aluminum?) that the fiberglass connects to enable the spacing to be increased between elements?

      3. Since the feed line for all bands are interconnected and there is no null or isolation between them, would developing a cutoff/relay switch for each individual band enable better response on each band? Basically isolating each band?

      4. With the above question in mind, would adding another wire for the same band, parallel to the single wire, like adding a second 20m wire, increase the s/n, swr, etc? or maybe building a sort of net like structure have any benefit? anyone try something like this?

      5. What other materials, besides fiberglass, have been suggested for the spreaders? Just curious here.

      6. What type of coax is used for enabling the antenna to turn without becoming tangled or broken? RG 213, RG 8, LMR 400? is there a best coax for this?

      7. Is there one thing that everyone that owns a hex beam wish they had or was built to make the antenna better? Really fishing here ;)


      I plan on experimenting with the hex beam, as this would garner the most fun for me and my amateur radio hobby. I thank you for reading and hope you did not fall asleep. Any response is appreciated and will keep everyone informed when I get going on this project.Thanks again.

      L.Paul Austin KT6G
    • Don Keith
      Paul, I think you will be pleased with the hex, and with the possibilities of experimentation.  I do have a few comments on your questions, based on limited
      Message 2 of 27 , Feb 2, 2010
      • 0 Attachment
        Paul, I think you will be pleased with the hex, and with the possibilities of experimentation.  I do have a few comments on your questions, based on limited knowledge but my own experiences building a broadband G3TXQ/K4KIO version from scratch.
         
        By the way, if you want to read about my hex project, see the article I wrote for eHam.net at: http://www.eham.net/articles/20575%c2%a0or at my web site: www.n4kc.com.
         
        1. I'd think it would be very difficult if not impossible trying to phase two hexbeams.  Some do advocate having two at different heights to take advantage of take-off angles, but you'd need one at about 50 feet and the other at about 90 feet to see much difference, I'd bet.
         
        2. The spacer dimensions are critical to the performance of the antenna, and you would never want to use anything conductive for a spacer.
         
        3. I see no advantage to isloating the feed to individual elements, and it unecessarily complicates a simple design.
         
        4. There are two elements for each band already...a driven element and a reflector.  Search the archives on this reflector.  There has been chatter about a 3-element version.
         
        5. Hard to beat fiberglass for spreaders.  Some use fishing poles, and that might work for versions with fewer bands.
         
        6. Any coax will work.  As with any rotatable antenna, you just leave enough loop to allow free turning 360 degrees.
         
        7. To me the most important part is the base plate.  It needs to be strong enough and drilled accurately enough that the spreaders are evenly arrayed and firmly held in place.  Some use plywood or cutting board plastic, but I just think you are asking for trouble.  I splurged and bought mine from one of the several places that offer it, and I have not regretted it.  Also read some of the sites--like K4KIO's--about the various methods of building the jumpers that run between the feed points on each set of elements.  Since you are using short pieces of coax and a bunch of connection points for a 5-band antenna, there are plenty of potential failure points.  Remember, if you feed from the top, as recommended, and you have the first jumper between the 20-meter and 17-meter elements fail, you lose all elements 17 through 10.
         
        Paul, good luck and let us know how you progress.  Mine's been up for going on two years and still working beautifully...and I know the idiot that built it, so that's amazing!
         
        73,
         
        Don Keith N4KC
        www.n4kc.com
        www.donkeith.com
         
         

        Hi fellow amateur radio operators,

        I am convinced that the hex beam will replace my dipole. So this summer, when I making some money again, I plan on purchasing a multi band hex beam antenna. I plan on mounting it on a 54 foot crank up tower and have the room and no HOA or CCR to contend with.

        I do have a few questions,

        1. Has anyone ever mounted two hex beams parallel, on say a 30 foot beam and run them in phase? would that even work? Would out of phase work better?

        2. With the spacing between elements being restricted by the length of the spreaders, would designing a spacer that was stiff (aluminum?) that the fiberglass connects to enable the spacing to be increased between elements?

        3. Since the feed line for all bands are interconnected and there is no null or isolation between them, would developing a cutoff/relay switch for each individual band enable better response on each band? Basically isolating each band?

        4. With the above question in mind, would adding another wire for the same band, parallel to the single wire, like adding a second 20m wire, increase the s/n, swr, etc? or maybe building a sort of net like structure have any benefit? anyone try something like this?

        5. What other materials, besides fiberglass, have been suggested for the spreaders? Just curious here.

        6. What type of coax is used for enabling the antenna to turn without becoming tangled or broken? RG 213, RG 8, LMR 400? is there a best coax for this?

        7. Is there one thing that everyone that owns a hex beam wish they had or was built to make the antenna better? Really fishing here ;)

        I plan on experimenting with the hex beam, as this would garner the most fun for me and my amateur radio hobby. I thank you for reading and hope you did not fall asleep. Any response is appreciated and will keep everyone informed when I get going on this project.Thanks again.

        L.Paul Austin KT6G



        [Non-text portions of this message have been removed]
      • Steve
        I have a few observations on some of the questions: 1)In principle you can phase two hexbeams just like you can any pair of identical antennas. But if you have
        Message 3 of 27 , Feb 2, 2010
        • 0 Attachment
          I have a few observations on some of the questions:

          1)In principle you can phase two hexbeams just like you can any pair of identical antennas. But if you have that much space and hardware available there may be easier ways of getting 2dB-3dB extra gain.

          2) Yes, you could increase the spacing by making the spreaders longer and get some small improvement in Gain; at each step the turn radius goes up, and eventually it turns into a full-sized Yagi or perhaps a Moxon. Most people are happy with the performance/size trade-offs of the conventional hexbeam or broadband hexbeam designs.

          3) There is no advantage in feeding the elements of a 5-band 20m thru 10m hexbeam separately. In fact the common feedpoint actually has some benefits in lowering the VSWR.

          4) I'm not sure what you mean by adding an extra wire. If you mean turning it into a 3-element array, there is a design here:
          http://www.karinya.net/g3txq/hexbeam/3el/
          You'll see that the turn radius has increased by 50%

          Adding a third element without significantly increasing the turn radius achieves little. The hexbeam is a parasitic array working on similar principles to a conventional Yagi. If you add extra elements to a 2-el Yagi without increasing the boom length, you'll see little improvement in its performance. It's the same with the hexbeam - you only see the full benefit of a third element once you let the array size increase significantly.

          73,
          Steve G3TXQ
        • BG
          Hello Paul, It sounds as if you ve done considerable research and settled in on a great antenna selection for your station. I have a few comments on your
          Message 4 of 27 , Feb 4, 2010
          • 0 Attachment
            Hello Paul,



            It sounds as if you've done considerable research and settled in on a great
            antenna selection for your station. I have a few comments on your question
            #4, and for the rest I will say, in general, that it sounds like you would
            be venturing into unchartered and untested territory with most of those
            ideas. For your question #2, I am not yet convinced that an aluminum center
            post could not be employed, and have considered trying that myself, but have
            not looked into it seriously. I would also be interested in knowing if
            anyone else has tried that.



            There has been considerable work done with a 3 element folded wire array,
            similar to a hexbeam in construction, with improved performance and
            construction components. I have to disagree with Steve's comments regarding
            a 3 element folded wire array, and I have posted some supporting data in the
            files section (OctoBeam.pdf). The OctoBeam (http://www.
            <http://www.OctoBeam.com> OctoBeam.com) has an 11 foot turn radius and an
            equal footprint when compared to the hexbeam, but it is slightly heavier.
            The OctoBeam exhibits optimum performance when the director element is as
            close as possible to the driven element. Increasing the separation distance
            between those 2 elements, as you would with a Yagi or a 3 element hexbeam,
            reduces the interaction and performance. The OctoBeam is manufactured using
            modern machine tools by highly skilled machinists and every antenna is
            inspected pursuant to a strict quality control program. Although a home
            brew version of this antenna is possible using the OctoBeam theory and
            geometry the construction and component quality would likely be reduced.



            It sounds like you intend to experiment, so have a lot fun with whatever
            antenna you deploy. Keep in mind that the OctoBeam does offer a lot of
            flexibility to enable customization for your Forward Gain and F/B choices by
            simple rearrangement of all 3 element geometries.



            73!



            Bill, KO6HL

            _____

            From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
            Of kt6gradio
            Sent: Sunday, January 31, 2010 10:28 PM
            To: hex-beam@yahoogroups.com
            Subject: [hex-beam] I want a hex beam



            Hi fellow amateur radio operators,

            I am convinced that the hex beam will replace my dipole. So this summer,
            when I making some money again, I plan on purchasing a multi band hex beam
            antenna. I plan on mounting it on a 54 foot crank up tower and have the room
            and no HOA or CCR to contend with.

            I do have a few questions,

            1. Has anyone ever mounted two hex beams parallel, on say a 30 foot beam and
            run them in phase? would that even work? Would out of phase work better?

            2. With the spacing between elements being restricted by the length of the
            spreaders, would designing a spacer that was stiff (aluminum?) that the
            fiberglass connects to enable the spacing to be increased between elements?

            3. Since the feed line for all bands are interconnected and there is no null
            or isolation between them, would developing a cutoff/relay switch for each
            individual band enable better response on each band? Basically isolating
            each band?

            4. With the above question in mind, would adding another wire for the same
            band, parallel to the single wire, like adding a second 20m wire, increase
            the s/n, swr, etc? or maybe building a sort of net like structure have any
            benefit? anyone try something like this?

            5. What other materials, besides fiberglass, have been suggested for the
            spreaders? Just curious here.

            6. What type of coax is used for enabling the antenna to turn without
            becoming tangled or broken? RG 213, RG 8, LMR 400? is there a best coax for
            this?

            7. Is there one thing that everyone that owns a hex beam wish they had or
            was built to make the antenna better? Really fishing here ;)

            I plan on experimenting with the hex beam, as this would garner the most fun
            for me and my amateur radio hobby. I thank you for reading and hope you did
            not fall asleep. Any response is appreciated and will keep everyone informed
            when I get going on this project.Thanks again.

            L.Paul Austin KT6G



            No virus found in this incoming message.
            Checked by AVG - www.avg.com
            Version: 9.0.733 / Virus Database: 271.1.1/2662 - Release Date: 02/01/10
            04:37:00



            [Non-text portions of this message have been removed]
          • Steve
            Bill, Perhaps I should have chosen my words more carefully to avoid being misunderstood :) My point was that adding extra elements to an array without
            Message 5 of 27 , Feb 4, 2010
            • 0 Attachment
              Bill,

              Perhaps I should have chosen my words more carefully to avoid being misunderstood :)

              My point was that adding extra elements to an array without increasing its size adds little to its **overall** performance. I've yet to see any evidence to the contrary, but if you have some I'm happy to be corrected.

              Any antenna design is a compromise, and with parasitic arrays it's generally the case that you can trade Forward Gain for F/B ratio **if you are constrained to a particular footprint**. The designer chooses a balance between the two which he feels is appropriate. The figures you published show a laudable Forward Gain, but the F/B has been compromised severely - 9.5dB compared to a 2-el hexbeam's 24dB.

              I maintain that the full benefit of a third element can be realised only by allowing the array to "grow". The 3-element design I published on my web site achieves the same Forward Gain as the Octobeam, but with 20dB F/B, and that without any serious attempt at optimisation. By comparison you would expect a full-sized 3-el Yagi to deliver 8dBi Forward Gain with a F/B of at least 30dB.

              Please don't think I'm "knocking" the Octobeam; it seems to offer another option for those who want extra Forward Gain at the cost of reduced F/B. But there will always need to be that trade-off unless the footprint increases.

              73,
              Steve G3TXQ
            • yash@aol.com
              Bill I took my center post and wrapped it in aluminum foil .Messed up the swr on some bands and others it did not. The foil was wrapped around the post,un
              Message 6 of 27 , Feb 4, 2010
              • 0 Attachment
                Bill
                I took my center post and wrapped it in aluminum foil .Messed up the swr on some bands and others it did not. The foil was wrapped around the post,un grounded and also grounded same result. I think that with a bit of tuning it could work.You could model it I suppose. My center post is carbon fiber 1 1/2" plenty strong enought.
                dale wt4t






                -----Original Message-----
                From: BG <ko6hl@...>
                To: hex-beam@yahoogroups.com
                Sent: Thu, Feb 4, 2010 1:49 pm
                Subject: RE: [hex-beam] I want a hex beam





                Hello Paul,

                It sounds as if you've done considerable research and settled in on a great
                antenna selection for your station. I have a few comments on your question
                #4, and for the rest I will say, in general, that it sounds like you would
                be venturing into unchartered and untested territory with most of those
                ideas. For your question #2, I am not yet convinced that an aluminum center
                post could not be employed, and have considered trying that myself, but have
                not looked into it seriously. I would also be interested in knowing if
                anyone else has tried that.

                There has been considerable work done with a 3 element folded wire array,
                similar to a hexbeam in construction, with improved performance and
                construction components. I have to disagree with Steve's comments regarding
                a 3 element folded wire array, and I have posted some supporting data in the
                files section (OctoBeam.pdf). The OctoBeam (http://www.
                <http://www.OctoBeam.com> OctoBeam.com) has an 11 foot turn radius and an
                equal footprint when compared to the hexbeam, but it is slightly heavier.
                The OctoBeam exhibits optimum performance when the director element is as
                close as possible to the driven element. Increasing the separation distance
                between those 2 elements, as you would with a Yagi or a 3 element hexbeam,
                reduces the interaction and performance. The OctoBeam is manufactured using
                modern machine tools by highly skilled machinists and every antenna is
                inspected pursuant to a strict quality control program. Although a home
                brew version of this antenna is possible using the OctoBeam theory and
                geometry the construction and component quality would likely be reduced.

                It sounds like you intend to experiment, so have a lot fun with whatever
                antenna you deploy. Keep in mind that the OctoBeam does offer a lot of
                flexibility to enable customization for your Forward Gain and F/B choices by
                simple rearrangement of all 3 element geometries.

                73!

                Bill, KO6HL

                _____

                From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                Of kt6gradio
                Sent: Sunday, January 31, 2010 10:28 PM
                To: hex-beam@yahoogroups.com
                Subject: [hex-beam] I want a hex beam

                Hi fellow amateur radio operators,

                I am convinced that the hex beam will replace my dipole. So this summer,
                when I making some money again, I plan on purchasing a multi band hex beam
                antenna. I plan on mounting it on a 54 foot crank up tower and have the room
                and no HOA or CCR to contend with.

                I do have a few questions,

                1. Has anyone ever mounted two hex beams parallel, on say a 30 foot beam and
                run them in phase? would that even work? Would out of phase work better?

                2. With the spacing between elements being restricted by the length of the
                spreaders, would designing a spacer that was stiff (aluminum?) that the
                fiberglass connects to enable the spacing to be increased between elements?

                3. Since the feed line for all bands are interconnected and there is no null
                or isolation between them, would developing a cutoff/relay switch for each
                individual band enable better response on each band? Basically isolating
                each band?

                4. With the above question in mind, would adding another wire for the same
                band, parallel to the single wire, like adding a second 20m wire, increase
                the s/n, swr, etc? or maybe building a sort of net like structure have any
                benefit? anyone try something like this?

                5. What other materials, besides fiberglass, have been suggested for the
                spreaders? Just curious here.

                6. What type of coax is used for enabling the antenna to turn without
                becoming tangled or broken? RG 213, RG 8, LMR 400? is there a best coax for
                this?

                7. Is there one thing that everyone that owns a hex beam wish they had or
                was built to make the antenna better? Really fishing here ;)

                I plan on experimenting with the hex beam, as this would garner the most fun
                for me and my amateur radio hobby. I thank you for reading and hope you did
                not fall asleep. Any response is appreciated and will keep everyone informed
                when I get going on this project.Thanks again.

                L.Paul Austin KT6G

                No virus found in this incoming message.
                Checked by AVG - www.avg.com
                Version: 9.0.733 / Virus Database: 271.1.1/2662 - Release Date: 02/01/10
                04:37:00

                [Non-text portions of this message have been removed]









                [Non-text portions of this message have been removed]
              • Steve
                Bill, With respect, it s rather simplistic to just move the Director forward and, because the performance drops, claim it proves bigger arrays don t always
                Message 7 of 27 , Feb 4, 2010
                • 0 Attachment
                  Bill,

                  With respect, it's rather simplistic to just move the Director forward and, because the performance drops, claim it proves bigger arrays don't always exhibit improved performance.

                  I just did something similar with my Octobeam EZNEC model. I moved the Director forward by 34"; but I then did some optimisation of the wires within that increased turn radius. The result was a Forward Gain of 7.8dBi and a F/B of 18.47dB. I only spent a few minutes optimising it, so there may have been more performance to come. The EZNEC plot is here:

                  http://www.karinya.net/g3txq/temp/octobeam_azimuth.png

                  Compare that performance with your "optimum position" figures of 7.47dBi Gain and 9.43dB F/B and you'll see that the bigger array does produce the better overall performance. It doesn't matter whether it's a Yagi, a Hexbeam, or an Octobeam, they all obey the same rules.

                  73,
                  Steve G3TXQ
                • BG
                  Steve, The model I used to get the results I published was not optimized. I simply wanted to demonstrate that by just moving the director reduced the
                  Message 8 of 27 , Feb 4, 2010
                  • 0 Attachment
                    Steve,



                    The model I used to get the results I published was not optimized. I simply
                    wanted to demonstrate that by just moving the director reduced the
                    performance for that configuration. I have other models that I used to get
                    results similar to yours with the director at the original design position.
                    Perhaps you would like to try optimizing your model before moving and then
                    optimizing again after the move. You should get results similar to mine.



                    Thanks,



                    Bill, KO6HL



                    _____

                    From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                    Of Steve
                    Sent: Thursday, February 04, 2010 3:04 PM
                    To: hex-beam@yahoogroups.com
                    Subject: [hex-beam] Re: I want a hex beam





                    Bill,

                    With respect, it's rather simplistic to just move the Director forward and,
                    because the performance drops, claim it proves bigger arrays don't always
                    exhibit improved performance.

                    I just did something similar with my Octobeam EZNEC model. I moved the
                    Director forward by 34"; but I then did some optimisation of the wires
                    within that increased turn radius. The result was a Forward Gain of 7.8dBi
                    and a F/B of 18.47dB. I only spent a few minutes optimising it, so there may
                    have been more performance to come. The EZNEC plot is here:

                    http://www.karinya <http://www.karinya.net/g3txq/temp/octobeam_azimuth.png>
                    net/g3txq/temp/octobeam_azimuth.png

                    Compare that performance with your "optimum position" figures of 7.47dBi
                    Gain and 9.43dB F/B and you'll see that the bigger array does produce the
                    better overall performance. It doesn't matter whether it's a Yagi, a
                    Hexbeam, or an Octobeam, they all obey the same rules.

                    73,
                    Steve G3TXQ



                    No virus found in this incoming message.
                    Checked by AVG - www.avg.com
                    Version: 9.0.733 / Virus Database: 271.1.1/2667 - Release Date: 02/03/10
                    23:35:00




                    [Non-text portions of this message have been removed]
                  • Steve
                    Bill, I m slightly confused now! Are you saying that you have achieved the same results I just published, with the Director back at the original position -
                    Message 9 of 27 , Feb 4, 2010
                    • 0 Attachment
                      Bill,

                      I'm slightly confused now! Are you saying that you have achieved the same results I just published, with the Director back at the "original position" - i.e. within the 11ft turn radius? if so I am impressed and confounded at the same time ;)

                      I did exactly what you just suggested: I had an optimised design; I increased the turn radius; I re-optimised the design, and I then had better performance figures.

                      To increase the size and not re-optimize demonstrates little except that a sub-optimum larger array may perform worse than an optimised smaller array; but I guess we already knew that.

                      My basic points remain: a bigger optimised array will have better overall performance than a smaller optimised array. And overall performance is much more dependent on the array size than it is on the number of elements.

                      Nearly midnight here, so forgive me if I'm not able to continue the discussion.

                      73,
                      Steve G3TXQ
                    • BG
                      Thanks Steve. I think we are saying the same thing. Except I didn t say I got the same results, I said I got similar results. You can view them here:
                      Message 10 of 27 , Feb 4, 2010
                      • 0 Attachment
                        Thanks Steve. I think we are saying the same thing. Except I didn't say I
                        got the same results, I said I got similar results. You can view them here:



                        http://www.norcalmfg.com/tmp/OB-20m-0DG-2.jpg



                        Every antenna is a compromise and an optimized OctoBeam with an 11 ft turn
                        radius is no exception. The exceptional characteristic to note about the
                        OctoBeam that makes it different from a conventional Yagi is that the
                        director element is excited by the driven element in much the same way the
                        reflector element is excited; at the ends.



                        Now, for fun, lets move the hexbeam driven element away from the reflector,
                        re-optimize and compare that to an optimized 2 element Yagi. ;-)



                        73!



                        Bill, KO6HL



                        _____

                        From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                        Of Steve
                        Sent: Thursday, February 04, 2010 3:45 PM
                        To: hex-beam@yahoogroups.com
                        Subject: [hex-beam] Re: I want a hex beam





                        Bill,

                        I'm slightly confused now! Are you saying that you have achieved the same
                        results I just published, with the Director back at the "original position"
                        - i.e. within the 11ft turn radius? if so I am impressed and confounded at
                        the same time ;)

                        I did exactly what you just suggested: I had an optimised design; I
                        increased the turn radius; I re-optimised the design, and I then had better
                        performance figures.

                        To increase the size and not re-optimize demonstrates little except that a
                        sub-optimum larger array may perform worse than an optimised smaller array;
                        but I guess we already knew that.

                        My basic points remain: a bigger optimised array will have better overall
                        performance than a smaller optimised array. And overall performance is much
                        more dependent on the array size than it is on the number of elements.

                        Nearly midnight here, so forgive me if I'm not able to continue the
                        discussion.

                        73,
                        Steve G3TXQ



                        No virus found in this incoming message.
                        Checked by AVG - www.avg.com
                        Version: 9.0.733 / Virus Database: 271.1.1/2667 - Release Date: 02/03/10
                        23:35:00




                        [Non-text portions of this message have been removed]
                      • Steve
                        Bill, We may be saying the same thing ;) Your figures illustrate well the design trade-offs: on the 3-element Octobeam the F/B has improved from 9.43dB to
                        Message 11 of 27 , Feb 5, 2010
                        • 0 Attachment
                          Bill,

                          We may be saying the same thing ;)

                          Your figures illustrate well the design trade-offs: on the 3-element Octobeam the F/B has improved from 9.43dB to 18.87dB, but the cost has been a loss of 1dB in Forward gain.

                          The other trade-off we haven't discussed is performance bandwidth. I wonder how you find the Gain and F/B hold up over the full band? I always felt that the move back to a "W" shape Reflector might mean some compromise in bandwidth.

                          My original comment, which you challenged, was that adding extra elements achieves little overall performance advantage unless you allow the array size to increase. I still maintain that is the case. A 2-element hexbeam gets within 0.75dB of the Forward Gain of a 3-element Octobeam of the same turn radius, and has a marginally higher F/B. That suggests to me there is a lot more performance potential in the 3-element Octobeam that could be released if it were larger.

                          But as I said earlier, just which design point is chosen amidst all the trade-offs is very much the personal choice of the designer.

                          73,
                          Steve G3TXQ



                          --- In hex-beam@yahoogroups.com, "BG" <ko6hl@...> wrote:
                          >
                          > Thanks Steve. I think we are saying the same thing. Except I didn't say I
                          > got the same results, I said I got similar results. You can view them here:
                          >
                          >
                          >
                          > http://www.norcalmfg.com/tmp/OB-20m-0DG-2.jpg
                          >
                          >
                          >
                          > Every antenna is a compromise and an optimized OctoBeam with an 11 ft turn
                          > radius is no exception. The exceptional characteristic to note about the
                          > OctoBeam that makes it different from a conventional Yagi is that the
                          > director element is excited by the driven element in much the same way the
                          > reflector element is excited; at the ends.
                          >
                          >
                          >
                          > Now, for fun, lets move the hexbeam driven element away from the reflector,
                          > re-optimize and compare that to an optimized 2 element Yagi. ;-)
                          >
                          >
                          >
                          > 73!
                          >
                          >
                          >
                          > Bill, KO6HL
                          >
                          >
                        • BG
                          Thanks Steve, You can see how the Gain and F/B hold up over the full band here: http://www.norcalmfg.com/case_studies Don t forget to click on the zoom icon
                          Message 12 of 27 , Feb 5, 2010
                          • 0 Attachment
                            Thanks Steve,



                            You can see how the Gain and F/B hold up over the full band here:



                            http://www.norcalmfg.com/case_studies



                            Don't forget to click on the zoom icon for a little better detail.



                            73!



                            Bill, KO6HL

                            2 is good. 3 must be better. How much better?

                            http://www.OctoBeam.com <http://www.octobeam.com/>





                            _____

                            From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                            Of Steve
                            Sent: Friday, February 05, 2010 7:37 AM
                            To: hex-beam@yahoogroups.com
                            Subject: [hex-beam] Re: I want a hex beam







                            Bill,

                            We may be saying the same thing ;)

                            Your figures illustrate well the design trade-offs: on the 3-element
                            Octobeam the F/B has improved from 9.43dB to 18.87dB, but the cost has been
                            a loss of 1dB in Forward gain.

                            The other trade-off we haven't discussed is performance bandwidth. I wonder
                            how you find the Gain and F/B hold up over the full band? I always felt that
                            the move back to a "W" shape Reflector might mean some compromise in
                            bandwidth.

                            My original comment, which you challenged, was that adding extra elements
                            achieves little overall performance advantage unless you allow the array
                            size to increase. I still maintain that is the case. A 2-element hexbeam
                            gets within 0.75dB of the Forward Gain of a 3-element Octobeam of the same
                            turn radius, and has a marginally higher F/B. That suggests to me there is a
                            lot more performance potential in the 3-element Octobeam that could be
                            released if it were larger.

                            But as I said earlier, just which design point is chosen amidst all the
                            trade-offs is very much the personal choice of the designer.

                            73,
                            Steve G3TXQ

                            --- In hex-beam@yahoogroup <mailto:hex-beam%40yahoogroups.com> s.com, "BG"
                            <ko6hl@...> wrote:
                            >
                            > Thanks Steve. I think we are saying the same thing. Except I didn't say I
                            > got the same results, I said I got similar results. You can view them
                            here:
                            >
                            >
                            >
                            > http://www.norcalmf <http://www.norcalmfg.com/tmp/OB-20m-0DG-2.jpg>
                            g.com/tmp/OB-20m-0DG-2.jpg
                            >
                            >
                            >
                            > Every antenna is a compromise and an optimized OctoBeam with an 11 ft turn
                            > radius is no exception. The exceptional characteristic to note about the
                            > OctoBeam that makes it different from a conventional Yagi is that the
                            > director element is excited by the driven element in much the same way the
                            > reflector element is excited; at the ends.
                            >
                            >
                            >
                            > Now, for fun, lets move the hexbeam driven element away from the
                            reflector,
                            > re-optimize and compare that to an optimized 2 element Yagi. ;-)
                            >
                            >
                            >
                            > 73!
                            >
                            >
                            >
                            > Bill, KO6HL
                            >
                            >



                            No virus found in this incoming message.
                            Checked by AVG - www.avg.com
                            Version: 9.0.733 / Virus Database: 271.1.1/2667 - Release Date: 02/03/10
                            23:35:00




                            [Non-text portions of this message have been removed]
                          • Steve
                            Bill, Thanks for sharing that data with us - the performance certainly seems to hold up quite well. By the way, the data for 14,000kHz appears to be missing
                            Message 13 of 27 , Feb 7, 2010
                            • 0 Attachment
                              Bill,

                              Thanks for sharing that data with us - the performance certainly seems to hold up quite well. By the way, the data for 14,000kHz appears to be missing from the set, unless it's something to do with my browser not seeing all the images.

                              I noted that the model in the results you posted in the PDF file here used tapered segmentation, whereas the model shown on your norcalmfg site uses linear segmentation. I wonder how different you found the results between the two models?

                              73,
                              Steve G3TXQ


                              --- In hex-beam@yahoogroups.com, "BG" <ko6hl@...> wrote:
                              >
                              > Thanks Steve,
                              > You can see how the Gain and F/B hold up over the full band here:
                              > Don't forget to click on the zoom icon for a little better detail.
                              >
                              > 73!
                            • Steve
                              Bill, The Octobeam performance charts you published here: http://www.norcalmfg.com/case_studies encouraged me to do some further work on my own EZNEC models
                              Message 14 of 27 , Feb 8, 2010
                              • 0 Attachment
                                Bill,

                                The Octobeam performance charts you published here:

                                http://www.norcalmfg.com/case_studies

                                encouraged me to do some further work on my own EZNEC models yesterday evening!

                                I eventually ended up with a model that produces almost identical performance figures to the ones you published - it has the same turn radius, the same number of segments, and the same Average Gain figure of 0.07dB; at each frequency my model exhibits the same azimuth shapes as yours, and it produces Forward Gain and F/B figures which are respectively within 0.08dB and 1dB of yours. You'll also see that secondary parameters such as Beamwidth, Sidelobe Gain and Front/Sidelobe ratio also match very closely. I've posted the results on my web site for comparison:

                                http://www.karinya.net/g3txq/modelling_data/

                                Given the almost identical results, I assumed that I must have a model very similar to yours; however, try as I may, and after much tearing out of hair which I can ill-afford to lose, I just cannot get a sensible set of VSWR figures; they range from 10.9:1 to 2.2:1 across the band! They are shown in my final chart.

                                To avoid any further unnecessary loss of hair caused by chasing the unobtainable, can you confirm that the model which produced those performance figures also delivers the <1.5:1 VSWR quoted on your web site?

                                73,
                                Steve G3TXQ


                                --- In hex-beam@yahoogroups.com, "BG" <ko6hl@...> wrote:
                                >
                                > Thanks Steve,
                                > You can see how the Gain and F/B hold up over the full band here:
                                >
                                > http://www.norcalmfg.com/case_studies
                                >
                              • BG
                                Hello Steve, I have not had the opportunity to look closely at my EZNEC linear and tapered segment models to compare them for differences, but I will do that
                                Message 15 of 27 , Feb 8, 2010
                                • 0 Attachment
                                  Hello Steve,



                                  I have not had the opportunity to look closely at my EZNEC linear and
                                  tapered segment models to compare them for differences, but I will do that
                                  soon and let you know what I find.



                                  Actually, your VSWR graph looks a little better than mine. The VSWR
                                  specification on my website came from actual measurement of the 6 OctoBeams
                                  in existence. Hopefully, you will be willing and able to explain the reason
                                  for differences between the computer models and the antennas.



                                  73!



                                  Bill, KO6HL



                                  _____

                                  From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                                  Of Steve
                                  Sent: Monday, February 08, 2010 2:33 AM
                                  To: hex-beam@yahoogroups.com
                                  Subject: [hex-beam] Re: I want a hex beam







                                  Bill,

                                  The Octobeam performance charts you published here:

                                  http://www.norcalmf <http://www.norcalmfg.com/case_studies>
                                  g.com/case_studies

                                  encouraged me to do some further work on my own EZNEC models yesterday
                                  evening!

                                  I eventually ended up with a model that produces almost identical
                                  performance figures to the ones you published - it has the same turn radius,
                                  the same number of segments, and the same Average Gain figure of 0.07dB; at
                                  each frequency my model exhibits the same azimuth shapes as yours, and it
                                  produces Forward Gain and F/B figures which are respectively within 0.08dB
                                  and 1dB of yours. You'll also see that secondary parameters such as
                                  Beamwidth, Sidelobe Gain and Front/Sidelobe ratio also match very closely.
                                  I've posted the results on my web site for comparison:

                                  http://www.karinya <http://www.karinya.net/g3txq/modelling_data/>
                                  net/g3txq/modelling_data/

                                  Given the almost identical results, I assumed that I must have a model very
                                  similar to yours; however, try as I may, and after much tearing out of hair
                                  which I can ill-afford to lose, I just cannot get a sensible set of VSWR
                                  figures; they range from 10.9:1 to 2.2:1 across the band! They are shown in
                                  my final chart.

                                  To avoid any further unnecessary loss of hair caused by chasing the
                                  unobtainable, can you confirm that the model which produced those
                                  performance figures also delivers the <1.5:1 VSWR quoted on your web site?

                                  73,
                                  Steve G3TXQ

                                  --- In hex-beam@yahoogroup <mailto:hex-beam%40yahoogroups.com> s.com, "BG"
                                  <ko6hl@...> wrote:
                                  >
                                  > Thanks Steve,
                                  > You can see how the Gain and F/B hold up over the full band here:
                                  >
                                  > http://www.norcalmf <http://www.norcalmfg.com/case_studies>
                                  g.com/case_studies
                                  >





                                  [Non-text portions of this message have been removed]
                                • Steve
                                  Hi Bill, As you know, in my hexbeam work I ve seen very close agreement between EZNEC predictions and what I measure in practice; any time there was a
                                  Message 16 of 27 , Feb 8, 2010
                                  • 0 Attachment
                                    Hi Bill,

                                    As you know, in my hexbeam work I've seen very close agreement between EZNEC predictions and what I measure in practice; any time there was a significant difference it was because I'd failed to include something significant in the model.

                                    I see nothing in the Octobeam geometry that would make it any more difficult to model than the hexbeam, so if your model is predicting very different VSWRs from those you measure I think you can be pretty sure that the model is flawed in some way and that the predictions for Gain and F/B are suspect.

                                    One thing I found when modelling the hexbeam was the importance of accounting for the terminations at the ends of the elements. The Octobeam performance and feedpoint impedance seems quite sensitive to the coupling between the tips of the Director and the "knees" of the Driver; I certainly see improved VSWRs - accompanied by lower gain - as I vary that gap on the model. I'm not sure how you terminate the Director wires, but could that be an area where the model is not complete because it doesn't properly account for the Driver/Director coupling?

                                    If you felt able to share the exact wire and gap dimensions and the termination method, I could delve more deeply; but I quite understand if you are reluctant to do that for "commercial-in-confidence" reasons.

                                    73,
                                    Steve G3TXQ




                                    --- In hex-beam@yahoogroups.com, "BG" <ko6hl@...> wrote:
                                    >
                                    > Hello Steve,
                                    >
                                    >
                                    >
                                    > I have not had the opportunity to look closely at my EZNEC linear and
                                    > tapered segment models to compare them for differences, but I will do that
                                    > soon and let you know what I find.
                                    >
                                    >
                                    > Actually, your VSWR graph looks a little better than mine. The VSWR
                                    > specification on my website came from actual measurement of the 6 OctoBeams
                                    > in existence. Hopefully, you will be willing and able to explain the reason
                                    > for differences between the computer models and the antennas.
                                    >
                                    > 73!
                                    > Bill, KO6HL
                                    >
                                  • BG
                                    Hi Steve, Please see my comments below. _____ From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf Of Steve Sent: Monday, February 08,
                                    Message 17 of 27 , Feb 8, 2010
                                    • 0 Attachment
                                      Hi Steve,



                                      Please see my comments below.



                                      _____

                                      From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                                      Of Steve
                                      Sent: Monday, February 08, 2010 9:17 AM
                                      To: hex-beam@yahoogroups.com
                                      Subject: [hex-beam] Re: I want a hex beam

                                      Hi Bill,

                                      As you know, in my hexbeam work I've seen very close agreement between EZNEC
                                      predictions and what I measure in practice; any time there was a significant
                                      difference it was because I'd failed to include something significant in the
                                      model.[BG] Your hexbeam models are the only ones I've ever seen published
                                      that agree well with your measured results. All others show significant
                                      difference in VSWR curves between model results and measured results.

                                      I see nothing in the Octobeam geometry that would make it any more difficult
                                      to model than the hexbeam, so if your model is predicting very different
                                      VSWRs from those you measure I think you can be pretty sure that the model
                                      is flawed in some way and that the predictions for Gain and F/B are
                                      suspect.[BG] I don't think there is anything unusual in the geometry and I
                                      can't find anything wrong with the model. In fact, my model is not
                                      different from yours except for possible wire lengths and gaps.

                                      One thing I found when modelling the hexbeam was the importance of
                                      accounting for the terminations at the ends of the elements. The Octobeam
                                      performance and feedpoint impedance seems quite sensitive to the coupling
                                      between the tips of the Director and the "knees" of the Driver; I certainly
                                      see improved VSWRs - accompanied by lower gain - as I vary that gap on the
                                      model. I'm not sure how you terminate the Director wires, but could that be
                                      an area where the model is not complete because it doesn't properly account
                                      for the Driver/Director coupling?[BG] Your model is the same as mine in
                                      that respect. What would need to be done to model the wire ends more
                                      accurately? What are you doing in your hexbeam models? I have experimented
                                      with a couple of different methods of terminating the wire ends and found no
                                      significant difference in VSWR caused by any method. You can see here how
                                      the wire ends are currently terminated and fixed relative to each other:

                                      http://www.norcalmfg.com/construction

                                      If you felt able to share the exact wire and gap dimensions and the
                                      termination method, I could delve more deeply; but I quite understand if you
                                      are reluctant to do that for "commercial-in-confidence" reasons.[BG] I
                                      don't have a problem sharing the dimensions. I believe your existing model
                                      is fairly close to mine, judging by your performance plots, but I will
                                      prepare a table of dimensions and e-mail them to you.

                                      73,
                                      Steve G3TXQ



                                      [Non-text portions of this message have been removed]
                                    • John Huggins
                                      Hi Steve and BG. Please let us know if you are using the same ground attributes and if so same height above ground. Also please share conductor materials,
                                      Message 18 of 27 , Feb 8, 2010
                                      • 0 Attachment
                                        Hi Steve and BG.

                                        Please let us know if you are using the same ground attributes and if so
                                        same height above ground. Also please share conductor materials, etc.
                                        There are quite a few variables in Moment of Method codes like NEC that
                                        can make things different.

                                        John

                                        BG wrote:
                                        >
                                        >
                                        > Hi Steve,
                                        >
                                        > Please see my comments below.
                                        >
                                        > _____
                                        >
                                        > From: hex-beam@yahoogroups.com <mailto:hex-beam%40yahoogroups.com>
                                        > [mailto:hex-beam@yahoogroups.com <mailto:hex-beam%40yahoogroups.com>] On
                                        > Behalf
                                        > Of Steve
                                        > Sent: Monday, February 08, 2010 9:17 AM
                                        > To: hex-beam@yahoogroups.com <mailto:hex-beam%40yahoogroups.com>
                                        > Subject: [hex-beam] Re: I want a hex beam
                                        >
                                        > Hi Bill,
                                        >
                                        > As you know, in my hexbeam work I've seen very close agreement between EZNEC
                                        > predictions and what I measure in practice; any time there was a significant
                                        > difference it was because I'd failed to include something significant in the
                                        > model.[BG] Your hexbeam models are the only ones I've ever seen published
                                        > that agree well with your measured results. All others show significant
                                        > difference in VSWR curves between model results and measured results.
                                        >
                                        > I see nothing in the Octobeam geometry that would make it any more difficult
                                        > to model than the hexbeam, so if your model is predicting very different
                                        > VSWRs from those you measure I think you can be pretty sure that the model
                                        > is flawed in some way and that the predictions for Gain and F/B are
                                        > suspect.[BG] I don't think there is anything unusual in the geometry and I
                                        > can't find anything wrong with the model. In fact, my model is not
                                        > different from yours except for possible wire lengths and gaps.
                                        >
                                        > One thing I found when modelling the hexbeam was the importance of
                                        > accounting for the terminations at the ends of the elements. The Octobeam
                                        > performance and feedpoint impedance seems quite sensitive to the coupling
                                        > between the tips of the Director and the "knees" of the Driver; I certainly
                                        > see improved VSWRs - accompanied by lower gain - as I vary that gap on the
                                        > model. I'm not sure how you terminate the Director wires, but could that be
                                        > an area where the model is not complete because it doesn't properly account
                                        > for the Driver/Director coupling?[BG] Your model is the same as mine in
                                        > that respect. What would need to be done to model the wire ends more
                                        > accurately? What are you doing in your hexbeam models? I have experimented
                                        > with a couple of different methods of terminating the wire ends and found no
                                        > significant difference in VSWR caused by any method. You can see here how
                                        > the wire ends are currently terminated and fixed relative to each other:
                                        >
                                        > http://www.norcalmfg.com/construction
                                        > <http://www.norcalmfg.com/construction>
                                        >
                                        > If you felt able to share the exact wire and gap dimensions and the
                                        > termination method, I could delve more deeply; but I quite understand if you
                                        > are reluctant to do that for "commercial-in-confidence" reasons.[BG] I
                                        > don't have a problem sharing the dimensions. I believe your existing model
                                        > is fairly close to mine, judging by your performance plots, but I will
                                        > prepare a table of dimensions and e-mail them to you.
                                        >
                                        > 73,
                                        > Steve G3TXQ
                                        >
                                        > [Non-text portions of this message have been removed]
                                        >
                                        >
                                      • Steve
                                        John, Yes, those parameters make a difference; but I am seeing gross differences between the model VSWRs and what Bill is measuring - 10:1 vs 1.5:1 - that
                                        Message 19 of 27 , Feb 9, 2010
                                        • 0 Attachment
                                          John,

                                          Yes, those parameters make a difference; but I am seeing gross differences between the model VSWRs and what Bill is measuring - 10:1 vs 1.5:1 - that indicates something much more fundamentally wrong with our models.

                                          73,
                                          Steve G3TXQ

                                          --- In hex-beam@yahoogroups.com, John Huggins <john@...> wrote:
                                          >
                                          > Hi Steve and BG.
                                          >
                                          > Please let us know if you are using the same ground attributes and if so
                                          > same height above ground. Also please share conductor materials, etc.
                                          > There are quite a few variables in Moment of Method codes like NEC that
                                          > can make things different.
                                          >
                                          > John
                                        • Steve
                                          Bill, Thanks, I look forward to getting your dimensions and checking my model. You mentioned that mine are the only hexbeam model results you ve ever seen
                                          Message 20 of 27 , Feb 9, 2010
                                          • 0 Attachment
                                            Bill,

                                            Thanks, I look forward to getting your dimensions and checking my model.

                                            You mentioned that mine are the only hexbeam model results you've ever seen published that agree well with measured results. Actually, I can't recall seeing any other in-depth comparison of measured azimuth patterns, F/B and VSWR results against model predictions, but perhaps I've not looked in the right places. I'm comfortable with my hexbeam comparisons - the model is publicly available for folk to play with and see the predictions, and the antenna's practical performance has been evaluated by many, including Leo and DX Engineering.

                                            One thought occurs to me - presumably you are making measurements on a multiband "stack", whereas my model - and I believe yours - is a monobander. Have you made practical measurements on a 20m monobander to compare directly with the monoband model?

                                            73,
                                            Steve G3TXQ
                                          • BG
                                            Hello John and Steve, I believe Steve and I are using the same OctoBeam parameters (free space) and possibly very close wire lengths and materials (14 AWG
                                            Message 21 of 27 , Feb 9, 2010
                                            • 0 Attachment
                                              Hello John and Steve,

                                              I believe Steve and I are using the same OctoBeam parameters (free space)
                                              and possibly very close wire lengths and materials (14 AWG insulated copper
                                              wire) and we are getting very similar Modeling results. I have previously
                                              given Steve OctoBeam geometry details and he is using his model generated
                                              from my spreadsheet of cartesian coordinates. However, measured VSWR
                                              results vary greatly with the OctoBeam models regardless whether mono-band
                                              or multi-band arrays are measured and analyzed.

                                              Steve has suggested that the OctoBeam wire end terminations are not
                                              described accurately within the model, so I have researched this and added
                                              similar large conductive cylinders (3/4" dia x 1" long) placed at the wire
                                              ends, as in Steve's BBHB model. The result is essentially the same; the
                                              OctoBeam VSWR curves produced by EZNEC vary greatly from the measured
                                              curves. I have also removed those cylinders from Steve's BBHB model and
                                              found that the new VSWR curves generated by EZNEC are similar in shape and
                                              value to those generated for the OctoBeam (~9:1 - 2.5:1 across the band).

                                              I have attempted to determine what difference, if any, others have found
                                              between their BBHB model curves and measured curves, and I have been able to
                                              find only a few occurrences published on the internet. Any of the
                                              commercial manufacturers publish either one or the other or none, but not
                                              both, and none supply the EZNEC model for download. Those that I have found
                                              show significant differences between modeled and measured VSWR curves. With
                                              the exception of the published curves by Allen Baker, KG4JJH (Thank you
                                              Allen!), http://www.kg4jjh.com/pdf/HexBeam-1.pdf and
                                              http://www.kg4jjh.com/pdf/HexBeam-2.pdf, I don't know if any of the others'
                                              BBHB models are using the 3/4" dia x 1" long cylinder termination. Allen
                                              does not use any cylinders at the wire ends in his BBHB model:

                                              http://www.kg4jjh.com/eznec/KG4JJH%20HexBeam.ez

                                              Hopefully Allen will read this and jump in with a few comments and possible
                                              guidance.

                                              The following are the only manufacturer's websites I researched:

                                              Traffie - measured curves, no modeled curves
                                              K4KIO - modeled curves, no measured curves
                                              DX Engr - no curves

                                              We are still working on the technical data at http://www.OctoBeam.com, but
                                              we intend to publish both modeled and measured data.

                                              73!
                                              Bill, KO6HL

                                              -----Original Message-----
                                              From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                                              Of John Huggins
                                              Sent: Monday, February 08, 2010 6:14 PM
                                              To: hex-beam@yahoogroups.com
                                              Subject: [hex-beam] Octobeam EZNEC Modeling - to help us better understand
                                              hexbeam models

                                              Hi Steve and BG.

                                              Please let us know if you are using the same ground attributes and if so
                                              same height above ground. Also please share conductor materials, etc.
                                              There are quite a few variables in Moment of Method codes like NEC that can
                                              make things different.

                                              John

                                              BG wrote:
                                              >
                                              >
                                              > Hi Steve,
                                              >
                                              > Please see my comments below.
                                              >
                                              > _____
                                              >
                                              > From: hex-beam@yahoogroups.com <mailto:hex-beam%40yahoogroups.com>
                                              > [mailto:hex-beam@yahoogroups.com <mailto:hex-beam%40yahoogroups.com>]
                                              > On Behalf Of Steve
                                              > Sent: Monday, February 08, 2010 9:17 AM
                                              > To: hex-beam@yahoogroups.com <mailto:hex-beam%40yahoogroups.com>
                                              > Subject: [hex-beam] Re: I want a hex beam
                                              >
                                              > Hi Bill,
                                              >
                                              > As you know, in my hexbeam work I've seen very close agreement between
                                              > EZNEC predictions and what I measure in practice; any time there was a
                                              > significant difference it was because I'd failed to include something
                                              > significant in the model.[BG] Your hexbeam models are the only ones
                                              > I've ever seen published that agree well with your measured results.
                                              > All others show significant difference in VSWR curves between model
                                              results and measured results.
                                              >
                                              > I see nothing in the Octobeam geometry that would make it any more
                                              > difficult to model than the hexbeam, so if your model is predicting
                                              > very different VSWRs from those you measure I think you can be pretty
                                              > sure that the model is flawed in some way and that the predictions for
                                              > Gain and F/B are suspect.[BG] I don't think there is anything unusual
                                              > in the geometry and I can't find anything wrong with the model. In
                                              > fact, my model is not different from yours except for possible wire
                                              lengths and gaps.
                                              >
                                              > One thing I found when modelling the hexbeam was the importance of
                                              > accounting for the terminations at the ends of the elements. The
                                              > Octobeam performance and feedpoint impedance seems quite sensitive to
                                              > the coupling between the tips of the Director and the "knees" of the
                                              > Driver; I certainly see improved VSWRs - accompanied by lower gain -
                                              > as I vary that gap on the model. I'm not sure how you terminate the
                                              > Director wires, but could that be an area where the model is not
                                              > complete because it doesn't properly account for the Driver/Director
                                              > coupling?[BG] Your model is the same as mine in that respect. What
                                              > would need to be done to model the wire ends more accurately? What are
                                              > you doing in your hexbeam models? I have experimented with a couple of
                                              > different methods of terminating the wire ends and found no
                                              > significant difference in VSWR caused by any method. You can see here how
                                              the wire ends are currently terminated and fixed relative to each other:
                                              >
                                              > http://www.norcalmfg.com/construction
                                              > <http://www.norcalmfg.com/construction>
                                              >
                                              > If you felt able to share the exact wire and gap dimensions and the
                                              > termination method, I could delve more deeply; but I quite understand
                                              > if you are reluctant to do that for "commercial-in-confidence"
                                              > reasons.[BG] I don't have a problem sharing the dimensions. I believe
                                              > your existing model is fairly close to mine, judging by your
                                              > performance plots, but I will prepare a table of dimensions and e-mail
                                              them to you.
                                              >
                                              > 73,
                                              > Steve G3TXQ
                                              >
                                              > [Non-text portions of this message have been removed]
                                              >
                                              >


                                              ------------------------------------

                                              Yahoo! Groups Links



                                              No virus found in this incoming message.
                                              Checked by AVG - www.avg.com
                                              Version: 9.0.733 / Virus Database: 271.1.1/2676 - Release Date: 02/08/10
                                              11:35:00
                                            • BG
                                              Hi Steve, I downloaded your BBHB model and had a look at the wires and I see now what you have done with the wire ends. I overlooked that detail originally,
                                              Message 22 of 27 , Feb 9, 2010
                                              • 0 Attachment
                                                Hi Steve,



                                                I downloaded your BBHB model and had a look at the wires and I see now what
                                                you have done with the wire ends. I overlooked that detail originally, so I
                                                wasn’t certain what you were referring to. What are you using that is ¾”
                                                diameter x 1” long at the wire ends? I had previously downloaded the file:
                                                General 6-Band G3TXQ Broadband Hex Beam.EZ from the files section and
                                                assumed this is your model. This model has no ¾” cylinders at the ends
                                                either, so the VSWR curves are similar to my OctoBeam modeled curves.



                                                I will keep working on this issue and see what I can come up with as well.
                                                I had originally tried to model the center conductor and the ground of the
                                                coax as they are actually connected to the wires, but that was very early on
                                                and I was told that was unnecessary. Do you think it is worthwhile to
                                                revisit that approach?



                                                Thanks and 73!

                                                Bill, KO6HL



                                                _____

                                                From: hex-beam@yahoogroups.com [mailto:hex-beam@yahoogroups.com] On Behalf
                                                Of Steve
                                                Sent: Tuesday, February 09, 2010 1:32 AM
                                                To: hex-beam@yahoogroups.com
                                                Subject: [hex-beam] Re: I want a hex beam





                                                Bill,

                                                Thanks, I look forward to getting your dimensions and checking my model.

                                                You mentioned that mine are the only hexbeam model results you've ever seen
                                                published that agree well with measured results. Actually, I can't recall
                                                seeing any other in-depth comparison of measured azimuth patterns, F/B and
                                                VSWR results against model predictions, but perhaps I've not looked in the
                                                right places. I'm comfortable with my hexbeam comparisons - the model is
                                                publicly available for folk to play with and see the predictions, and the
                                                antenna's practical performance has been evaluated by many, including Leo
                                                and DX Engineering.

                                                One thought occurs to me - presumably you are making measurements on a
                                                multiband "stack", whereas my model - and I believe yours - is a monobander.
                                                Have you made practical measurements on a 20m monobander to compare directly
                                                with the monoband model?

                                                73,
                                                Steve G3TXQ



                                                No virus found in this incoming message.
                                                Checked by AVG - www.avg.com
                                                Version: 9.0.733 / Virus Database: 271.1.1/2676 - Release Date: 02/08/10
                                                11:35:00




                                                [Non-text portions of this message have been removed]
                                              • Steve
                                                Bill, On my early hexbeam models I saw a consistent frequency offset between the modelled and measured VSWR curves; it wasn t great - about 70kHz on 20m -
                                                Message 23 of 27 , Feb 9, 2010
                                                • 0 Attachment
                                                  Bill,

                                                  On my early hexbeam models I saw a consistent "frequency offset" between the modelled and measured VSWR curves; it wasn't great - about 70kHz on 20m - but it annoyed me because I couldn't understand it! I must make clear that, unlike what we are seeing on the Octobeam model, the *shape* and *values* of the modelled VSWR curve were the same shape and values that I measured - they were simply offset 70kHz in frequency. I had several exchanges of eMails with Roy Lewallen about it and Roy convinced me it must be something I had omitted from the model, and that there there was no mechanism within EZNEC that would cause that sort of offset.

                                                  The only thing I could think of which I hadn't included in the model was the brass connector blocks I used to attach the wires to the spacer cords. I'd also noted that the frequency offset did not scale with frequency, and was therefore most likely some sort of "end effect". I did some measurements on the effects of those blocks, which is written up here:

                                                  http://www.karinya.net/g3txq/hexbeam/end_spacers/

                                                  The addition of the blocks was enough to change the resonant frequency of a 10m hexbeam driver by as much as 260kHz, depending on the size of the block and the length of terminal screw sticking out from it.

                                                  Of course you can't easily model in EZNEC a half-round brass block with a screw sticking out of it!! The cylinders I included in the model were my attempt to model it at least in some form - they certainly eliminated most of the frequency offset.

                                                  The end spacing on the hexbeam is large enough that the inclusion or exclusion of those blocks doesn't have a major effect on the VSWR, the Gain or the F/B, just the tuning; but, given the proximity of the Director tips to the Driver "knees" on the Octobeam I wondered if the end effects might be greater.

                                                  Of course, you must use the EZNEC Transmission Line capability to model the differential-mode effects of the coax. I'm doubtful that common-mode issues could be causing the effects you are seeing, but if you think it might be a factor you should model the braid - *not* the inner conductor - as a set of wires of appropriate diameter connecting one side of the Driven elements. You can also include a load to represent any common-mode choke.

                                                  I know this is a crazy question, but I have to ask it - are you using insulated wire, but modelling bare copper wire?

                                                  73,
                                                  Steve G3TXQ




                                                  --- In hex-beam@yahoogroups.com, "BG" <ko6hl@...> wrote:
                                                  >
                                                  > Hi Steve,
                                                  >
                                                  >
                                                  >
                                                  > I downloaded your BBHB model and had a look at the wires and I see now what
                                                  > you have done with the wire ends. I overlooked that detail originally, so I
                                                  > wasn't certain what you were referring to. What are you using that is ¾"
                                                  > diameter x 1" long at the wire ends? I had previously downloaded the file:
                                                  > General 6-Band G3TXQ Broadband Hex Beam.EZ from the files section and
                                                  > assumed this is your model. This model has no ¾" cylinders at the ends
                                                  > either, so the VSWR curves are similar to my OctoBeam modeled curves.
                                                  >
                                                  >
                                                  >
                                                  > I will keep working on this issue and see what I can come up with as well.
                                                  > I had originally tried to model the center conductor and the ground of the
                                                  > coax as they are actually connected to the wires, but that was very early on
                                                  > and I was told that was unnecessary. Do you think it is worthwhile to
                                                  > revisit that approach?
                                                  >
                                                  >
                                                  >
                                                  > Thanks and 73!
                                                  >
                                                  > Bill, KO6HL
                                                  >
                                                • Steve
                                                  Bill, This may be pure coincidence, but if I set your Director/Driver gap to 4 , assume 0.015 thick PVC insulation, and the beam at 30ft over average ground,
                                                  Message 24 of 27 , Feb 9, 2010
                                                  • 0 Attachment
                                                    Bill,

                                                    This may be pure coincidence, but if I set your Director/Driver gap to 4", assume 0.015" thick PVC insulation, and the beam at 30ft over average ground, my model predicts a VSWR of less than 1.8:1 across the 20m band with a minimum of 1.5:1. Measured through 100ft of RG213 that would translate to VSWRs between 1.4:1 and 1.6:1

                                                    73,
                                                    Steve G3TXQ
                                                  • Steve
                                                    Bill, Reading carefully the details on your web site, I see that the wire you are using **is** insulated! 73, Steve G3TXQ
                                                    Message 25 of 27 , Feb 10, 2010
                                                    • 0 Attachment
                                                      Bill,

                                                      Reading carefully the details on your web site, I see that the wire you are using **is** insulated!

                                                      73,
                                                      Steve G3TXQ

                                                      --- In hex-beam@yahoogroups.com, "Steve" <steve@...> wrote:
                                                      >
                                                      > Bill,
                                                      >
                                                      > This may be pure coincidence, but if I set your Director/Driver gap to 4", assume 0.015" thick PVC insulation, and the beam at 30ft over average ground, my model predicts a VSWR of less than 1.8:1 across the 20m band with a minimum of 1.5:1. Measured through 100ft of RG213 that would translate to VSWRs between 1.4:1 and 1.6:1
                                                      >
                                                      > 73,
                                                      > Steve G3TXQ
                                                      >
                                                    • Lou Campbell
                                                      I had called Mike a month ago to say hello and we talked about him going to dayton this year. He had originally told me he  was not going....but now see him
                                                      Message 26 of 27 , Apr 9, 2011
                                                      • 0 Attachment
                                                        I had called Mike a month ago to say hello and we talked about him going to dayton this year. He had originally told me he  was not going....but now see him on the list inside at booth 11...so if you are going to dayton stop by...and say hello to him..i will be there alot...so you may get to say hello to me also...have met many hex nuts there and hope to meet more..
                                                           c ya at dayton...
                                                         I wlkl also be at the dx/contest forums also..and at most of the dx/contest suits at the crown also as for the 1st time I have a room at the crown this year...


                                                        ED NX7TT

                                                        Idaho Falls, Id

                                                        --





                                                        [Non-text portions of this message have been removed]
                                                      • bur52@aol.com
                                                        Great job Ed ! Say hello to mike for me, if he is there. 73 John/k3jvb In a message dated 4/9/2011 9:01:03 A.M. Eastern Daylight Time, nx7tt@yahoo.com writes:
                                                        Message 27 of 27 , Apr 9, 2011
                                                        • 0 Attachment
                                                          Great job Ed !

                                                          Say hello to mike for me, if he is there.
                                                          73
                                                          John/k3jvb


                                                          In a message dated 4/9/2011 9:01:03 A.M. Eastern Daylight Time,
                                                          nx7tt@... writes:




                                                          I had called Mike a month ago to say hello and we talked about him going to
                                                          dayton this year. He had originally told me he was not going....but now
                                                          see him on the list inside at booth 11...so if you are going to dayton stop
                                                          by...and say hello to him..i will be there alot...so you may get to say
                                                          hello to me also...have met many hex nuts there and hope to meet more..
                                                          c ya at dayton...
                                                          I wlkl also be at the dx/contest forums also..and at most of the
                                                          dx/contest suits at the crown also as for the 1st time I have a room at the crown
                                                          this year...

                                                          ED NX7TT

                                                          Idaho Falls, Id

                                                          --

                                                          [Non-text portions of this message have been removed]






                                                          [Non-text portions of this message have been removed]
                                                        Your message has been successfully submitted and would be delivered to recipients shortly.