Woud a disguised Cobweb work the same?

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by G1YBB, Sep 8, 2015.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Left-3
ad: L-Geochron
ad: Left-2
ad: abrind-2
  1. G1YBB

    G1YBB XML Subscriber QRZ Page

    it's the file on this post:

    the first box that pops up is this:

    When it opens I see this:

    though in the editor I do have 93 wires:

    Running a frequency sweep 14 to 28 gets this:

    I'm guessing its a scaling issue maybe. Metres imported as mm or inches maybe... Away to check
    Edit: Just noticed the Z axis is 50,ooo inches, or 4000+ feet...
    Edit2: Looking at the numbers in the editor, X and Y's in order of 33 inches and a Z of 300 inches looks what I would expect, 25 feet above ground etc.
    Last edited: Feb 14, 2021
  2. AC6LA

    AC6LA Ham Member QRZ Page

    Ah, perfect. Now we are on the same page!

    The 4nec2 message that begins with "This setting does only affect the display of wire dimensions ..." is for information only. It is not an error message. It will appear any time an EZNEC file is opened which has "Units" as anything other than "Meters".

    If you look closely I believe you'll see that the model actually has 94 wires.


    That 94th wire is for an EZNEC "virtual segment". Notice that it is out in space by ~100 wl in X/Y/Z. (At 14.175 MHz 100 wl is ~83265 inches.) When this *.ez model is viewed with EZNEC the "virtual segment" wire is not displayed. But 4nec2 does not recognize that wire as anything special so when the model is viewed with 4nec2 the view is "zoomed out" to include all wires. Hence the actual antenna wires 1-93 become so small as to be invisible.


    If you zoom in (Pg up) several times, and keep the X/Y/Z axes origin in the center of the view, the antenna wires will eventually become visible.

    And that is the real problem. The EZNEC model has a 4:1 (or 50:12.5) transformer between the wire with the source (the "virtual segment" wire, number 94) and the center of the 20m element (wire 63).


    But 4nec2 does not recognize the transformer part of the EZNEC model file. So as far as 4nec2 knows there is no connection between the source (wire 94) and the center of the 20m element (wire 63). That's why the 4nec2 frequency sweep shows an almost infinite SWR from 14 to 28 MHz. That's the SWR that would apply to a very short wire out in space, fed with a voltage source, and with no connection to a bunch of other wires.

    4nec2 does not recognize any of the features that have been added to EZNEC after EZNEC v. 3.0. In addition to transformers that includes wire insulation, loss in transmission lines, and L networks.

    To make this EZNEC model work with 4nec2 it is necessary to a) put the source directly on one of the element feedpoints rather than on an "out in space" wire and b) use 4nec2 Settings > Char-impedance of 12.5 ohms rather than 50 ohms as a substitute for the lack of a transformer. And to mimic putting the balun/transformer at the "hub" rather than at one of the element feedpoints, as you would like, a short 1-segment wire can be put at the X=Y=0 location (the hub location).

    The attached zip contains diamond cobweb models in *.weq, *.ez, and *.nec formats with a 12.5 ohm transmission line between the hub and either the 20m element (wire 63) or the 10m element (wire 1) as well as four Zo 12.5 ohm TLs connecting the five element centers. Note that there is a slight difference in the end-to-end element lengths for the two different versions.

    Since 4nec2 does not recognize EZNEC insulation, to make the 4nec2 results match the EZNEC results it is necessary to use the 4nec2 interface to add a custom "LD 7" wire insulation card. Such a card has been added to both *.nec model files to simulate typical THHN insulation. If the EZNEC files are opened with 4nec2 the wires will be bare (not insulated) and the results will not be valid.

    Dan, AC6LA

    Attached Files:

  3. G1YBB

    G1YBB XML Subscriber QRZ Page

    That's awesome, many thanks Dan. I can work with that now. I was trying to make sense of the model after I posted and I did manage to zoom in enough (not easy!!) to see the wires, but I would never have deduced the source out in space thing.

    I've added wire insulation into 4NEC2 models recently with some success, so hopefully I can get a good first stab at this.
    (links to my modelling and the builds with pics for interest 6m moxon and 20m moxon)
    OMG I just realised MoxGen comes from you! On behalf of myself and the greater radio community a huge THANKS for that, such a useful calculator!
    BTW I LOVE my 20m moxon. I work Stateside after Stateside on it when the wx is calm enough to keep it in one place. However I really miss being multiband (as I think I damaged the disguised cobweb running an amp at the band edges with higher SWR and overheating) hence I plan to build a more substantial conventional style cobweb.
    I also intend to hopfully deal with bandwidth by tuning it for SSB optimum but fit small bullet connectors at the ends of the wires and add a small extension for the CW/data end of the bands. I feel this should work. The 20m moxon was made for SSB then I decided to solder on short extensions and it works 1:1 at CW end now (when not pointed at the house, it goes to 1.1:1 then)


    To all:
    I wonder whilst this thread has been running, has anyone watching this thread built one of the diamond hub centre fed cobwebs by any chance?
  4. G1YBB

    G1YBB XML Subscriber QRZ Page

    Playing with the model trying to implement physical insulated wire I may use, trying to find the best option, I 'lost' a band.
    The insulation shortening the wire electrical length had dropped the 20m dip off my SWR plot range.
    This got me thinking, is there a harmonically related reason why I couldn't add a 30m wire a little further out?
    I know 40m wouldn't work as it would have/cause issues with 15m. I guess 10m is pretty closely related to 30m but would it just miss actual issues?
    I suspect not as I have never heard anyone of adding 30m to a cobweb before.
    I'm almost tempted to scale up a bit and go 12-30m, would that work?
  5. G1YBB

    G1YBB XML Subscriber QRZ Page

    OK so I should have googled beforehand. I've turned up a couple of threads here where 30m (and 40m) was added to a 10-20m cobweb. 40m is a bit too big really but I think 30m could go nicely size wise. I have been aclimatising the neighbours and YL with a 20m moxon so this would be smaller by comparison.
    Just need to work out an easy way to add a sixth wire to the model without a lot of trig and a calculator...
  6. N4UFO

    N4UFO Ham Member QRZ Page

    I purchased one of the first run of the MFJ cobweb kits here in the US. Went to the local store and bought some fiberglass reflector posts and added a 30m wire. Worked really great! And my thinking was same as yours, 40m would be a bit too much. Besides, this cobweb was mounted atop a 30 foot tall mast that was insulated from ground and had over 1000 feet or radials running out from it's base. So a little top loading (look closely and you will see a wire loop on top of the cobweb center) and the whole thing loaded up as a fantastic (almost) full size 40m vertical. - Here are some pictures with and without the 30m wire just for comparison:

    cobweb 30m.jpg cobweb.jpg

    The cobweb is a great antenna. I moved and traded the one above for other gear. Thinking of homebrewing another one for the new QTH. - 73, Kevin N4UFO
    Last edited: Feb 17, 2021
    G1YBB likes this.
  7. G1YBB

    G1YBB XML Subscriber QRZ Page

    Thanks Kevin, good to see. It looks no more than a couple of feet per spreader. I've been to-ing and fro-ing with my choice of antenna to put on a new wind up mast I'm building from scratch (below) as I love the gain of my moxon but miss multiband. The yagi in the picture (2 ele as most bang for the buck on gain) is a DK7ZB design that is multiband but by adjusting the elements. So I've re-warmed to the cobweb (which will also not overhang other people's gardens tho he said it's fine) for instant band switching. Adding 30m wire to 10-20m is another tick in the pro column. The major tick n the con column is the narrow band, but I feel I can deal with it with plug in wire extensions for the CW/data end as mentioned above.
    As this one is going on a more substanstial mast that the one I started this (now extremely long!) thread about I can build it a bit more manly...
    Also it would be great to see the new style cobweb that evolved in this thread brought to reality.

    PS. Yes I'm jealous of your antenna helping XYL!! :) My YL is pretty tolerant to be fair but not much more. I have tried and tried to get her to help carry my antenna gear to the top of our local GW mountain but no dice!

    Last edited: Feb 17, 2021
  8. N4UFO

    N4UFO Ham Member QRZ Page

    And rightly you should be! :D:cool: She stood outside just a couple weeks ago and held ropes and things while I twirled an arborist bag around (while standing on a ladder!) to get ropes up over tree limbs to hang the OCFD you saw on my page there. :)

    Seriously, I used to have a 3 element mini beam up about 5 feet lower than the cobweb was at... same tripod/mast, etc. I decided to replace it when some connections and cables needed replacing and I had gotten worried about it all coming down in a strong wind. (Had one set of guy ropes on the beam, went to two with the cobweb.) Honestly, I didn't miss the beam that much when I went to the cobweb. Now there is something to be said for the idea that 'you can't miss what you don't hear'; because that is the real advantage of a directional antenna such as a yagi or a moxon... better reception of weak signals. That said, I think the Cobweb has better reception that most other 'non-beam' antennas... it was quieter and the pattern more 'omni'. I think the fact that the 'ends' are brought around near each other. I'm sure I am not saying this right but I'm told that noise comes off the end of the antennas. Hence, loops are quieter than a dipole. Well, a cobweb is not a loop, but it's close. And that advantage of being quieter does not completely make up for the side and rear rejection of a beam, but it certainly helps!

    As for the bandwidth... 30, 17 & 12m are NOT an issue. On 20, 15 & 10m I tuned it towards the low end of the bands. (I mostly run CW and digital anyway.) On the occasions when I ventured up to the phone section of the band, my tuner allowed it operate okay. I say 'okay' because while I couldn't tell any degradation in efficiency. I would assume there must have been some. After my history of working with compromise antennas (mini beams and reduced size vertical dipole with coils & cap hats) if it was not noticeable, who cares. - Oh, one last thing... if you are considering 6m... if you can get it to tune/work you are better than I am. ;)

    Good luck & 73!

    Kevin N4UFO
  9. WA6FYD

    WA6FYD Ham Member QRZ Page

    I have an MFJ Cobweb antenna still in the box while I dread the tuning. When I had a Traffie full size Hexbeam it worked like tuned circuits. No tuning. I just assembled and got on the air. Somebody tell me how to make this easy. What analyzer should I use? I need this antenna because I'm a prisoner of an HMO. I have a 20 MTR dipole at 6ft height around the wooden fence that works pretty well.
  10. AK5B

    AK5B Ham Member QRZ Page

    Anything lightweight that you can attach at the feedpoint would be best; I love my RigExpert AA-170 for stuff like this. If cost is an object then get a cheap ($60) NanoVNA and an adapter so you can attach it without an extra length of coax.

    Too bad you still don't have the elephant cage; must have been super!


    Jeff (I home-brewed a cobweb about ten years ago; now I am untangling it from being in storage---for so long that my cobweb has its own cobwebs, now) :D

Share This Page

ad: w5yi