Some help with 4NEC2

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by N3DT, Dec 21, 2018.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Left-2
ad: Left-3
ad: MessiPaoloni-1
ad: Subscribe
ad: K5AB-Elect-1
  1. N3DT

    N3DT Ham Member QRZ Page

    I made a model of the current QST cube slot antenna, but for some reason wires 3 and 7 are 'not connected or crossing' to wire 11. I just don't see it. Can someone help me here? Here's the notepad edit.

    CM 2M slot cube
    GW 1 5 4.25 4.25 72 4.25 -3.75 72 0.25
    GW 2 5 -4.25 4.25 72 -4.25 -4.25 72 0.25
    GW 3 5 -4.25 -4.25 72 3.75 -4.25 72 0.25
    GW 4 5 -4.25 4.25 72 4.25 4.25 72 0.25
    GW 5 5 4.25 4.25 64.5 4.25 -3.75 64.5 0.25
    GW 6 5 -4.25 4.25 64.5 -4.25 -4.25 64.5 0.25
    GW 7 5 -4.25 -4.25 64.5 3.75 -4.25 64.5 0.25
    GW 8 5 -4.25 4.25 64.5 4.25 4.25 64.5 0.25
    GW 9 5 4.25 -3.75 72 4.25 -3.75 64.5 0.25
    GW 10 5 3.75 -4.25 72 3.75 -4.25 64.5 0.25
    GW 11 5 3 -4.25 72 3 -4.25 64.5 0.25
    GS 0 0 0.0254
    GE -1
    GN 2 0 0 0 5 0.002
    EX 0 11 3 0 1 0 0 0
    FR 0 0 0 0 146 0

    Attached Files:

  2. VA3RR

    VA3RR Ham Member QRZ Page

    I haven't run it up in 4NEC2, but just looking at the text file, I think wire 7 should read:

    GW 7 5 -4.25 -4.25 64.5 3.75 -4.25 72 0.25

    and wire 11 should read:

    GW 11 5 3.75 -4.25 72 3 -4.25 64.5 0.25
  3. N3DT

    N3DT Ham Member QRZ Page

    That doesn't work. Wire 7 has the same Z dimensions. And wire 11 is the feed, so it should have the same X1 and X2 dimensions to remain vertical. I've run wires into others before, I just don't get why it thinks they either don't connect or cross unless it's the thickness of wires 7 and 3, actually that may be it, I'll change the dimensions of wire 11 to reflect the thickness of the wires. The thing looks fine in the geometry view. Thanks.
  4. N3DT

    N3DT Ham Member QRZ Page

    I tried shortening wire 11 by .25" each end, and got no geometry errors, but then the wires were not connected. It just ended up being a short vertical dipole of unknown freq with the rest of the antenna disconnected. Baffled.
  5. AC6LA

    AC6LA Ham Member QRZ Page

    From the QST in Depth page for January 2019, here's a diagram:


    Those are the copper pipe cut lengths so with the corner fittings it looks like you need to add 1/2" (1/4" on each end) for each dimension. But that's not the main problem. Here's the EZNEC view of the NEC deck from post #1.


    Note that wire 11 joins wire 3 and wire 7 in the middle of a segment. That's a big no-no so EZNEC won't even bother to do the calculation, instead showing this:


    I'm really surprised 4nec2 does not catch that error. Wires 3 and 7 need to be broken into two wires each with wire 11 joining at the intersections.

    Dan, AC6LA
    KF6A and WZ7U like this.
  6. PG0DX

    PG0DX Ham Member QRZ Page

    Hello All,

    Dan, it does show the "error"
    However, you need to "validate" the model.
    That is done by pressing f3 (the geometry field) and then use the validate option.

    Its similair in "actions" as the average gain validation in Eznec.
    (although they have nothing to do with each other of course....just to explain "how" the steps works).

    Below is the "warning" 4nec2 provides: wavelength=2.053 mtr.

    Error: Wires 3 (tag 3) and 11 (tag 11) are crossing or not connected.
    Error: Wires 7 (tag 7) and 11 (tag 11) are crossing or not connected.

    Hope it was of use,

    Kind regards,
    Henry PG0DX

    (Edit...Almost forgot: you could set the function on "automatic" then it behaves like Eznec)
    W9XMT likes this.
  7. AC6LA

    AC6LA Ham Member QRZ Page

    Thank you, Henry. And I see now that Dave actually did mention those errors in his first post.

    Dan, AC6LA
    PG0DX likes this.
  8. N3DT

    N3DT Ham Member QRZ Page

    Thanks, that should get me going. Pretty sure I took care of the elbow overlap in my dimensions, but what I'll do is change all those numbers to symbols so I can optimize it. I just wanted to get it working first. I wanted to compare the pattern to other things I have.

    I have an idea of a 2M interlaced H Doublebay with both of them different enough so that when they're combined I get 50±j0, I've seen it done with full wave loops so it should be possible. Omni with gain too, at least horizontal. I think there was a QST article a while ago about the FW loops as an omni as I remember.
  9. KX4O

    KX4O Ham Member QRZ Page

    I'm interested in the gain you all find in your modeling. I get about 1.3 dBi freespace along and around the H-plane. That's not bad for an antenna this small and seems to jive with Chu-Wheeler limits, but the article suggests the gain is comparable to the regular J so wonder what I am missing.
  10. N3DT

    N3DT Ham Member QRZ Page

    I do all my modeling over my ground so I can compare similar antennas with my location. I got about equal gain and similar patter from the 2M slot and my 2M GP. I show about 2.5-2.7dB with the max gain at about 80° (10° above the horizon). So I think for the effort it's no better than a GP (or a JPole). But it is small at only about 8" high and 9" square. It could easily be covered with a plastic salad bowl. Which I guess was his reason. I still contend the GP is the easiest to make and feed.

    Once I got the connections straightened out it modeled fine but I ended up with slightly different dimensions than the author got. Probably just a function of how the optimizer worked on it.

Share This Page