Help with a 4NEC2 antenna model

[A.Z.]

Hello all, I'm unsure if this is the right section to post this, if not please let me know and I'll move it elsewhere

As for the topic, days ago an old friend of mine which is a "hardcore" CBer asked me if I could model his "Sirio Vector 4000" (the antenna is the same as the "Avanti Sigma IV" and others), so I found some time and wrote a 4NEC2 model file for such an antenna, the model is the following

Code: [Select]


CM Avanti Sigma 4 / Sirio Vector 4000
CM use "auto segmentation" in NEC
CE

' base values
SY freq=27.210                  ' calc frequency
SY test=freq                    ' test frequency

' materials
SY alum=37700000                ' alluminium
SY copp=58000000                ' copper

' pipes
SY pip0=0.0055                  ' link at cage bottom
SY pip1=0.0080                  ' 16   mm diam
SY pip2=0.0100                  ' 20   mm diam
SY pip3=0.0125                  ' 25   mm diam


SY wfac=1                       ' shortening factor
SY wave=(300/freq)*wfac         ' Lambda

SY ftop=0.50                    ' upper section factor
SY fbot=0.25                    ' lower section factor
SY ltop=(wave*ftop)             ' upper section length
SY lbot=(wave*fbot)             ' lower section length

SY ffac=0.81                    ' feedpoint adj
SY flow=(lbot*ffac)             ' lower split
SY fupr=(lbot-flow)             ' upper split

SY hfac=0.61                    ' height factor
SY base=(wave*hfac)             ' base from ground
SY hbot=flow+base               ' lower section apex
SY hmid=hbot+fupr               ' upper section apex
SY htop=hmid+ltop               ' antenna apex

SY rfac=0.2754                  ' cage calc factor
SY rads=(wave*rfac)             ' cage elements length
SY urad=(base+rads)             ' ring height

SY stub=0.02                    ' cage link at bottom
SY uspc=0.336                   ' ring radius

SY segl=31                      ' long elements
SY segm=5                       ' midsize elements
SY segs=1                       ' short elements

SY fedw=1                       ' feed element
SY feds=segm                    ' feed segment

' ring data
SY rad=uspc
SY inc=30
SY a01=15
SY a02=a01+inc
SY a03=a02+inc
SY a04=a03+inc
SY a05=a04+inc
SY a06=a05+inc
SY a07=a06+inc
SY a08=a07+inc
SY a09=a08+inc
SY a10=a09+inc
SY a11=a10+inc
SY a12=a11+inc

' vertical elements
GW  1 segm      0         0  base     0       0 hbot pip2
GW  2 segm      0         0  hbot     0       0 hmid pip3
GW  3 segl      0         0  hmid     0       0 htop pip3
               
' cage links
GW  4 segs      0         0  base  stub       0 base pip0
GW  5 segs      0         0  base -stub       0 base pip0
GW  6 segs      0         0  base     0    stub base pip0
GW  7 segs      0         0  base     0   -stub base pip0
               
' cage elements
GW 10 segm   stub         0  base  uspc       0 urad pip1
GW 11 segm  -stub         0  base -uspc       0 urad pip1
GW 12 segm      0      stub  base     0    uspc urad pip1
GW 13 segm      0     -stub  base     0   -uspc urad pip1

' cage ring
GW 31 3  rad*sin(a01) rad*cos(a01) urad   rad*sin(a01+inc) rad*cos(a01+inc) urad pip1
GW 32 3  rad*sin(a02) rad*cos(a02) urad   rad*sin(a02+inc) rad*cos(a02+inc) urad pip1
GW 33 3  rad*sin(a03) rad*cos(a03) urad   rad*sin(a03+inc) rad*cos(a03+inc) urad pip1
GW 34 3  rad*sin(a04) rad*cos(a04) urad   rad*sin(a04+inc) rad*cos(a04+inc) urad pip1
GW 35 3  rad*sin(a05) rad*cos(a05) urad   rad*sin(a05+inc) rad*cos(a05+inc) urad pip1
GW 36 3  rad*sin(a06) rad*cos(a06) urad   rad*sin(a06+inc) rad*cos(a06+inc) urad pip1
GW 37 3  rad*sin(a07) rad*cos(a07) urad   rad*sin(a07+inc) rad*cos(a07+inc) urad pip1
GW 38 3  rad*sin(a08) rad*cos(a08) urad   rad*sin(a08+inc) rad*cos(a08+inc) urad pip1
GW 39 3  rad*sin(a09) rad*cos(a09) urad   rad*sin(a09+inc) rad*cos(a09+inc) urad pip1
GW 40 3  rad*sin(a10) rad*cos(a10) urad   rad*sin(a10+inc) rad*cos(a10+inc) urad pip1
GW 41 3  rad*sin(a11) rad*cos(a11) urad   rad*sin(a11+inc) rad*cos(a11+inc) urad pip1
GW 42 3  rad*sin(a12) rad*cos(a12) urad   rad*sin(a12+inc) rad*cos(a12+inc) urad pip1


GE  1

GN  2  0  0  0  13  0.005

LD  5   0  0  0  alum

EK

EX  0 fedw feds 0 1 0 0

FR  0  1  0  0  test  0

EN


now, the model works w/o problems, even if I was forced to set "auto segmentation" in NEC due to the cage elements junctions creating warnings, but the problem is that running the model I get a gain of 5.89dBi (see attached image) which seems too good to be true, so I'm here asking if someone with more expertise than me in 4NEC2/NEC would be so kind to revise the model and tell me where the error is

[tomnut]

Nice coding in the 4nec2 model!

NEC2 is sensitive to segments at acute angles like this, especially near the feedpoint.
I can't see how many segments you have, but it's possible the bottom four are almost overlapping where the meet.

The right way to model this would be to extend a single wire upwards for some distance. Make it a bit fatter if you like (but respect the length/diameter ratio, at least 2:1).
Then at some point say 1/5 of the way up, branch out with four wires at about 45 degrees, and then continue with the almost-parallel wires all the way to the ring.
Maybe you have done this, I can't quite tell from the file.

You could also just model the ring with four wires, no need to make it circular, the effect will be very small.

To test the model, apart from throwing no errors,
- switch to a perfect ground plane
- turn on integration over radiation pattern
- confirm that the integrated radiated power is the same as the input power (I think... could be a factor of 2 because of the hemisphere)

[A.Z.]

Thank you, nothing special, I'm far below the level of knowledge of people like Owen Duffy or Cebik, my only merit is using a vanilla editor to create/modify the NEC models, instead of using the built-in "easy" editor

As for the segments, they are defined by the "segl", "segm" and "segs" symbols in the model, the warnings I'm getting from NEC are the following

Code: [Select]

Warn.: Wire 1, seg 5 (tag 1), EX-src: Nearby segm-length or radius differs too much.
Error: Wire 1, seg 1 (tag 1), seg-len (0.489) differs more than 5 * seg-len (0.02) for wire 4
Error: Wire 1, seg 5 (tag 1), seg-len (0.489) differs more than 5 * seg-len (0.073) for wire 2
Error: Wire 2, seg 6 (tag 2), seg-len (0.073) differs more than 5 * seg-len (0.489) for wire 1
Warn.: Wire 2, seg 6 (tag 2), len/rad at junction (5.842) below 6.
Warn.: Wire 2, seg 10 (tag 2), len/rad at junction (5.842) below 6.
Warn.: Wire 4, seg 42 (tag 4), len/rad at junction (3.636) below 6.
Error: Wire 4, seg 42 (tag 4), seg-len (0.02) differs more than 5 * seg-len (0.622) for wire 8
Warn.: Wire 4, seg 42 (tag 4), len/rad at junction (3.636) below 6.
Warn.: Wire 5, seg 43 (tag 5), len/rad at junction (3.636) below 6.
Error: Wire 5, seg 43 (tag 5), seg-len (0.02) differs more than 5 * seg-len (0.622) for wire 9
Warn.: Wire 5, seg 43 (tag 5), len/rad at junction (3.636) below 6.
Warn.: Wire 6, seg 44 (tag 6), len/rad at junction (3.636) below 6.
Error: Wire 6, seg 44 (tag 6), seg-len (0.02) differs more than 5 * seg-len (0.622) for wire 10
Warn.: Wire 6, seg 44 (tag 6), len/rad at junction (3.636) below 6.
Error: Wire 7, seg 45 (tag 7), seg-len (0.02) differs more than 5 * seg-len (0.489) for wire 1
Warn.: Wire 7, seg 45 (tag 7), len/rad at junction (3.636) below 6.
Error: Wire 7, seg 45 (tag 7), seg-len (0.02) differs more than 5 * seg-len (0.622) for wire 11
Warn.: Wire 7, seg 45 (tag 7), len/rad at junction (3.636) below 6.
Error: Wire 8, seg 46 (tag 10), seg-len (0.622) differs more than 5 * seg-len (0.02) for wire 4
Error: Wire 9, seg 51 (tag 11), seg-len (0.622) differs more than 5 * seg-len (0.02) for wire 5
Error: Wire 10, seg 56 (tag 12), seg-len (0.622) differs more than 5 * seg-len (0.02) for wire 6
Error: Wire 11, seg 61 (tag 13), seg-len (0.622) differs more than 5 * seg-len (0.02) for wire 7

and I've tried changing the segments to no luck, so I was forced to enable "auto segmentation" even if I don't like that, as for the problematic areas, I'm attaching a pic showing them

Hope that you may spot where my mistakes are, because sincerely I'm here scratching my head