Canadian TV, Computing and Home Theatre Forums banner

1 - 9 of 9 Posts

·
Registered
Joined
·
80 Posts
Discussion Starter #1
Have been working on a indoor antenna.

The DTV antennas that I have tried are very picky about
the location on the window still. Including vhf channesl

Only one spot on the window still works.

So I enhanced the vhf capability of the 4 bay bow-ties.

This vhf uhf 4 bay bow-tie has:

UHF: 10.3 to 11.84 to 11.27 db
VHF: 4.83 to 5.01 to db 5.08db

Here is the NEC file:

Code:
CM Kosmic SuperQuad 4-Bay, NO Reflector, 4nec2 by holl_ands, 1Mar2010
CM Simple SOURCE Wire. Autosegment(21).  Several "Too Sharp" Warnings.
CE
SY Rsrc=0.016       ' SOURCE wire Radius. Adjust for AGT=1.0: UHF=0.041 & HiVHF=0.041
SY Rbow=0.0404       ' Radius (in inches) of BOWTIE elements
SY Rfeed=0.0404       ' FEEDLINE wire Radius (in case they're different)
SY ZBowII=14       ' Distance between the Centers of the two INNER bowties
SY ZBowOI=14       ' From Center of INNER bowtie to Center of OUTER bowtie
SY BowLen=9.0        ' Bow Half-Length - Assume all SAME (Reality +/- 0.25+ in)
SY BowSep=7.00     ' Bow Tine Separation - Assume all SAME (Reality +/- 0.25 in)
SY FedSep=1.56       ' Separation (in inches) between two FEEDLINE wires
SY Hop=1.25              ' Separation between Feedlines at Crossover
SY ZCross=5.83       ' From Center of Feedline Crossover to Center of OUTER bowtie
SY ZClen=4.0       ' From Center of Feedline Crossover to Inflection point
SY Cond=1.67e7       ' Conductivity (Copper=3.0e7, Alum=2.0e7, StainlessSteel=1.67e7)
' Calculated from above INPUT Values:
SY ZBowInr=ZBowII/2          ' Distance from antenna center to center of INNER bowtie
SY ZBowOut=ZBowII/2+ZBowOI   ' Distance from antenna center to center of OUTER bowtie
SY Z1=ZBowOut+BowSep/2
SY Z2=ZBowOut
SY Z3=ZBowOut-BowSep/2
SY Z4=ZBowOut-ZCross+ZClen    ' Very long crossover region
SY Z5=ZBowOut-ZCross
SY Z6=ZBowOut-ZCross-ZClen    ' Very long crossover region
SY Z7=ZBowInr+BowSep/2
SY Z8=ZBowInr
SY Z9=ZBowInr-BowSep/2
SY YBowN=-FedSep/2
SY YBowP=FedSep/2
SY YBow=(BowLen^2-(BowSep/2)^2)^0.5
SY Ymax=YBow+FedSep/2
'  #    segs    X1      Y1    Z1    X2      Y2        Z2    radius
' SIMULATED BALUN SOURCE ON GW1:
GW 1    3    0.0    YBowN    0.0    0.0    YBowP        0.0    Rsrc
' GW2 Not used
' GW3 Not used
' INNER BOWTIES:
GW  4    21    0.0    Ymax     Z7    0.0        YBowP         Z8    Rbow
GW  5    21    0.0    Ymax     Z9    0.0        YBowP         Z8    Rbow
GW  6    21    0.0    YBowN     Z8    0.0        -Ymax         Z7    Rbow
GW  7    21    0.0    YBowN     Z8    0.0        -Ymax         Z9    Rbow
GW  8    21    0.0    Ymax    -Z7    0.0        YBowP        -Z8    Rbow
GW  9    21    0.0    Ymax    -Z9    0.0        YBowP        -Z8    Rbow
GW 10    21    0.0    YBowN    -Z8    0.0        -Ymax        -Z7    Rbow
GW 11    21    0.0    YBowN    -Z8    0.0        -Ymax        -Z9    Rbow
' OUTER BOWTIES:
GW 12    21    0.0    Ymax     Z1    0.0        YBowP         Z2    Rbow
GW 13    21    0.0    Ymax     Z3    0.0        YBowP         Z2    Rbow
GW 14    21    0.0    YBowN     Z2    0.0        -Ymax         Z1    Rbow
GW 15    21    0.0    YBowN     Z2    0.0        -Ymax         Z3    Rbow
GW 16    21    0.0    Ymax    -Z1    0.0        YBowP        -Z2    Rbow
GW 17    21    0.0    Ymax    -Z3    0.0        YBowP        -Z2    Rbow
GW 18    21    0.0    YBowN    -Z2    0.0        -Ymax        -Z1    Rbow
GW 19    21    0.0    YBowN    -Z2    0.0        -Ymax        -Z3    Rbow
' CROSS-OVER FEEDLINE:
GW 20     3    0.0    YBowP     Z2    0.0        YBowP         Z4    Rfeed
GW 21     3    0.0    YBowN     Z2    0.0        YBowN         Z4    Rfeed
GW 22     1    0.0    YBowP     Z6    0.0        YBowP         Z8    Rfeed
GW 23     1    0.0    YBowN     Z6    0.0        YBowN         Z8    Rfeed
GW 24     9    0.0    YBowN     Z6    Hop/2        0.0         Z5    Rfeed
GW 25     9    0.0    YBowP     Z4    Hop/2        0.0         Z5    Rfeed
GW 26     9    0.0    YBowP     Z6    -Hop/2    0.0         Z5    Rfeed
GW 27     9    0.0    YBowN     Z4    -Hop/2    0.0         Z5    Rfeed
GW 28     3    0.0    YBowP    -Z2    0.0        YBowP        -Z4    Rfeed
GW 29     3    0.0    YBowN    -Z2    0.0        YBowN        -Z4    Rfeed
GW 30     1    0.0    YBowP    -Z6    0.0        YBowP        -Z8    Rfeed
GW 31     1    0.0    YBowN    -Z6    0.0        YBowN        -Z8    Rfeed
GW 32     9    0.0    YBowN    -Z6    -Hop/2    0.0        -Z5    Rfeed
GW 33     9    0.0    YBowP    -Z4    -Hop/2    0.0        -Z5    Rfeed
GW 34     9    0.0    YBowP    -Z6    Hop/2        0.0        -Z5    Rfeed
GW 35     9    0.0    YBowN    -Z4    Hop/2        0.0        -Z5    Rfeed
GW 36     11    0.0    YBowP     Z8    0.0        YBowP        0.0    Rfeed
GW 37     11    0.0    YBowN     Z8    0.0        YBowN        0.0    Rfeed
GW 38     11    0.0    YBowP    -Z8    0.0        YBowP        0.0    Rfeed
GW 39     11     0.0    YBowN    -Z8    0.0        YBowN        0.0    Rfeed
GS 0 0 0.0254            ' Convert above from INCHES to METERS for NEC
GE 0                    ' No Ground Plane
EK 0                    ' Enable Extended Kernel
LD 5 0 0 0 Cond 0            ' Conductivity
EX 0 1 2 0 1 0            ' GW1 is SOURCE wire
GN -1                    ' Free Space
' FR Freq Sweep choices in order of increasing calculation time (fm holl_ands):
' FR 0 0 0 0 470 0        ' Fixed Freq
FR 0 29 0 0 470 12        ' Freq Sweep 470-806 every 12 MHz - OLD UHF BAND
' FR 0 34 0 0 410 12        ' Freq Sweep 410-806 every 12 MHz - Even Wider Sweep
' FR 0 39 0 0 470 6        ' Freq Sweep 470-698 every 6 MHz - PREFERRED FOR UHF
' FR 0 77 0 0 470 3        ' Freq Sweep 470-698 every 3 MHz
' FR 0 153 0 0 470 1.5        ' Freq Sweep 470-698 every 1.5 MHz
' FR 0 71 0 0 300 10        ' Freq Sweep 300-1000 every 10 MHz - WIDEBAND SWEEP
' FR Hi-VHF choices:
' FR 0 15 0 0 174 3        ' Freq Sweep 174-216 every 3 MHz
' FR 0 29 0 0 174 1.5        ' Freq Sweep 174-216 every 1.5 MHz - PREFERRED
' FR 0 43 0 0 174 1        ' Freq Sweep 174-216 every 1 MHz - Hi-Rez
' FR 0 26 0 0 150 6        ' Freq Sweep 150-300 every 6 MHz - WIDEBAND SWEEP
' FR Lo-VHF choices:
' FR 0 35 0 0 54 1        ' Frequency Sweep every 1 MHz for Ch2-6
' FR 0 36 0 0 75 1        ' Frequency Sweep every 1 MHz for Ch5 + Ch6 + FM
' FR 0 28 0 0 54 6        ' Wide Freq Sweep every 6 MHz for Ch2-13
' FR 0 64 0 0 54 12        ' Super Wide Freq Sweep 54-810 every 12 MHz
' RP choices in order of increasing calculation time:
' RP 0 1 1 1510 90 90 1 1 0 0    ' 1D Gain toward 0-deg Azimuth - SIDE GAIN
' RP 0 1 1 1510 90 0 1 1 0 0        ' 1D Gain toward 90-deg Azimuth - FORWARD GAIN
' RP 0 1 1 1510 90 180 1 1 0 0    ' 1D Gain toward 270-deg Azimuth - REVERSE GAIN
' RP 0 1 37 1510 90 0 1 5 0 0     ' 2D (Left only) Azimuthal Gain Slice
RP 0 1 73 1510 90 0 1 5 0 0         ' 2D Azimuthal Gain Slice - PREFERRED
' RP 0 73 1 1510 90 0 5 1 0 0     ' 2D Elevation Gain Slice
' RP 0 73 73 1510 90 0 5 5 0 0     ' 3D Lower Hemisphere reveals antenna (use Fixed Freq)
' RP 0 285 73 1510 90 0 5 5 0 0    ' 3D Full Coverage obscures antenna (use Fixed Freq)
EN
 

·
Registered
Joined
·
120 Posts
Hi Lassar
Your modifications, to the original Kosmic superquad, increases raw gain for Hi-VHF, while increasing the SWR and the impedance mismatch. I think the minor increase in raw gain will be offset by the increase in SWR and the impedance mismatch. I think that you will find the original design will give you just as good, if not better results. Your actual gain is always less than your raw gain. The lower the SWR, the less the gain will be reduced. Impedance mismatch will have more affect, as your cable run, from the antenna to the receiver, increases. The original Kosmic Superquad is a pretty mature design and it will be difficult to improve on, without changing the basic design. Adding a loop allows you more control over SWR and impedance in the Hi-VHF band.
 

·
Registered
Joined
·
4,870 Posts
Hi Lassar,
As ljavener mentioned... he is spot on.
If you are using 4nec2, note that the gain figures are the raw gain, before any mismatch loss due to vswr. Cheap and dirty way to rescale any antenna by adding one new symbol and modifying one line of Nec code.
I found holl_ands' original model from his website and get a much nicer looking freq response by frequency scaling it. I look for the existing minimum SWR in the UHF band, then shift that to where I'd want it (eg in the center of the new UHF Band). The UHF Net gain is now better than 10 dBi net across the new UHF Band. And VHF Remains about
2.1 dBi Net. Roughly the same as a dipole.
I am using Nikiml's python scripts in linux. That gives ya the raw and net gain figures.
I also sometimes run the model thru 4nec2 in 'wine' in order to get #segs from it's AGT test as was explained by holl_ands some time ago.

Code:
lassar model
171.0-219.0 net gain - ave 1.56dBi
171.0-219.0 raw gain - ave 4.95dBi

holl_ands rescaled
171.0-219.0 net gain - ave 2.10dBi
171.0-219.0 raw gain - ave 3.49dBi

467.0-611.0 net gain - ave 11.39dBi
467.0-611.0 raw gain - ave 11.66dBi
...
...
...
CE
SY Fscale=647/533 'old min vswr / desired new min vswr'
SY Rsrc=0.041
...
...
...
GS 0 0 0.0254*Fscale' Convert above from INCHES to METERS for NEC
 

·
Registered
Joined
·
4,870 Posts
I have never seen an antenna model where the Net gain = Raw gain, despiite AGT=1.0.
Unless the VSWR was extremely low across the entire bandwidth, the two figures may get pretty close.
eg - for practical purposes, I am sure there is a point of limited return for swr, but it should be possible to get AGT=1.0 if we are following holl_ands' guidance.

Not an expert by any means, but my understanding of the Average Gain Test is to determine how well the model 'converges' in terms of segments per wavelength, which affects the model accuracy, a measure of generating an accurate radiation pattern. All has to do with how NEC does it's calculations I guess. Some reading from cebik may help. Full disclosure,
I have not read the entire thing myself.
A quote from cebik...
The average gain figure that results from the test may be higher or lower than 1.0. One proposed gradation of model merit uses the following dividing points:
GAVE Value Range Significance
0.95 - 1.05 Model is considered to have passed the test
and is likely to be highly accurate.
0.90 - 0.95 and 1.05 - 1.10 Model is quite usable for most purposes.
0.80 - 0.90 and 1.10 - 1.20 Model may be useful, but adequacy can be
improved.
<0.80 and >1.20 Model is subject to question and should be
refined.


The user may develop more strict limits for the adequacy of a model based on the specific tasks within which the model plays a role.
 

·
Registered
Joined
·
120 Posts
Doesn't net gain = raw gain minus AGT
Let me give my beginners try, to explain this as best as I can. Hopefully one of the experts will let us know if I'm on the right track.

AGT is a function of the software, and how well it will properly calculate your model. It gives you an idea of what effect your model is going to have on the results(better model-better AGT-better results). This is definitely one thing that effects your results.

The SWR number also effects the results of your antenna. I can honestly say that I don't fully understand SWR. What I do know is that your actual gain is affected by the SWR. The closer to 1.0 the SWR, the less that it affects the results. It looks like majortom is using "Nikiml's python scripts in linux" additional software available on Nikiml's website, to calculate actual gain in addition to raw gain. I haven't had a chance to look into this. What I know is that lower SWR number are better, and that SWR of 2.7 seems to be a target to stay below. Yurii posted a spreadsheet that showed how much reduction you will see, for different SWR figures. Click on SWR_Filtering to bring up spreadsheet. I'm not sure if this is valid for all antennas, or just specific ones. It shows that a SWR of 10 will cause approximately -5db drop in actual gain. SWR of 2.7 seems to cause a reduction of approximately 1db, so that may be why it's a common target.

There's a lot to be considered when designing a DIY antenna. It takes a lot of time to understand what all of the numbers mean, and how each is affected by changes to your antenna. Keep in mind that all of the experts were beginners at one time. Keep trying, and keep asking questions, but also spend some time reading the information that's out there, before you frustrate yourself making designs that don't work. As majortom stated, click on The Average Gain Test to see part of a very good tutorial.

Lawrence
 

·
Registered
Joined
·
80 Posts
Discussion Starter #7
After putting this NEC file thru the opimizer wringer,
forgot to adjust source wire to put AGT close to 1 (.99 to 1.01).

After adjusting the source wire, so AGT is almost 1., the VHF varies from
180 Mhz: 4.43 db to 216 Mhz: 4.72 db

Channel 11 (201 Mhz ) comes in at 4.6 db.

Mclapp 9.5x9: 201 Mhz comes in at 3.42 dB.

This antenna should give me 1.18 more db then the Mclapp m4.
 

·
Registered
Joined
·
80 Posts
Discussion Starter #8
Doesn't 4nec2 take swr & impedance into account, when it calculates the antenna gain?
 

·
Registered
Joined
·
120 Posts
Doesn't 4nec2 take swr & impedance into account, when it calculates the antenna gain?
The simple answer is no. I'm not sure if 4nec2 can or could calculate actual gain, but I know that it does not. Some of the python scripts, written by Nikiml, can take SWR into account, and give raw and actual gain figures.

Impedance mismatch is harder to calculate, because it's affect is downstream of the antenna. If the impedance of the antenna does not match the impedance rating of your wiring/cabling, between your antenna and your receiver, you will have even more loss. Most television antennas are designed for 300ohm. Signal is then converted to 75ohm with a 4:1 balun. Signal is then sent down 75ohm rated cable to the receiver. The farther the impedance matching is, from this design spec of 300ohm, the more signal loss you will have in your cabling. You always want to get the best impedance match that you can, but I would be less concerned if your antenna location is very close to your receiver. The longer the cable, the more loss the mismatch will cause. Remember that even if impedance is perfectly matched, your going to have some signal loss, per foot of cable. The loss is just worse, if the impedance of the antenna, does not match the cabling system(remember most antenna systems use 75ohm cables, but are designed for 300ohm, because we use a 4:1 balum to convert 300ohm to 75ohm).

It's almost impossible to get a perfect SWR and impedance match, at all frequencies. Better numbers are going to create a better antenna design. You are going to have loss, due to high SWR, no matter how long the cable is, so I'd concentrate on lowering the SWR as much as possible, and try to keep impedance within 1/2 and 2 times the designed specification(usually 300ohm).
 
1 - 9 of 9 Posts
Top