9 band antenna? Anything is possible!

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by SP3L, Oct 28, 2015.

ad: L-HROutlet
ad: l-rl
ad: L-MFJ
ad: Subscribe
ad: Left-2
ad: Left-3
ad: FBNews-1
  1. AC6LA

    AC6LA Ham Member QRZ Page

    Jacek,

    To calculate efficiency you can use NEC's average gain function. From the EZNEC Help file:
    The same thing is discussed in the fifth bullet here:

    http://ac6la.com/aeuse3d.html

    I'm afraid I can't remember all the details of doing this with 4nec2, sorry.

    Here's a comparison of efficiency for 3 models: 1) The SP3L dipole as shown in post #1 with a 9:1 balun. 2) A T2FD with a 16:1 balun, using Danny's suggestion to increase the length and hence reduce resistor losses at the low frequencies. The length was made the same as the SP3L dipole, 53.4 m (175 ft). To annoy Carl I did not make any other changes to the T2FD such as tilt angle, wire spacing, or phase of the moon. 3) Danny's 80-10 OCFD using the model that he kindly posted.

    All models include 100 ft of RG-8X with realistic loss. The SP3L and T2FD were at the same height, 12m (39.37 ft). Danny's OCFD was left at the model height of 30 ft. That made only a minor difference in the calculated efficiency. All models were over the same Average ground.

    All models assumed a perfect (loss-less) balun. I don't know how to include balun loss in a model, perhaps G3TXQ can jump in with some suggestions.

    SP3La.gif

    Dan, AC6LA
     
    KC8VWM and SP3L like this.
  2. G3TXQ

    G3TXQ Ham Member QRZ Page

    That's quite difficult!

    I would expect that in these applications the core loss due to common-mode signals would dominate, rather than the winding loss due to differential-mode signals. To include common-mode losses in the model you would need to know the common-mode impedance of the balun at each frequency, and you would need to be able to model the feedline length/routing/grounding accurately.

    Steve G3TXQ
     
    KC8VWM likes this.
  3. W1YW

    W1YW Ham Member QRZ Page

    This is wrong.

    You have to get the loss from three factors for antenna system efficiency as considered in this thread:

    1) ohmic loss of 'antenna'
    2) mismatch loss balun to antenna
    3) insertion loss of balun.

    Coaxial insertion loss would be nice to include, but is common to all antennas in any case, thus not relevant in said comparision. Its certainly true that most hams are not aware that 3 dB losses are common with even good coax and longer lengths-- see E 74 M data.

    Hint: you can get 1) by comparing each antenna to itself with perfect conductors and then copper conductor, for the gain, at each frequency.

    73
    Chip W1YW
     
    Last edited: Nov 15, 2015
  4. W1YW

    W1YW Ham Member QRZ Page

    A rare example: Carl and I are in absolute agreement here.

    73
    Chip W1YW
     
  5. SP3L

    SP3L Ham Member QRZ Page

    Dan,
    it was simpler than I thought. 4nec2 calculates antenna radiation efficiency every time you generate the radiation pattern. But this result is not shown in the plot window but in the main window of the program (see the blue ink value in the figure below).
    upload_2015-11-15_12-41-38.png

    The "Efficiency" (a line above in the main window) is a ratio of the radiated power versus supplied power. The "Radiation Efficiency" takes additionally into account losses in the ground. So, this is a more interesting value for the antenna user.

    And here are the antenna radaition efficiency results (without losses in the cable and in the balun) for SP3L and monoband dipoles at 12 m height as calculated by 4nec2:

    Frequency SP3L [%] Dipole [%]
    3.750 76.12 69.7
    7.150 72.18 77.92
    10.125 80.03 75.86
    14.175 64.65 72.55
    18.118 68.83 73.6
    21.225 73.97 75.85
    24.940 74.33 73.91
    28.500 59.63 72.41

    upload_2015-11-15_13-1-34.png
     

    Attached Files:

  6. KM1H

    KM1H Ham Member QRZ Page

    Nope Im not annoyed as Ive grown to expect that from you when the work required becomes too hard so all you use is one of the worst examples of the junk on the market. Reminds me of a certain W8 who does even less work before the insults start to cover up and apply damage control.

    Im in deep shock Chip but QRZ and Eham antenna forums can do that to people! Too many times I need hip waders to get thru some of the pure BS sprouted here/there over the years.

    OTOH the ones honestly looking for help which can be newbies to old timers, the several who understand the subject, the few who really and deeply understand the subject and are willing to help are the high points.

    Carl
     
  7. AC6LA

    AC6LA Ham Member QRZ Page

    Hi Jacek,

    I've done a little boning up on the way 4nec2 reports efficiency. It is true that what 4nec2 shows as "Efficiency" (aka "Power Efficiency") takes into account wire loss and load loss (if any). And then "Radiation Efficiency" includes those loss factors as well as ground loss.

    "Radiation Efficiency" is computed from the Average Power Gain as shown at the very end of the NEC output listing, divided by 2 unless the model is in free space.

    SP3Lb.png

    However, that calculation assumes that the model is absolutely perfect. To get a slightly more accurate figure for "Radiation Efficiency" you need to determine how far off the model is from perfection, usually due to things like bad source placement or angled wire junctions near the source. To do that you run the "Average Gain Test".

    SP3Lc.png

    4nec2 will temporarily remove all loss from the model; that is, the model will have no wire loss, no load loss, and no ground loss. When the model has no loss elements the radiated power should be exactly equal to the supplied power. If it's not, the model has some imperfection (which may be very minor).

    As before, look at the Average Power Gain as shown at the very end of the NEC output listing, divided by 2 unless the model is in free space.

    SP3Ld.png

    After dividing by 2 a perfect model with no loss elements would show an Average Power Gain of exactly 1.0. That figure is also called the "Average Gain Test" (AGT) value. To get a more accurate Radiation Efficiency you can divide the original factor (0.7173) by the "imperfection factor" (1.076) to get 66% efficiency vs 71%. Note that in some cases the "no loss" Average Power Gain will be less than 1, not greater than 1. In that case the original efficiency would have been under-estimated instead of over-estimated.

    Please do not misinterpret this post. It is in no way a criticism of your model. It's very rare that a model is absolutely perfect and in any case the difference between 66% and 71% efficient is pretty meaningless. I was just curious as to why the slight discrepancy between what 4nec2 shows vs the methodology recommended by W7EL.

    BTW, I'm using the model that you supplied in post #1. Apparently that's not quite the same as what you are currently using.

    Dan, AC6LA
     
    EA1DDO likes this.
  8. G3TXQ

    G3TXQ Ham Member QRZ Page

    Just to expand on Dan's comments above, Cebik quoted the following "rule of thumb" for assessing model accuracy:
    • AG>0.8dB "Model is subject to question and should be refined"
    • AG>0.4dB "Model may be useful, but adequacy can be improved"
    • AG>0.2dB "Model is quite usable for most purposes"
    • AG<0.2dB "Model is considered to have passed the test and is likely to be highly accurate"
    Steve G3TXQ
     
  9. KF6A

    KF6A Ham Member QRZ Page

    It should also be noted that, "the Average Gain Test is a necessary and not a sufficient condition of model adequacy". - W4RNL
     
  10. SP3L

    SP3L Ham Member QRZ Page

    Thank you very much, Dan. Your feedbacks are invalueable. And your knowledge of antenna simulation tools is really awesome. Together with the other posters (e.g. G3TXQ) you make this thread very interesting to read.


    Meanwhile, I thought a little bit about voltage rating for the balun.
    I simulated all three versions of the antenna again to find out their highest impedances (magnitude), in order to calculate what voltage must be expected at the antenna terminals when power is applied. And here are the results.

    Antenna input impedance (magnitude) [ohm]
    upload_2015-11-16_11-46-0.png
    This correspondes to the following voltage levels [Vrms] for maximum antenna impedance:
    upload_2015-11-16_11-47-51.png

    And the same for the 6 m band - impedance magnitude [ohm]:
    upload_2015-11-16_12-7-2.png
    and voltage [Vrms]:
    upload_2015-11-16_12-8-2.png


    This is a high impedance antenna and high voltage will be present at the balun terminals in all three versions. Up to 200W, the voltage should not be a huge problem but making a balun for this antenna for 1 or 1.5 kW power output can be challenging. I wonder. if amongst the readers, there are the hams experienced with balun design and making? Are the baluns for 1 kW feasible?
     

Share This Page

ad: chuckmartin