3/8 wavelength , an experience

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by IK3UMT, Jun 8, 2010.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-rl
ad: l-gcopper
ad: L-rfparts
ad: Subscribe
  1. IK3UMT

    IK3UMT Ham Member QRZ Page

    HI all,

    I just built a 3/8 wavelenght vertical (HyGain AV-640, Cushcraft R8 and so on principle)
    It is a simple monoband version with a single radiator for 10mt band, just for evaluation.
    I used an old Cushcraft MN8 (R8) matching box rebuilt with a 9 turns RG142 on FT240-43 core choke connected to a 10 turns bifilar winding on 2 stacked FT240-61 core 4:1 transformer, feeding a 11,3 ft vertical wire through a 18pF capacitor.
    Counterpoise is made by 6x3,3ft stainless steel radials

    SWR is within 1:1,5 28 to 29 MHz
    The antenna is placed 13ft above ground

    As propagation is very poor , i have only listened to few EU stations (no qso yet)

    I built the antenna to evaluate its performances versus the antenna I actually use , a Butternut HF9V over a 30X33ft elevated radials groundplane (10ft above real ground, not so much "elevated"...)
    If true "the higher the better", an AV-640 @ 30ft heigth would be a better solution for higher bands

    The first things left me disappointed was the higher noise level than HF9V and, worse, the lower signal received (1 to 2 S point less)
    Could this be due to overall antenna system dimension ??

    I hope in a better 10m opening tomorrow morning to better check performance and comparision.

    Any comment is very appreciated

  2. K6CLS

    K6CLS Ham Member QRZ Page

    thanks for posting your experiences with this. very interesting.
  3. K4EEZ

    K4EEZ Ham Member QRZ Page

    yup tnx for the post
    i am still trying to work out a myth or true deal
    a loop wire antenna is spouse to be low noise as its a DC short
    right now I have an inverted Vee wire from #14 THHN with jacket on cut for 40 meters, it is up at center point 21 ft, it talks well and Rx well (East/West) but floor noise is always S7 or more... anyone below a good S7 and forget it, but i do try and pull the signals out and work Dx last station i spoke to was J88CF out in the St. Vincent & The Grenadines Villa, so it works but the noise floor is bad...

    so i need to bust a myth, will a loop wire be better for lower floor noise or ?

    but good on the Vertical antenna, i have a Gap Titan Dx, that works well for DX not real good for local but for Dx its good but again its the noise floor thats a killer... 73
  4. KI6J

    KI6J Ham Member QRZ Page

    Ciao Federico,

    Thank you for sharing your experiment. 3/8 wl. sounds interesting.

    First, I have a few questions about the principle of operation. If you feed the element through an 18pf capacitor, are you reducing the electrical length to 1/4 wl.? If so, then what is the purpose of the 4:1 balun? I understand the feedpoint impedance of a 1/4 wl. ground plane to be about 35 ohm. Clearly there is something that I am missing, since your results show reasonable SWR.
    Second, have you done antenna modeling with the 3/8 element? Is there a gain advantage anywhere in the elevation plot that is superior to the 1/4 wl. element? Do you know what the theoretical feedpoint impedance of the 3/8 element before it is transformed by your matching network?

    I've thought about the 3/8 wl. radiator, but considered bottom loading it, resonating it as a half wave with no ground image needed, and tapping the loading coil at the 50 ohm point.

    I look forward to the results of your testing.

  5. G4LNA

    G4LNA Ham Member QRZ Page

    Hi Federico,

    I have been playing around here with the antenna program, I tried entering in your 3/8 wave and compared it to an normal 1/4 wave with slopping radials. There appears to be very little difference between the 3/8 and the 1/4 in the gain that you would notice, it seems less than 1db.

    Have you tried doing away with all the matching networks at the bottom and connecting the feeder straight onto the element and radials? The reason I say that is when I had finished the results where as follows:


    Notice the feed impedance Z: 72.552 -j9.798 Ohm which give a SWR of 1.5 for 50 Ohms which is a pretty close match so that any matching unit and feeder can handle with little loss at that frequency.

    Just one other observation about these antennas, they both have a very low angle of radiation, which is great when you are working towards the horizon, but if you are working Sporadic E then the cloud can quite often be at a high angle, that's where my cloud warmer comes into its own :)
  6. IK3UMT

    IK3UMT Ham Member QRZ Page

    A 3/8 radiator presents an impedance of 200-250 ohm +Jsomething (inductive reactance) , the purpose of capacitor is to resonate that inductance to leave the only resistive part.
    This will be matched to feedline by mean of 4:1 transformer.
    Further, since from feedpoint two equal and opposite currents must flow,the current opposite to vertical radiator one, would flow down the coax rather than along little radials (coax braid is the easiest way) so a common mode choke is mandatory, and it will dissipate a quite good part of power supplied.
    That's why RG142 was used , there is some heat to manage !

    My intention is to evaluate this kind of antenna versus my actual HF9V over its good groundplane.
    If results were good, a so light and slim antenna (its multiband commercial version) would be easy to be installed high on the clear.

    Unfortunately, first tests aren't so fair,
    Antenna is noisy ( 1-2 point S more of noise vs HF9V)
    Received and transmitted signal is always at least 2 S points less than reference antenna.

    I remember you , HF9V is installed 11.5ft above ground, this one is 13ft.
    They are 46ft far each other on clear space.

    This is a 10mt version, probably performances would be better on other bands versions.....

    That's all

  7. IK3UMT

    IK3UMT Ham Member QRZ Page

    This is a lucky limit case:
    If you consider the whole antenna lenght, 3/8 radiator plus 1mt radial, you are very near to half wavelenght on 10mt
    That's why of such a good matching to 50ohms , it is just like to move feedpoint along a vertical half wavelenght dipole.
    Consider instead, that 48" whips are used in commercial multiband version 6 to 40mt , if you try to simulate a 20 mt 3/8 lambda radiator fed against a 48" diameter groundplane you'll be so far from a good direct match.

    These antennas are mostly based on end-fed principle (but not halfwave end-fed) over a restricted groundplane to collect return current.

    Yes, these antennas have a good takeoff angle , if installed high above ground it becomes even lower with many higher angle lobes : try with MMANA @ 15-20mt from ground.....

  8. IK3UMT

    IK3UMT Ham Member QRZ Page

    ERRATA !!!

    ERRATA !!, now it works !!!

    Wrongly, following the 4:1 schematic where unbalanced line ground is connected to common point of two windings, i have connected 4:1 common point to radials plate, my mistake....

    The common point instead has to be connected to the hot pole of choke.

    With this configuration , all values were returned back to initial simulation ones.

    I've got to replace capacitor with a 12 pF one (as modeled) and bring radiator back to 12.8ft
    Further I've added an air wound RFC between radiator and radials as per original schematics.

    Now antenna has a 1:1 SWR at 28.300 MHz (R=47 X=0 on MFJ259) and within 1:1,3 28 to 29 MHz

    Even better, it is no more noisy and RX and TX signal are almost the same as my Butternut HF9V, sometimes even higher due to some difference in radiation angle.

    I'm so curious to check/compare on DX qso, but propagation is bad these days.

    Next step will be to put antenna 80 to 100ft high from ground to see if "the higher, the better" is really true.....

Thread Status:
Not open for further replies.

Share This Page

ad: elecraft