An Experiment Closer to the Ground(ing)

Discussion in 'Antennas, Feedlines, Towers & Rotors' started by KK4NSF, Jan 15, 2019.

ad: L-HROutlet
ad: l-rl
ad: Left-2
ad: MessiPaoloni-1
ad: L-MFJ
ad: Subscribe
ad: Left-3
  1. KK4NSF

    KK4NSF Ham Member QRZ Page

    View attachment 519712 View attachment 519713 After our unsuccessful but very funny experiment using a drone-based vertical antenna, my friends and I did some research and came across the work of Rudy Severns (N6LF) and Al Christman (K3LC) which explored the use of radials of various lengths and numbers. N6LF's work was especially enlightening, since it dealt with the improvements radials make on physically short verticals such as Hamsticks, base-loaded whips and so forth. (Hence my question last week about hamsticks vs base-loaded whips... I wanted you guys' opinions.)

    So after reading the papers, and your opinions we did our own experiment to see what we could achieve. The goal here was to see how good we could make a physically short antenna perform, with an eye toward using it on our upcoming MiniDXpedition to Ship Island where there are no trees or much of anything to work with, and everything has to fit in backpacks and carried about half a mile across sand dunes.

    As you can see from the pics, the antenna is simply a tripod mounted vertical with a radial plate, and sixteen 25' radials. The pic only shows the Hamstick, but we also tested a base loaded vertical 8' whip. For test equipment, we didn't have a lot to work with except a Kenwood with an SWR, a Watt-meter, and an SDR radio.... even so, we were able to do some comparisons that told us a lot.

    First, we tested both of the antennas mounted on my Jeep where they normally live..... and tuned them to best SWR. Both tuned out at ~1.5:1, and worked like we expected them to: ok but not great. The Base loaded antenna worked a lot better on both transmit and recieve, but neither one was spectacular. antenna 1.jpg antenna 2.jpg
    Next we tested both antennas mounted on the radial plate. Both easily tuned to 1:1 SWR by simply adjusting the antenna.... no tuner required. As predicted by N6LF there was a very profound increase in received signal strength, less noise, and an apparent strong increase in radiated signal.... as evidenced by signal reports from other operators.

    Actually... I'm very impressed by this experiment. N6LF's numbers and our results here seem to indicate that the problem with small vertical antennas is not so much a fault of the antenna itself, but rather with the lack of a proper ground for it to work against. Our results were so good that we are considering the design for our DXpeditions.... but will still need to do some further testing before we decide.

    Of course this is not an Earth-Shattering new discovery. Short verticals with radials have been around for quite some time....BUT given the results we had with them, I think they may have been given the short stick by hams, and hold a lot more promise than most folks realize.

  2. K7TRF

    K7TRF Premium Subscriber QRZ Page

    Nice experiment!

    Lack of good RF Return is definitely an issue with any vertical antenna, short or not. Not surprised that improving the radial system improved performance. The next obvious step would be to repeat the experiment with longer antennas, for instance using the good ground plane system try a 16' base loaded whip. You should see another dramatic improvement as the whip length goes up as the radiation resistance of a shortened vertical antenna increases roughly with the length (relative to 1/4 wavelength) squared.

    IOW, your tests show the value of a good radial system but unless you compare to longer whips it doesn't really say anything about how severely shortened antennas compare to longer yet not quite full length antennas. I wouldn't lay all the problems folks have with shortened antennas on the ground plane, sure that's a big factor but going from 8' to 16' or more for the whip should still show substantial performance improvements if you'll work 20m or lower frequency bands.

    Also from the DXpedition sense, if you can travel with an 8' radiator you can travel with a couple or a few screw together 8' radiators so running something like a 16' whip (basically quarter wavelength on 20m so very little base loading needed) or 24' (great for lower frequency bands) shouldn't be a deal breaker. But sure if your plan is to use a single base loaded antenna for all HF bands up to 10m then 8' is about as long as you can go and you'll trade some performance on 40m and especially 80m to get that 10m coverage.
    KK4NSF likes this.
  3. KK4NSF

    KK4NSF Ham Member QRZ Page

    yep! we are going to order one of MFJ's long telescopic antennas.... and see how much we can gain using it.
    K7TRF likes this.
  4. W9KEY

    W9KEY XML Subscriber QRZ Page

    New ham here, so I've been reading quite a number of antenna books. Our local library has an old (20th Edition / 2003) ARRL Antenna Book.

    I suppose things may have been updated in the 16 years since it was published, but Chapter 3, "The Effects of Ground" is fascinating reading. Ten pages into that chapter is a Table showing "Optimum Ground-System Configurations" for vertical antennas- which highlights effects of 6 different quantities of radials (16, 24, 36, 60, 90, and 120) and their "optimum" length.

    Of particular interest - as you increase the number of radials, you should ALSO increase the length of EACH radial - the reason for that is clearly explained in that section and has to do with soil coverage. So, for example:
    • With 16 radials, each should be 0.100 wavelength (totaling 1.6 wavelengths of wire) ; power loss 3.0 dB.
    • With 24 radials, each should be 0.125 wavelength (totaling 3 wavelengths of wire) ; power loss 2.0 dB.
    • With 36 radials, each should be 0.150 wavelength (totaling 5.4 wavelengths of wire) ; power loss 1.5 dB.
    • With 60 radials, each should be 0.200 wavelength (totaling 12 wavelengths of wire) ; power loss 1.0 dB.
    • With 90 radials, each should be 0.250 wavelength (totaling 22.5 wavelengths of wire) ; power loss 0.5 dB.
    • With 120 radials, each should be 0.400 wavelength (totaling 48 wavelengths of wire) ; power loss 0 dB.

    Obviously, the total length of wire required increases dramatically as more radials are used. For a 40 meter vertical, with 120 radials, you would need 1.2 miles of radial wire! (but 36 radials would only require 709 feet).

    The gauge (diameter) of the radial wire is not important, other than being mechanically large enough to avoid breakage during deployment. They recommend any size from #12 to #22 copper. The table includes other interesting stuff, like feed point impedance.

    The good news - others have already done some of the work - thus saving you a LOT of experimental wire!
    KK4NSF and AI3V like this.
  5. KK4NSF

    KK4NSF Ham Member QRZ Page

    That generally corresponds with N6LFs work, but differs in the number and lengths recommended. A link to his work is here:

    Keep in mind that radial lengths and numbers are highly dependent on soil types, moisture content and so forth. The term "optimum" in the ARRL Antenna books is geared toward most efficient use of the wire...."band for your buck"... in "average soil" which is undefined. With even a small number of radials, like our 16, increased length does offer increased efficiency but the law of diminishing returns starts to kick in. Where this happens varies with soil types.

    Our goal was to see how much a good set of radials would improve a physically small vertical, for use on MiniDXpeditions.... which we determined was quite substantial.

Share This Page