Topband
[Top] [All Lists]

TopBand: How many radials ?

To: <topband@contesting.com>
Subject: TopBand: How many radials ?
From: n7cl@sparx.mmsi.com (Eric Gustafson)
Date: Sun, 20 Oct 1996 15:45:33 -0700
 >From: dj8wl@uugate.wa7slg.ampr.org
To: <topband@contesting.com>
 >Date: Sun, 20 Oct 96 12:42:25 mst
 >
 >
 >How many radials?
 >
 >Quite a few years ago I tried to find out how many radials I
 >would have to *plant* to get a decent ground system for a new
 >T-antenna to be put into service on 160m....
 >

Snip...


 >The feedpoint impedance of the T-antenna was measured with a rf
 >impedance bridge. I began laying 50m long radials on the
 >ground connecting more and more of them to the common ground
 >connection at the rf bridge.
 >
 >The result can be seen in the following table:
 >
 >no. of radials:       2     5     9     14     20     30     50
 >feedpoint (ohms)     122    66    48    39     35     32     29
 >

Snip...


 >I stopped after having added 50 radials and looking at the
 >diagram I decided that I would have to add at least 50 more
 >radials to see a major improvement.
 >

This is another very interesting empirical determination.
Peter's perception of diminishing returns based on feedpoint
impedance measurements came at a radial tip to tip spacing of
0.039 wavelengths.  This is very close to the 0.03 wavelength
value I referred to in an earlier posting.  Given that he was
using a physically short radiator and that the measurement
methods were completely different, these values would appear to
be in extremely good agreement.  Only 15 more radials would have
put Peter's screen at the 0.03 wavelength spacing assuming equal
distance between radials.

Good job Peter!  Your method is also probably the only one that
an average Ham could consider using to make the determination for
himself under the constraints of his particular installation.


 >A few years later I learned that I was within 2 db of the
 >optimum.

Just out of curiosity, what method was used to make that
determination (This is not a sharpshooting attempt)?


 >You do not have to bury radials, just lay them on the ground !
 >

In fact, it may be better _not_ to bury them if it isn't
necessary.  But the difference is probably not great if the
burial is shallow.

 >Peter, DJ8WL  <dj8wl @ uugate.wa7slg.ampr.org>
 >


If I were putting up a ground mounted vertical, I think I would
(will) do the following:

1.  Determine how long the radials _can_ be under the constraints
    of the antenna site.  I would not consider radials longer
    than 0.375 wavelength unless I was _very_ rich or extremely
    obsessive.

2.  Calculate how many radials are necessary to get to 0.03
    wavelength tip to tip distance.

3.  Put in that many radials.

4.  Stop worrying if the system is optimized and start working
    DX.

For example, using Peter's 50 meter long radials on top band we would
get the following calculation.


    50/160 = 0.31 (Radial length in terms of wavelength)

    ARCTAN(0.03/0.3125) = 5.48 (number of degrees between radials
                                to get to 0.03 wavelength spacing)

    360/5.48 = 65.65 (number of radials required to get almost
                      all of the possible ground loss improvement
                      when constrained to radials of this length)

Sixty five looks like a good number to me in this case.  As we
have seen, fifty also works.

73,  Eric

-- 
------------------------------------------------------------------------------
Eric Gustafson  N7CL                  | The mountains are high and the Emperor
6730 S. Old Spanish Trail             | is far away.
Tucson, AZ 85747                      |
                                      | You can't work 'em
INTERNET: n7cl@mmsi.com               | if you can't hear 'em.
------------------------------------------------------------------------------

--
FAQ on WWW:               http://www.contesting.com/topband.html
Submissions:              topband@contesting.com
Administrative requests:  topband-REQUEST@contesting.com
Sponsored by Akorn Access, Inc & KM9P

<Prev in Thread] Current Thread [Next in Thread>