[Top] [All Lists]

[TowerTalk] Efficiency of Vertical Radial Fields

To: <>
Subject: [TowerTalk] Efficiency of Vertical Radial Fields
From: (Jim Reid)
Date: Sun, 21 Jun 1998 11:08:44 -0900


  Recall my intentions to discover the impact of added
elevated radials,  beyond four,  to an elevated vertical antenna.

My plan was to use the NCDXF beacons,  attenuators,  and my
horizontal beam/long wires to discover any signal efficiency
improvement as radial wire count were to be doubled,  three

A couple days ago,  Eric,  N7CL,  sent to me an email which I found
very eye "opening" with respect to my intentions.  With his permission,
I want to share some of his thoughts with you,  and he granted this
earlier this morning.  So below is the meat of  his message:

Reid wrote:
>>2.  The background RF noise level out here is VERY low.  However,
>>It is a bit surprising to me,  and a few others,  that there would
>>be 40 or 50 dB of "dynamic" or FT-1000D sensitivity range on 
>>down below the "no S meter movement" of the rig!!  None of
>>yesterday's "loudest" beacon signals ever even "tweaked"
>>the D's S meter.


>Two things.  First, this is why you don't want to be trying a
>comparison like this using signals propagated by the ionosphere.
>You will not get valid comparison data using propagated signals.
>Not even if both the test and reference antennas are the same
>polarization.  They can't be in exactly the same place so
>multipath effects will be causing a bigger (and varying)
>difference than the effect you are trying to measure.
>Second, when you are using a locally generated test signal, use
>enough signal so that the s-meter reads at least s-5 for your
>weak reference signal level (the one you will be attenuating the
>other signal to match).  then you can use your eye and an
>indicator to resolve differences that you couldn't possibly sort
>out with your ears.

[Since your objective is to measure]

> the relative effeciency for various elevated
>radial configurations under a vertical,  the
>reference antenna (and the local emitter) should also be
>vertically polarized.  Otherwise the measurements will be
>interesting but meaningless.  The local emitter need only be a
>wavelength or two away from the antenna under test.
>Also, if possible, try to do your tests when the ionosphere isn't
>working well for the band.  This is less critical for vertical
>tests since the verticals don't illuminate the high angles much
>anyhow.  But it will minimize interference from distant signals.
>The quantity you are trying to measure is unaffected by arrival
>or takeoff angle.  So there is no reason to worry about testing
>against signals arriving from various elevation angles.  The
>ground screen you are installing is not big enough to affect the
>takeoff angle of the test antenna either.  So the change in
>efficeincy will not be masked by pattern changes as you add to
>the screen.

Again,  Reid:

>>6.  Finally,  a couple mentioned that the most gain increase
>>I just might experience with adding all 32 additional radials
>>would probably be about 5 dB at most.  How would I be
>>able to tell the difference with my present method?

>Five dB is indeed about the most you can expect to get.  But it
>should be easily measurable if you get set up right to do the
>measurement.  If you do get 3 dB or more, you have significantly
>reduced the near field losses for that antenna.  I would consider
>it worth doing for 3 dB.

>>Guess I probably could not,  except subjectively.  I really
>>do need to find a calibrated attenuator somewhere for
>>it to be worth while to continue this exercise.

>Yes, you really can't do the measurement without a reasonably
>good 1 dB step attenuator.
>Good luck.  And have fun (most important).
>73, Eric  N7CL

I found Eric's input to be almost mind boggling,  hi!  

Someone email'd that MFJ sells a switchable step
attenuator,  for around $65 or so;  however,  another said
such units could not really be depended upon for accuracy
at the dB levels I will be attempting to discriminate.  So,  am
glad I have the SMA "stack able"  fixed attenuators on order.

Now I must modify my plan,  and most importantly determine
how to set up a nearby vertical radiator to implement Eric's
method.  I suspect that my "older" MFJ - 259 will radiate
sufficient signal to show an S5 indication on the FT-1000D
meter,  if the test transmitter set-up is only two or three
wavelengths away at 20 meters.  I can just,  perhaps hang
a wire from a palm tree,  running the output of the 259
into the wire at the bottom--ought to be good enough
for such  test.  Will be sure to have fresh batteries in the
unit for each test.  I suppose I can use the horizontal
beam as sort of a reference measure of starting out
with the same signal amplitude from the test xmitr
set up.

Any comments on this approach to Eric's test method to
determine the added efficiency of adding radials?

73,  Jim,  KH7M

FAQ on WWW:     
Administrative requests:

<Prev in Thread] Current Thread [Next in Thread>
  • [TowerTalk] Efficiency of Vertical Radial Fields, Jim Reid <=