antenna_ground_system_experiment_6.pdf

Vertical antenna ground system experiment No. 6

Rudy Severns N6LF

9 August 2008

Midway through the first five vertical ground system experiments Paul Thompson, W8IEB, brought

me his BigIR SteppIR vertical to run some tests. Some results using this vertical have been given in

prior experimental reports but in looking over my notebook on the experiments I could see that there

were a number of notes not covered in earlier reports which may be of interest to some users.

The following is not really an "experimental report" but rather just a collection of notes on what I

noticed while running other experiments. Paul has long since taken his antenna back home but if I

have the opportunity to test another SteppIR vertical in the future I will extend this report into

something more complete. Paul worked with me in taking many of the measurements and the hours

consumed in changing radial combinations.

Some mechanical comments

The electrical connection for the radial system is a simple 10-32 machine screw sticking out of the

bottom of the reel housing. During testing, which ended up using a fairly large number of radials, I

found this small screw to be inadequate. I have suggested to SteppIR that this be made larger (at

least a 1/4-20) and longer. Inside the reel housing the connection from the shell of the UHF

connector to the screw is just a wire. What I think needs to be done is to add a small piece of tinned

copper, perhaps 1" x 4" or so, with the UHF feedpoint connector mounted on it. I also suggested two

1/4-20 bolts, through the copper strip, sticking out of the case to which the radials can be attached.

During the experiments we changed radial lengths and numbers frequently and found the single

machine screw to be a real pain. So I made up a circular aluminum plate shown in figure 1. The

plate is 1/8" aluminum which I salvaged so there are a number of spurious holes not related to it's

SteppIR function. The plate is about 15" in diameter, the diameter limited by what available at the

time. The center hole is 2" and passes over the center support post on the housing. The large hole

next to it is a clearance hole for the UHF connecter. Because of the flange on the housing, the plate

mounts about 1/2" below the housing. This recesses the UHF connector so that a fairly large hole is

needed to get your fingers in there to insert the coax connecter. The hole in the picture is about 1.5"

but 2" would have been better.

Outside of the two large holes but on a line with them, there is a small hole for the 10-32 machine

screw which protrudes from the bottom of the housing. The holes are identified by the notes and

arrows on the photo.

Figure 1, ground radial anchor plate for the SteppIR vertical.

Because of the 0.5" flange on the reel housing I had to place a 0.5" spacer around the 10-32 screw.

I used a 0.5" piece of 1" copper pipe for that job. Almost anything would do.

Around the perimeter of the plate are 15 1/4-20 SS bolts to attach radials. Because we were making

frequent changes in radials wing-nuts were used on these bolts. Notice that there are 15 not 16

bolts which may seem odd. The reason for that is the limited diameter of the plate due what I had on

hand. It was not large enough to reach the outside one edge of the housing. The flange blocks the

use of that hole. For most users who are not running an experiment eight or even four bolts would

be plenty and you could use a smaller diameter plate also..

Later in this note I recommend using thirty or more radials (which is in line with SteppIR's

recommendations) and explain why. For portable operation such as field day or DXpeditions this

many radials could be a real nuisance. If you use the radial plate with eight bolts then the radials

can be made up in eight groups of four each with a common terminal or even four groups with eight

radials on a common ring terminal. When you use this many radials they do not have to be very

large wire, #18 stranded hook-up wire is fine. These could be coiled up in groups and placed in a

container for transport. This would be much more practical for deployment in the field. In my

antenna experiments I used #18 stranded wire and with a bit of practice it was surprising how quickly

radials could be laid down or gathered up.

Figure 2 shows the radial plate attached to the SteppIR.

Figure 2, radial base plate installed on the SteppIR.

If SteppIR modifies the housing to have two 1/4-20 bolts protruding then the radial plate should be

solid. Otherwise you have to be careful not to shear off the 10-32 screw if somebody trips over the

radials.

Some electrical tests

One of the objects of testing the SteppIR was to develop a radial system for field day which would

work well on all bands from 40 through 10m. SteppIR recommends using "20 to 30 radials 16' or

more long". I thought this was little odd since on 40m a 1/4-wave is about 33' and on 10m about 8'.

I though that a broadband radial system should have a few radials for each band, each about 1/4-

wave on that band. As a first pass we decided to try four radials on each band cut to 1/4-wave. Boy

was that a mistake!

There are two serious problems with that approach. First, as illustrated by the work reported in

experiments 3, 4 and 5, four physically 1/4-wave (free space 1/4-wave) radials lying close to ground

can be very lossy. As shown in the earlier reports, when using only a few radials (8 or less) it is

necessary to shorten the radials so that they are close to resonance over the particular soil at the

site.

The second problem observed was the coupling between multiple different length radials lead to

unexpected resonances and the resonant length for the vertical was often very different from the

preset lengths from SteppIR. We also noticed that some of the resonances led to very large drops in

signal strength.

Seeing this, the next step was to determine the optimum radial length on each band, 40, 30, 20, etc,

etc, by observing the signal as the radials were trimmed down from the physical 1/4-wave. For this

experiment we used 8 radials on each band. In all cases we were able to optimize the radials and

the resulting vertical heights for resonance were close to those provided by SteppIR. But when we

put all these radials on the base plate simultaneously we had a real rats nest, totally impractical for

portable use. More importantly the strange resonances were back. In short, this idea didn't work

either!

What works is to do what SteppIR suggests: use a bunch of radials! When all else fails, read the

instructions! We found that 30 or more radials eliminated all the resonance and coupling effects and

was provided an efficient ground system. Radial lengths of 20' to 25' seemed to work noticeably

better than 16' radials and that was over good soil. Poor soils would probably benefit from even

longer radials if you want to operate down to 40m which is of course the bread-and-butter band for

DXpeditions while the sunspot cycle remains low.

These remarks are just observations. No doubt a more extensive and careful set of experiments

would have been better but we really didn't have the time for that. But even still I think these

observations deserve consideration until we have more to work with.

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: