qex-ground-systems-part-5.pdf

Rudy Severns, N6LF

PO Box 589, Cottage Grove, OR 97424; n6lf@arrl.net

Experimental Determination of

Ground System Performance for

HF Verticals

Part 5

160 Meter Vertical Ground System

How much will the signal strength and feed point

impedance change as radials are added?

This experiment was actually the irst

of the series of experiments on ground

systems that have been the subject of this

series of articles. The experiment involved

measuring the change in signal strength as

radials are added to the ground system of a

vertical antenna, beginning with four radials

and going up to 64 radials. The intent was

to determine the additional gain in signal

for each doubling of radial number, and to

determine the point of vanishing returns. In

addition, the changes in feed point imped-

ance due to changing radial number were

of interest.

While the results of this initial experiment

were quite interesting, a more important

result was an appreciation of the dificulties

of making these measurements accurately.

This experience led to a modiication in the

test procedure and a shift to 40 m verticals,

which have been described earlier.

Figure 1 — This photo shows the antenna base with radials attached.

Test Antenna Description

The test frequency for this experiment

was 1.800 to 2.000 MHz. The vertical was

125 feet of no. 12 AWG insulated copper

wire suspended from a Dacron line hung

between two 150 foot poles.

At the base of the antenna there was an

18 inch diameter copper disk, as shown in

Figure 1. The inner ends of the radials and

The terrain around the antenna was not

lat, but rather on a narrow ridge about 40

to 50 feet wide. The result is that many of

the radials were in part bent down at about

a 45° angle as they ran down the steep slope

on either side. Along the ridge, however, the

radials are more or less level.

the shield of the coax feed line were attached

to the disk. There were also two galvanized

5 ⁄ 8 inch × 4 foot ground stakes connected to

the disk. The radials were 130 foot lengths of

no. 12 insulated (THHW) wire lying on the

ground surface. Radials were put down in

the sequence of 4, 8, 16, 32 and 64.

QEX – July/August 2009 15

The test antenna was erected 700 feet

to the east of my house with a 50 foot deep

gully in between. The ridge is in a Douglas ir

forest with 100 plus foot trees within 50 feet

of the test antenna at some points. The radial

system ran along the ridge and also down the

sides of the ridge into the forest.

To excite the test antenna, between the

house and the antenna there was a 700 foot

length of 1 5 ⁄ 8 inch coax, with an additional

75 feet of ½ inch coax. Both were Andrews

heliax.

Table 1

Typical Test Data for Received Signal Strength with P o = 50 W.

Number of Radials

Corrected Signal Strength

Relative Signal Strength

–30.1 dBm

0.0 dBm

–29.3 dBm

0.8 dBm

–28.9 dBm

1.2 dBm

–28.0 dBm

2.1 dBm

–27.7 dBm

2.4 dBm

2.5

Measurement Equipment

The signal source was a Yaesu FT1000MP

transceiver with two Bird Model 43 wattme-

ters on the output (forward and reflected

power). The wattmeters were used to set the

forward power to a constant 50 W and also to

measure relected power to calculate SWR.

The SWR measurement is needed to correct

for the power relected from the antenna and

not radiated. This correction was applied to

the received signal amplitude.

The receiving antenna was a 10 foot

vertical wire driven against a 4 foot ground

stake, next to my house. The receiver was an

HP3585A spectrum analyzer. The amplitude

resolution was about ± 0.1 dB.

Base impedance measurements were

made at the antenna using an N2PK vector

network analyzer (VNA). The impedance

measurements were accurate to better than

1%.

The test procedure was very straight-

forward. For each number of radials, the

FT1000MP output was adjusted to 50 W and

received signal strength on the spectrum ana-

lyzer recorded along with the SWR for that

measurement and the input impedance at the

base of the antenna.

2.0

1.5

1.0

0.5

160 m Te st Vertical

August 2006

Run 2

0.0

Number of Radials

QX0907-Severns2

Figure 2 — Here is a graph of the typical signal strength change with radial number.

4 Radials

160 m Test Vertical

August 2006

Run 2

8 Radials

Test Results

Three complete runs were made to verify

repeatability of the measurements. Each run

included a complete stepping through the

number of radials in the sequence, 4, 8, 16,

32 and 64. Typical received (and corrected

for SWR) signal strengths versus radial num-

ber are given in Table 1. This data is graphed

in Figure 2.

The data in Figure 2 has one obvious odd-

ity. You would expect that the incremental

difference as the radial numbers are doubled

would be monotonically decreasing as the

radial number rises. The step between 16

and 32 radials does not do this and it appears

that the value for 16 radials is too small. This

anomaly was noted during the experiment,

however, and checked carefully as the radial

count was redone three times. The anomaly

was there in all three cases. I have no expla-

nation for this other than the irregularity of

16 Radials

32 Radials

64 Radials

1.80

1.82

1.84

1.86

1.88

1.90

1.92

1.94

1.96

1.98

2.00

Frequency (MHz)

QX0907-Severns3

Figure 3— This graph gives the resistive part of the base impedance over the 160 m band for

different radial numbers.

16 QEX – July/August 2009

the site, which forced the radial layout to

be far from lat or level. Later experiments

with more regular radial systems on other

antennas all showed the expected monotonic

decrease in improvement with increasing

radial number.

In any case, it’s pretty clear that 32 radi-

als do a good job and by 64 radials you are

well into the region of vanishing returns. I

certainly could not justify doubling the radial

count to 128!

The results of feed-point impedance mea-

surements are given in Figures 3, 4 and 5.

As discussed in Part 2 of this series, we

would expect the resonant frequency to vary

with the number of radials, due to the shift in

radial resonance because of soil loading. The

40 m experimental work was done over an

essentially lat pasture and the resonant fre-

quency change was regular and monotonic.

The gross irregularity of the ground surface

in this earlier experiment, however, resulted

in the erratic frequency changes shown in

Figure 5. This problem was a primary reason

for moving the experimental site from the

narrow ridge to a pasture. Unfortunately, the

150 foot support poles were not available in

the pasture so it was necessary to change the

experimental frequency to 40 m to make the

vertical height manageable.

160 m Te st Vertical

August 2006

Run 2

1.9 MHz

Number of Radials

QX0907-Severns4

Figure 4 —This graph shows the base resistive component versus radial number at 1.9 MHz.

1.920

160 m Te st Vertical

August 2006

Run 2

1.915

Summary

This initial experiment helped me to

understand the problems inherent in mak-

ing accurate comparisons between different

ground systems. I had to change the site,

the test frequency, the test instrumentation

and the test methodology to get to the point

where I could have conidence in the test

results and draw conclusions from them.

This experiment was by no means a fail-

ure, however. We can see that the change

in signal strength is very much in line with

what we saw in the 40 m work. It also sup-

ports the conclusion that we should use at

least 16 radials, but when we use more than

32 radials we are definitely reaching the

point of vanishing returns. For most amateur

installations the Standard Broadcast ground

system of one hundred twenty 0.4-wave-

length radials could not be justiied by any

useful increase in signal strength.

1.910

1.905

1.900

1.895

Number of Radials

QX0907-Severns5

Figure 5 — This graph shows the antenna resonant frequency for different numbers of

radials.

Rudy Severns, N6LF, was irst licensed as

WN7WAG in 1954 and has held an Extra class

license since 1959. He is a consultant in the

design of power electronics, magnetic compo-

nents and power-conversion equipment. Rudy

holds a BSE degree from the University of

California at Los Angeles. He is the author of

two books and over 80 technical papers. Rudy

is an ARRL Member, and also an IEEE Fellow.

QEX – July/August 2009 17

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: