Small Antennas For Small Spaces: Projects and Advice for Limited-Space Stations

Table of contents :
001LR_Page_001......Page 3
001LR_Page_002......Page 4
001LR_Page_003......Page 5
001LR_Page_004......Page 6
001LR_Page_005......Page 7
001LR_Page_006......Page 8
001LR_Page_007......Page 9
001LR_Page_008......Page 10
001LR_Page_009......Page 11
001LR_Page_010......Page 12
001LR_Page_011......Page 13
001LR_Page_012......Page 14
001LR_Page_013......Page 15
001LR_Page_014......Page 16
001LR_Page_015......Page 17
001LR_Page_016......Page 18
001LR_Page_017......Page 19
001LR_Page_018......Page 20
001LR_Page_019......Page 21
001LR_Page_020......Page 22
001LR_Page_021......Page 23
001LR_Page_022......Page 24
001LR_Page_023......Page 25
001LR_Page_024......Page 26
001LR_Page_025......Page 27
001LR_Page_026......Page 28
001LR_Page_027......Page 29
001LR_Page_028......Page 30
001LR_Page_029......Page 31
001LR_Page_030......Page 32
001LR_Page_031......Page 33
001LR_Page_032......Page 34
001LR_Page_033......Page 35
001LR_Page_034......Page 36
001LR_Page_035......Page 37
001LR_Page_036......Page 38
001LR_Page_037......Page 39
001LR_Page_038......Page 40
001LR_Page_039......Page 41
001LR_Page_040......Page 42
001LR_Page_041......Page 43
001LR_Page_042......Page 44
001LR_Page_043......Page 45
001LR_Page_044......Page 46
001LR_Page_045......Page 47
001LR_Page_046......Page 48
001LR_Page_047......Page 49
001LR_Page_048......Page 50
001LR_Page_049......Page 51
001LR_Page_050......Page 52
001LR_Page_051......Page 53
001LR_Page_052......Page 54
001LR_Page_053......Page 55
001LR_Page_054......Page 56
001LR_Page_055......Page 57
001LR_Page_056......Page 58
001LR_Page_057......Page 59
001LR_Page_058......Page 60
001LR_Page_059......Page 61
001LR_Page_060......Page 62
001LR_Page_061......Page 63
001LR_Page_062......Page 64
001LR_Page_063......Page 65
001LR_Page_064......Page 66
001LR_Page_065......Page 67
001LR_Page_066......Page 68
001LR_Page_067......Page 69
001LR_Page_068......Page 70
001LR_Page_069......Page 71
001LR_Page_070......Page 72
001LR_Page_071......Page 73
001LR_Page_072......Page 74
001LR_Page_073......Page 75
001LR_Page_074......Page 76
001LR_Page_075......Page 77
001LR_Page_076......Page 78
001LR_Page_077......Page 79
001LR_Page_078......Page 80
001LR_Page_079......Page 81
001LR_Page_080......Page 82
001LR_Page_081......Page 83
001LR_Page_082......Page 84
001LR_Page_083......Page 85
001LR_Page_084......Page 86
001LR_Page_085......Page 87
001LR_Page_086......Page 88
001LR_Page_087......Page 89
001LR_Page_088......Page 90
001LR_Page_089......Page 91
001LR_Page_090......Page 92
001LR_Page_091......Page 93
001LR_Page_092......Page 94
001LR_Page_093......Page 95
001LR_Page_094......Page 96
001LR_Page_095......Page 97
001LR_Page_096......Page 98
001LR_Page_097......Page 99
001LR_Page_098......Page 100
001LR_Page_099......Page 101
001LR_Page_100......Page 102
001LR_Page_101......Page 103
001LR_Page_102......Page 104
001LR_Page_103......Page 105
001LR_Page_104......Page 106
001LR_Page_105......Page 107
001LR_Page_106......Page 108
001LR_Page_107......Page 109
001LR_Page_108......Page 110
001LR_Page_109......Page 111
001LR_Page_110......Page 112
001LR_Page_111......Page 113
001LR_Page_112......Page 114
001LR_Page_113......Page 115
001LR_Page_114......Page 116
001LR_Page_115......Page 117
001LR_Page_116......Page 118
001LR_Page_117......Page 119
001LR_Page_118......Page 120
001LR_Page_119......Page 121
001LR_Page_120......Page 122
001LR_Page_121......Page 123
001LR_Page_122......Page 124
001LR_Page_123......Page 125
001LR_Page_124......Page 126
001LR_Page_125......Page 127
001LR_Page_126......Page 128
001LR_Page_127......Page 129
001LR_Page_128......Page 130

Citation preview

-----ARRL.~s~----

Projects and Advice for Limited-Space Stations

- -- - -ARRL,s - - - --

Small Antennas tor Small Spaces Projects and Advice for Limited-Space Stations

Steve Ford, WBSIMY

t.
----

21

22

23

24

25

26

27

0.4

fl:

0.2

'

28

29

30

31

32

33

of the anten na (the 11ear:field region), where the losses can be very high.s The puivose of the ground system is to reduce these near-field losses, increasing efficiency and allowing us to radiate as much of the antenna input power as poss ible, which ultimately improves our s ignal.

Overview of the Experiments T his wo rk started with a 160 me te r ve11ical with which T vaiied the number of Yi wave radials and measured the change in signal strength for a fixed input power. This was interesting and educational but J realized that repeatedly laying down and pic king up some 8000 feet of #12 AWG wire was not practical for more extensive investigations. I thus changed the test frequency to 7 .2 MHz initially, and later added experime nts for multiband ground systems (40 through 10 meters). This initial experime nt also stimulated me to use the much more accurate measurement procedure that is outlined in the sidebar on the QST ln Depth Web site.9 Twent tJu-ough several rounds of experiments, each one answering some questions but, of course, always generating more. In the following three sections we'll consider radials for vertical monopoles - on and above the ground and finally, radial systems for multiband verticals.

Round OneRadials on the Ground This sel of experiments used four different m1te nnas: a Y. wave vertical, an Vs wave vertical with base loading, an Vs wave vertical with sufficient top loading to be resonant at 7.2 MHz and a 40 meter mobile whip. 1 stm·ted with a single 4 foot ground stake (zero radials) and then progressively added Y. wave radials, measuring tJ1e changes in signal strength with each increase in radial num-

'

I

' ' . I

I

I

'

,_,_ 1/4-wave vertical' 1-f-

1-1I1- 11-11-11-1-

7.2 MHz

; '

I

I I

I I I

. '

I

0.0

0

5

10

15

-

I

I I I

4 Radials 23 Sept 07

QX0901 ·Sovcms2-02

Figure 2 - Effect on signal strength of shortening radial lengths. The 0 dB reference is four 33 foot radials.

I I

'

Radial Length (feel) QX0901-Sevems2-04

I I

I

,__

0.0 20

'

I

~

.

3.5

Sinewave Trendline

20

-

-_,

"

25

30

35

Distance from base (feet)

Figure 3 - Measured current distribution on a radial.

bcr. The results are shown in Figure I. Note that the graph is in terms of the improvement in s ignal for a given input power for each antenna over tJ1e single ground stake with no radials. The graph does not compare the relative merit of each antenna. Obviously a sho11, lossy mobile whip will yield less s ignal, typically JO dB less, than a full size Y. wave vertical. The signal improvement metric gives us a direct idea of bow much is gained for a given improvement in tJ1e ground system.

How Many Radials? This graph shows several things. First it makes clecu· just how importru1t a rad ia.l system is. IL can make a differe nce of many dB in o ur signal strength. Keep in mind that the soil over which the expe rimen ts were done would be classified as good to ve ry good. Over average or poor soils the s ignaJ improve ments could be many dB greater than shown here. The second thing the graph shows is the point of diminishing ret11111s. Laying down a syste m with at least 16 radials will give you most of the obtainable improvement. As we go to 32 and the n 64 radials the improvement ge ts progressively smaller. It's arguable that the improvement from going from 32 to 64 radials is worth the cost and clearly the standard 120 radial BC ground syste m would be ove rkill. A final point the graph makes is that tile sho rrer and more heavily loaded your venical, rhe 111.ore you have 10 gain from improving the ground system. The sho1ter the vertical, the higher will be the fi eld inte nsity (for a given i nput powe r) in the near field of the antenna and the lower will be the radiation resistance. This leads to much higher ground losses, which translates to mo re improvement whe n you reduce these losses by improving the ground system.

How Long Should They Be? Radials Y. wave in length are known to be effective in ground systems, bul I wondered what the penalty would be from using shmt er radials. I was expecting to see a fairly unifom1 decrease in signal strength (due to an increase in ground loss) as the radials were shortened. That is 110 1 what I found. Figure 2 shows the results of an experiment in which I measured the s ignal strength while progressively shortening the radials in four and eight radial systems. Surprisingly, s horte ning the radial lengths increased the signal strength - not by just a little bit, but by more than 3 dB. This is ce1tainly counterintuitive, but I was seeing clues that helped explain what was happening. 1 noticed that with only the ground stake the resonant frequency of the vertical was much lower than expected and, as I added more radials, tJ1e resonant frequency increased slowly. Most of the change occu1Ted between 4 and 16 radials and had pretty much leveled out by the time I had 64 radials. This suggested to me that the radials might be self-resonant below 7.2 MHz. To check this out 1 measured the cu1Tent distribution on a radial and found it to be sinusoidal. The results are shown in Figure 3. The maximum current point has been moved from the base of the antenna out onto the radials and th_is substantiaUy increases the ground loss. The radials are resonant below the band and this affects the antenna. A wire, close to ground, can be heavily loaded by the ground, decreasing its resonant freque ncy. The extent of the loading will depend on the characteristics of the soil. Figure 3 shows that the maximum current point is 10 to 11 feet away from the base. Looking at Figure 2 we see that the maximum signal occurs when we have shortened the radial by this amount. Figure 3 also illustrates a difference

Appendix • 39

OX0903-Severns3-01

6

co ~ c

5

c

2

'

...... :-Y- 'i

1- r 3

·;;;

;-.

,

ea.

.£

.

~

i-!-"-

4

E >

I-'---,..l- I .J.-1. ·-

,_·- -T-+- ~_-:-:-+

I-

~

Q) Q)

I

1-F-

l

0 0

-~

11

10

: :·:_t

-. '

~·~

' I

30

=

I I I

40

50

c"'

4

E Q) > 0

3

.E

2

c:

·;;;

-

l..L..J.-j ,.l.J~. I~ ....,.-..---. ____.... -••

60

_,_,__ -

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Height of radials above ground (ft)

Figure 5 - Signal improvement with four radials and the antenna base at different heights. F = 7.2 MHz.

Table 1 Relative Signal Strengths for 4, 8, 16 and 32 Radials, Comparing Lengths of 33' and 21' Normalized to Four 33' Radials (dB) 33' Radials

Normalized To Four 33' Radials {dB) 21' Radials

Gain Change {dB)

0 2.26 3.76 4.16

3 .08 3.68 3.95 4.04

+3 .08 +1.42 +0.19 - 0 .12

between buried bare wire rad.ials and radials lying on o r very near the surface of the soil. The current dis tribution on a buried bare radial wi ll usuall y decrease exponentially from th e base regardless of its lenglh . 1o You will not see the standing wave shown in F igure 3 except in very poor soils. The insulated radial lying on the ground surface behaves much more like a radial in an elevated radial system in that it has a sine wavelike current distribution. A buried insulated wire will be somewhere in between these two cases depending on the burial depth an d soil characteristics. You can also see in Figure 2 that the s ignal increases as the radial numbers increase. To check this out J extended the experiment to 32 radia ls, comparing 33 to 21 fool radials. The res ults are given in Table I. The results in Table l indicate that the excess loss due to radial resonance has pretty much disappeared by the time you reach 16 radials. This leads 10 some advice - rather than trying to dete1mine the optimum radial length, which will vary with every installation due to soil differences, just use at least 16 radials. If you are l imited by the meal amount of wire available, you're better off to use a larger number of shorter radials rather than a few long ones. I didn' t have time to rnn an extensive set

Appendix

-~

- I-

70

=

40

t

I

Figure 4 - Signal improvement as a function of radial number. All radials lying on the ground surface, F 7.2 MHz.

4 8 16 32

'I"'·:~-+· l

•·i

.. ,-r.

Radial Number

Number of Radials

.. -+-'-'-~~~~~1~

' - -- -•---' -.f...-t-

::::!.

a.

-

Radial Length 33' h 33.5' No Ground Stakes. Choke Isolated

-~

20

=

co

5

Q)

I-+ ~1::;-~k,_!__. L ,__::J_

'

-

---, ~

1~

.

·-t-

.~

~.

1-

~_;.t

-~t~

~.

QX0903-Sevems3-02

6 ~'-'-'~-'--

'

of experiments comparing different radi al le ng th and radial number combinatio ns (each with the same total length of wire). but I did model that situation with EZNEC. 11 The modeli ng predicted, particularly with short verticals, that it was often advantageous to reduce the length of lhc radia ls and increase their number. The modeling showed that there is a coJTelation between vertical height and optimum radial lengths. More details can be found in the modeling repo11 and in the workofothers. 12-t 5

Round Two - Elevated Radials Over the past few years there has been a lot o f discussion about the re lative merits of ground systems using a large number o f surface or buried radials versus onJy a few elevated radials. This stems from NEC modeling tlrnt indicated that four radials elevated 8 foet or so above ground could be just as e ffective as 120 buried radials . Many of us, including me, simply could not believe that. 1 dec ided the best way to address this ques ti o n would be to directl y compare two ante nnas, one w ith a large number of ground radials and the otl1e r with o nl y a few elevated radials. The same antenna was used in both cases, a simple Y. wave vertical. For the surface tests I used Y. wave radials and vari ed the number from 4 to 64. For the

devated tests I used four Y. wave radials. T he elevated radials were placed at 0, 6, 12 and 48 inches above ground. The res u lts are shown in Figures 4 and 5. T he 0 dB po int in th e graphs is normalized to the s ig nal strength for the case of four y, wave radials lyi ng on the st11face (0 dB ). What you see in the graphs is the improvement as you eithe r add more su1face radials or elevate the antenna and the four radials above ground. T he most s uiking thing shown by the graphs is that four elevated radials at a height of 48 inches are withi n 0.2 dB of 64 radials ly ing on the ground. This wou ld seem to suppon Lhe predictions from NEC modeling. A detailed view of the results with different e levated configurations is provided on the QST ln Depth We b site.

Round Three Multiband Ground Systems Whi le s ing le band verticals a rc frequently used, multiband ve rticals are even more popular but I'd not seen any experime ntal work re lated to multiband ground systems. So I did some. The experiments were pe1formed in n vo phases. T he first was for rad ials lyi ng on the ground and the seco nd was for e levated radials. These represent two typical scenarios for amateurs, he lping lo answer a related question: " Do l put the an te nna in the backyard or up on the roof?" For this series of tests I used a SteppJR IJl vcrti cal. 10 The motor driven SteppJR can be adjusted to be resonant anywhere be tween 40 and 6 mclcrs. For these expe riments I made up four sets of thirty-two Y.l wave radials, one set for each band (40, 20. 15 and 10 meters). I then tried several different configurations starting w ith sets of 32 s ingle balld radials, one se1a1 c1 lime. In this way I had a Yi wave vertical over a ground system of thir ty-two y, wave rad ia ls on each band. These antennas were

then measured individually on each band. I then tried groups of four and eight (32 total) ~ radials for each band, connected all at the same time. Next I uied 32 radials each 32 feet Jong, followed by 16, 8