A Simple Dummy Antenna
Grant Bingeman, KM5KG
Have you ever needed a quick and dirty dummy antenna to make some transmitter tests? What can you do during Field Day or in the middle of that holiday trip
when you need a dummy load, but you left it at home? This may sound like an unusual solution, but it works -- just bury a short dipole a few inches below
ground and transmit into that.
I say "short" dipole because the permittivity of earth is quite a bit higher than that of air, so a wavelength is much shorter along a buried wire than it is along a
wire suspended in air. A wavelength in air is about 300/f meters, where f is measured in MHz. A wavelength in dirt or any medium other than free space is
300/Nf, where N is a factor based on the dielectric constant (relative permittivity) and conductivity of that medium (Equation 1). Thus a wire looks longer in any
medium that has a greater permittivity or greater conductivity than air.
N = { er2 + 3.23(108) (S/f)2 } 1/4 .....................................Equation 1
where er = the dielectric constant, usually about 13 for dirt
and S = the conductivity in Siemens/m, typically .001 in rock and .03 in "good" soil
Consider the case of average soil having a conductivity of 5 mS/m and a relative permittivity of 13 at 14.2 MHz. Then N becomes the fourth root of 169 +
3.23(108) (.005/14.2)2 which works out to be 3.8. This means that what we call the 20 meter band in air looks more like the 5 meter band in average dirt. That is,
a half wave dipole at 14.2 MHz is about 10 meters long in air, but only about 2.6 meters long when buried in the earth. The earth surrounding the wire dipole
forces us to reduce its length by a factor of 3.8 in this case, in order to achieve antenna characteristics we normally attribute to a half-wave dipole.
We must keep in mind that the permittivity and conductivity of the earth may be frequency sensitive. And even if these parameters were constant, the frequency
term in Equation 1 results in a value for N which ranges from 9.5 at 1 MHz to 3.6 at 50 MHz, when er is everywhere 13 and S is .005. At lower frequencies the
conductivity term predominates, but at the higher frequencies the permittivity term has the greatest effect. This inverse frequency effect in Equation 1 is why the
input resistance to a buried dipole can be held fairly constant over a wide frequency range -- the electrical length of a buried dipole tends to change more slowly
than that of a dipole in free space.
I decided to measure the input impedance to a short dipole as I moved it closer and closer to ground, eventually burying it one inch below the surface. My local
earth had a conductivity of about 30 mS/m and a dielectric constant of about 15. Thus N was about 6.4 at 14.2 MHz, so I decided to use a six foot length of bare
12 gauge copper wire for each arm of my dipole. I thought about using a balun to isolate my impedance measuring equipment (an HP vector impedance meter)
from the antenna, then decided it was better to forego the balun since the test equipment would then act as a model of the real stray reactances and coaxial cable
outer conductor radiation of an actual Field Day transmitter. That is, I figured in an emergency where you needed a quick and dirty dummy load, you might not
have a balun handy either. But I made some balanced measurements as well, just for the purpose of comparison, because I did notice that the impedance readings
changed slightly when I touched the case of the vector impedance meter.
So with my vector impedance meter plugged into an extension cord and sitting on a wooden table three feet above ground, with its five foot long probe cable
hanging in the air, I measured my dipole's 15 MHz input impedance at several heights close to ground. As you can see in Table 1 the dipole passes through
resonance as it moves below ground. Also since the system losses increase the closer we get to ground, the input resistance also increases. The probe was
actually lying on the ground for the buried dipole case. The vector impedance meter probe is a metal cylinder about six inches long.
Table 1: 12 foot Dipole Close to Ground, 15 MHz
Height Above Gnd Unbalanced Input Impedance Balanced Input Impedance
18 inches .......... 86 - j614 ohms ............. 31 - j414 ohms
12 ................. 85 - j535 .................. 29 - j394
6 .................. 84 - j533 .................. 35 - j398
3 .................. 87 - j492 .................. 40 - j355
0 .................. 92 - j77 ................... 98 - j219
-1 ................ 135 + j39 .................. 156 + j84
I also measured the dipole's balanced and unbalanced input
impedances over a wide range of frequencies, for two
conditions: the dipole lying directly on top of the ground
(Figure 1), and buried one inch below the surface (Figure 2).
The earth was moist and the conductivity figure of 30 mS/m
is the verified local average value in the standard broadcast
(medium wave) band. Whether or not that value remains the
same at the higher frequencies allocated to amateur radio is a
matter of speculation, and will remain so until somebody
makes a full set of ground-wave field intensity measurements
at HF for a carefully installed monopole using professional
test equipment. Refer to FCC Rules and Regulations 73.184
and 73.186 for procedure details.
Figure 1 is a plot of the unbalanced input R and X when the
12 foot dipole is lying on top of the green grass. Note in
Figure 2 that the impedance is quite a bit lower when the
same dipole is buried one inch below the surface (this change
is most noticeable when the earth is moist). I don't
recommend transmitting into a dipole which is simply lying on
the ground, as there may be safety issues, hot spots, etc. And
the buried situation actually presents a reasonable VSWR
relative to 50 ohms over a surprising portion of the amateur
radio bands. Of course if you had an antenna tuner, you could easily match the buried dipole's impedance to exactly 50 ohms resonant. Note that the input
reactance of the buried dipole is everywhere positive with this particular soil. That is, there appear to be no obvious resonances because the load is so lossy. Of
course, this is what we want in a dummy load. The higher resistance and reactance at the higher frequencies may be caused (at least in part) by RF currents on
the co-ax outer conductor and test equipment. However, comparing the unbalanced results of Figure 2 to the balanced case of Figure 4, perhaps the stray RF
currents in our set-up are not too significant. What we may be seeing is an actual change in the characteristics of the earth at the higher frequencies. The
dissipation factor of various materials is certainly frequency sensitive, as borne out by typical properties of materials tables, such as those in the RF electrical
engineer's bible, Reference Data for Radio Engineers. We will take a quick look at NEC4.1 modeling of buried wires later in this article, and see corroboration of
the measured increase in resistance at 14 MHz compared to 7 Mhz.
When you are building your buried dipole dummy load, if you
notice that there is not much difference between the
impedance you measure when the dipole is lying on the
ground, and when it is buried, then your dirt is probably too
dry. Try adding water, and this should bring the impedance
down considerably, but avoid wetting the center of the dipole
where your co-ax is connected. Comparing Figures 1 and 2,
you can see that the buried dipole impedance is shifted to the
left -- note the parallel-resonant peak in resistance near 40
MHz in Figure 1 slides down to about 14 MHz in Figure 2.
This argues for a value of N of about 3 for the wet condition.
You can use insulated wire if you wish (Figure 3), but this is
probably not a good idea at high power levels, since the
insulation may melt or burn. A thin layer of insulation with a
dielectric constant of three is typical, and tends to increase the
input impedance to the buried dipole as well as restore the
resonances that disappeared in the somewhat lossier bare wire
case depicted in Figure 2. Note in Figure 3 that the resonance
appears near 22 MHz, arguing for an effective value of N of
about 2. At any rate, if you are going to use insulated wire,
keep in mind that the measured values in Figure 3 are for
common PVC insulation, which is much lossier at higher
frequencies than Teflon, for example.
Note that if you pour salt water into your ditch, you can
drastically increase both the permittivity and conductivity.
"Pure" salt water has a dielectric constant near 80 and a
conductivity of 3000 to 5000 mS/m. Thus you may be able to
use a really short dipole, since would approach a value
perhaps as high as 70 in salty mud. So if your "groundtenna"
looks electrically short, add a dash of salt then water liberally.
However, if your ground is simply dry you don't need the salt
-- plain water poured into the trench will really help to pull
down the resistance and reactance closer to 50 ohms
resonant.
I looked at the radiation from this buried dipole using
NEC4.1, and found some interesting field intensities.
Obviously the connection to the buried dipole will allow some
radiation from the exposed above-ground RF input current.
The 1000 watt near-field intensities three feet directly above
the center of the buried dipole are about 14 V/m peak and
100 mA/m peak, when we allow an inch of dipole to protrude
above ground. This is a pretty strong field. Using an MFJ
9420 travel radio, I put about ten watts at 14.2 MHz into my
buried dipole, but would recommend that when using this
kind of dummy load at 1000 watts you do so only for short
periods until you are comfortable with its stability and performance, radio-frequency interference, etc.. If you notice any RF on the transmitter case, then I would
advise you to insert a balun at the dipole input, or make a choke with several turns of your co-ax.
Figure 4 shows the measured input impedances with a balun
connected to the buried 12 foot dipole. The performance of
the balun is suspect above 30 MHz, so you may want to take
the higher-frequency data with a grain of salt. Therefore we
should not compare the parallel-resonance point with that of
the unbalanced data in the previous figures. But in general the
performance of the balanced and unbalanced buried dipoles is
similar enough to provide reasonable dummy load function
over much of the amateur radio bands.
So it appears that a quick and dirty dummy load can be
fashioned from a bare copper wire dipole buried an inch
below ground in moist soil. This dipole can be quite a bit
shorter than a conventional half-wave resonant dipole because
a wavelength in the earth is a lot shorter than it is in the air. A
12 foot buried dipole seems to provide a reasonable
impedance match in moist earth for the 160 to 40 meter
bands, but a shorter dipole may not necessarily work better
for higher frequencies. Refer to Figure 5 where the reactance
of an 8 foot buried dipole above 10 MHz is considerably
higher than that of the 12 foot dipole, and the resistance is
about the same. This shorter dipole is indeed resonant at a
higher frequency, about 58 MHz compared to the 40 MHz of the 12 foot dipole. This makes sense, since we would expect the ratio of the resonant frequencies
to scale according to the ratio of the dipole lengths. I invite the reader to make his own measurements to see if a reasonable buried antenna dummy load can be
made for the higher-frequency ham bands that does not require a balun or impedance matching network.
In case you are wondering exactly how permittivity, relative
permittivity and dielectric constant are related, it all ties back
to the permittivity of free space, e0 which is 8.85 pF/m. If the
permittivity of the earth is 115, then the relative permittivity is
13 = 115/8.85. This figure is also called the dielectric constant
(Equation 2). Permittivity describes the interaction of a
medium with an electric (as opposed to magnetic) field.
Permeability describes the interaction of a medium with a
magnetic field. The permeability of free space, µ0 is 1.26
uH/m. The intrinsic impedance of free space is the familiar
377 ohms, per Equation 3, which equation can also be applied
to transmission lines, in which case we call it the
characteristic or surge impedance. You can use a form of
Ohm's Law to determine magnetic field, H in amps/meter, if
you know the electric field, E in volts/meter, using Equation 4
below.
er = e / e0 .................Equation 2
Z0 = sqrt { µ0 /e0 } ..............Equation 3
H = E / Z0 ...............Equation 4
Note that these equations apply to a far-field situation, however, and if you want to know what the electric and magnetic near fields are likely to be where you
are standing directly over a buried wire, I recommend you either measure them or at least model them using NEC3 or NEC4, which allow buried wires in their
models.
Using NEC4.1 to model a dipole buried three inches in 30 mS/m soil, using one inch model segments and #12 AWG copper wire, with a feed-point connection
one inch above ground, I developed the following 7.1 and 14.25 MHz input impedance data (Table 2). As you can see, the 72 inch (12 foot dipole) values are in
reasonable agreement with the measured values. We can also see that there is a practical limit to the dipole length where additional wire does not have any
significant effect. This tends to indicate that most of the power has been absorbed by the earth in just the first few feet of wire, at least at the higher frequencies.
Thus there would be little point in using more than about three feet of wire for each arm of the buried dipole at 7 and 14 MHz and this soil conductivity. The
shorter length also appears to offer a lower reactance, hence a better impedance match to 50 ohm coaxial cable.
Table 2: 12 Gauge Bare Copper Dipole buried 3 inches in 30 mS/m Earth
dipole arm length 7.1 MHz impedance 14.25 MHz impedance
12 inches ........ 141 - j18 Ohms .... 128 - j30 Ohms
22 ................ 99 - j1 ........... 94 + j2
32 ................ 81 + j11 .......... 86 + j24
42 ................ 73 + j22 .......... 90 + j38
52 ................ 70 + j30 .......... 97 + j43
62 ................ 71 + j36 ......... 103 + j42
72 ................ 73 + j40 ......... 105 + j39
82 ................ 76 + j41 ......... 105 + j37
92 ................ 78 + j42 ......... 104 + j37
102 ............... 79 + j41 ......... 103 + j36
Because we are dealing with a very lossy medium, the current distribution on buried wires is not as clear-cut as it would be on a similar above-ground wire,
especially at higher frequencies. That is, even with the slower velocity of electromagnetic propagation on such a very lossy antenna, the wire current may be
overwhelmed by the high losses, and disappear before an obvious standing wave pattern can emerge.
So in summary it appears that a quick and dirty dummy load can be fashioned by simply burying a short horizontal dipole in the ground. If you try this, take
precautions to cover the feed point with a plastic bucket or other barrier to prevent accidental contact with the RF voltage, and wind a choke balun if you
suspect any RF voltage on your transmitter case. Also keep in mind that there will indeed be a significant amount of radiation into the air above the buried dipole,
which means that you can still interfere with other communications and there will be an increased possibility of electromagnetic radiation hazard to somebody
standing near the buried dipole.
Underground Communication
At this point it is convenient to extend the discussion of buried wires to include underground RF communication. NEC4.1 allows us to model insulated wires, so
we can fashion an air space around a buried dipole, thereby simulating an installation associated with an underground bunker. Assume that the antenna consists
of a six foot long, #12 AWG copper dipole buried three feet below the surface of 5 mS/m ground with a dielectric constant of 13. Our reference case will be a
buried bare wire in direct contact with the earth with no air space around it. The input impedance to this non-chambered antenna at 14.25 MHz is 87 - j89 ohms.
So you can see that burying the wire deeper in the soil with no part of the antenna above ground changes the input impedance considerably. Interestingly the one
kilowatt electric field intensity one kilometer above ground from this wire buried three meters under ground is about 10 mV/m rms, which is relatively large. The
law of reciprocity tells us that such an antenna needs to be buried deeper than three feet in order to avoid damaging the equipment connected to it in the event of
a nuclear burst, lightning, or other EMP producing event. The propagation loss within 5 mS/m soil is about 2.2 dB/m at 14.25 MHz. Thus if the dipole were
buried very near the surface, the one kilowatt field a kilometer up would be about 21 mV/m.
But this also tells us that this unusual dummy load serves as an invisible radiator, albeit a rather inefficient one! It could provide a very temporary, not very
practical solution to ham radio operators who live in residential areas that restrict tower and antenna erection, until they can at least install a decent attic antenna.
A half-wave dipole radiating one kilowatt and suspended three meters above 5 mS/m earth produces a field intensity of about 320 mV/m rms one km above
ground. Thus the short buried dipole field is about 24 dB below this [20 log (21/320)], which makes it a fairly effective dummy load and a poor radiator. Only
about four watts of the 1000 watts input to this antenna is radiated, and the rest is dissipated as heat in the earth and wire. This is a rather expensive way to
operate QRP! As I said, it is not a very practical antenna.
Keep in mind that the propagation of an electromagnetic wave through a lossy medium such as dirt does not behave the same way as the above-ground surface
wave behaves. The fact is that attenuation of the RF energy below ground is directly related to the ground conductivity. In other words, higher ground
conductivity means higher losses. This is opposite to surface wave behavior, where the least attenuation occurs over salt water, which has a conductivity of about
5000 mS/m and a dielectric constant of 80. By contrast, propagation within (not over) salt water is very poor compared to propagation within rock, for example.
In fact, there is no attenuation of an EM wave traveling through a perfect insulator. The EM wave may travel more slowly through an insulator, or lossless
dielectric, but that is another matter. As a practical matter, RF transmission through typical dry dirt from a buried transmitter to a buried receiver is a very poor
method of communication because of the rapid attenuation of the EM wave at HF. However, signal attenuation through solid rock is not as great. And at VLF,
underground communication is indeed practical because of the lower losses exhibited at lower frequencies.
Grant Bingeman is a Principal Engineer at Continental Electronics in Dallas, Texas. His e-mail address is DrBingo@compuserve.com.
www.qsl.net/km5kg