Successful Eavesdropper antenna modification
Woofle
GROUNDING MATTERS:
I live in a house with a lot of electrical noise sources. Despite
putting up 75 feet of copper wire into the Hi-Z input of my Drake R8,
and having the radio room at the corner of the house pretty far from
the computers, I couldn't hear the weaker signals because they were
swamped by buzzing, hash and noise. Even very strong broadcasters
would fade briefly into a buzzing racket whenever the signal drops
below, say, S5. Sometimes the noise would abate or be particularly
bad, with no apparent cause. This was irritating.
Hoping to improve this situation the other day, I bought an Antenna
Supermarket Eavesdropper antenna, and 100' of RG/8X coax, and erected
the antenna as far from the house
as I could manage. I figured the coax and separation from the noise
sources would help reduce the interference. Bingo -- reception was
vastly improved. But still, the weaker signals faded into the
interference instead of into the noise and you could hear buzzing.
As I want to hear some tropical band stuff, this was still not good
enough.
So, after reviewing John Doty's post on minimizing noise and grounding
for shortwave antennas
(http://www.anarc.org/naswa/badx/antennas/low-noise_antenna.html), I
resolved to adapt those principles to my (resonant) setup. The
results were significant, especially with regard to noise at the lower
frequencies. In fact, at 4 Mhz, I can barely discern an increase in
noise level when I switch between the antenna and the unused antenna
input during the day, with computers, fans, etc. all running. Of
course, at night the noise rises as lightning, etc. propagate in.
Here's what I did:
Mod #1: Remove the PVC center insulator device on the Eavesdropper
and replace it with a 1:1 Balun made for 1.8-30Mhz. This, of course,
requires soldering. One nice "side-effect" of this change if you're
near strong BCB stations is that BCB AM signals are attenuated by the
balun. Apparently, it acts as a high-pass filter and cuts off below
1.8 Mhz. I could discern no difference in sensitivity at 5 Mhz, and
later in the evening heard a variety of signals on 60 and 75 meters
that I had never heard before (or had heard before only under
exceptional circumstances.)
Mod #2: Drive a copper ground stake into the soil directly below the
antenna and solder a short length of copper wire to the coax shield,
and connect the other end to the ground stake. Weather-seal the whole
mess.
Mod #3: Repeat Mod#2, but located the grounding rod near where the
coax enters the house. Again, weatherseal.
The result is very quiet reception. Signals now fade into hiss, but
only if they are very weak (for example, VNG, the 5 kW time station in
Australia on 16.000 Mhz will fade into hiss from time to time.)
As a result, I went from having trouble listening to Radio Netherlands
without reception being marred by buzzing noises to hearing the
following:
16000 kHz VNG Llandilo Australia (hovering just above the noise)
5025 Khz Radio Rebelda, Cuba (good signal)
5077 kHz Caracol, Columbia (good signal)
4919 Khz Radio Quito, Quito Equador (fair signal)
4914.4Khz Radio Cora, Lima Peru (weak)
6070Khz CRFX/CFRB (fair-to-weak signal)
4890 Khz NBC Port Moresby, Papua New Guineas (fair signal before
sunrise)
This experience really does suggest that a good direct ground
can really help out with noise...
John Doty
> Mod #1: Remove the PVC center insulator device on the Eavesdropper
> and replace it with a 1:1 Balun made for 1.8-30Mhz. This, of course,
> requires soldering. One nice "side-effect" of this change if you're
> near strong BCB stations is that BCB AM signals are attenuated by the
> balun. Apparently, it acts as a high-pass filter and cuts off below
> 1.8 Mhz. I could discern no difference in sensitivity at 5 Mhz, and
> later in the evening heard a variety of signals on 60 and 75 meters
> that I had never heard before (or had heard before only under
> exceptional circumstances.)
Good idea, good observation!
The balun is probably not acting as a high-pass filter. Most 1:1 baluns pass
signals down to zero frequency. I suggest instead that you've reduced the
response of your antenna to the vertically polarized radiation that AM
broadcasters transmit. I assume your dipole is horizontal, so it should not
respond to vertically polarized radiation. However, the coax shield also acts
as a "random wire" antenna, predominantly vertically polarized at mediumwave
frequencies. Without the balun, energy picked up in this mode flows to the
dipole and then couples into the inside of the cable and back to your
receiver. The balun blocks this path.
> 16000 kHz VNG Llandilo Australia (hovering just above the noise)
> 5025 Khz Radio Rebelda, Cuba (good signal)
> 5077 kHz Caracol, Columbia (good signal)
> 4919 Khz Radio Quito, Quito Equador (fair signal)
> 4914.4Khz Radio Cora, Lima Peru (weak)
> 6070Khz CRFX/CFRB (fair-to-weak signal)
> 4890 Khz NBC Port Moresby, Papua New Guineas (fair signal before
> sunrise)
Nice log! Where are you? What time of day was it? How high was the antenna? I
ask these questions because I'm trying to get a feel for the tradeoff between
vertical and horizontal polarization between 3-10 MHz. Theory suggests that
horizontal polarization should be better at higher shortwave frequencies and
vertical polarization should be better at mediumwave, but at the lower
shortwave frequencies the question is interesting.
Please answer to the group, as there may be others interested in this.
--
John Doty "You can't confuse me, that's my job."
Home: j...@w-d.org
Work: j...@space.mit.edu
Woofle
>Woofle wrote:
>
>> Mod #1: Remove the PVC center insulator device on the Eavesdropper
>> and replace it with a 1:1 Balun made for 1.8-30Mhz. This, of course,
>> requires soldering. One nice "side-effect" of this change if you're
>> near strong BCB stations is that BCB AM signals are attenuated by the
>> balun. Apparently, it acts as a high-pass filter and cuts off below
>> 1.8 Mhz. I could discern no difference in sensitivity at 5 Mhz, and
>> later in the evening heard a variety of signals on 60 and 75 meters
>> that I had never heard before (or had heard before only under
>> exceptional circumstances.)
>
>Good idea, good observation!
>
The whole thing was motivated by your web page. I still wish to thank
you for sharing the insights and methods so that others can experiment
and benefit from your experience. I'm hoping that I am doing
something similar that might be of interest to someone who is in a
very weak signal area, like I am, who might already have a trap dipole
installed.
Sometimes, you get the right idea for entirely wrong reasons. I think
this may be the case in terms of my reasoning for why I went to Ham
Radio Outlet and purchased a 1:1 balun. Fortunately, Nature is not
offended when you do the right thing for entirely wrong reasons,
because it ends up working even if the theory on which it is built is
entirely spurious.
(Flash back to those elaborate mechanisms with "epicycles" which
made an earth-centric solar system still accord rather closely with
observations, despite being entirely wrong.. :)
Anyways, my original thinking for removing the center insulator and
replacing it with a balun was that with just a center insulator, I
would be effectively grounding one leg of the dipole when grounding
the coax shield. To me, that would mean that 1/2 of the dipole would
be ineffective. With a balun, this would no longer be true. Very
simplistic thinking, but then, I'm a programmer, not a physicist!
Also, I wanted magnetic instead of electric coupling so as to reduce
static voltage transfer from the antenna to the feedline, hoping that
would reduce noise due to wind blowing on the antenna, etc.
>The balun is probably not acting as a high-pass filter. Most 1:1 baluns pass
>signals down to zero frequency. I suggest instead that you've reduced the
>response of your antenna to the vertically polarized radiation that AM
>broadcasters transmit. I assume your dipole is horizontal, so it should not
>respond to vertically polarized radiation.
Yes, the antenna is horizontal. The Eavesdropper is an 11-band
horizontal trap dipole design which is effectively 1/2 wavelength on
the major shortwave broadcast bands. For the first time in my life,
out West here, I am using my Drake R8 without the preamp on and
getting good results. (I do plan to also install a random-wire
untuned antenna with a 9:1 un-un one of these days, just to see how it
works by comparison. Certainly it will work better outside the bands
this dipole was cut/tuned/trapped for.)
>However, the coax shield also acts
>as a "random wire" antenna, predominantly vertically polarized at mediumwave
>frequencies. Without the balun, energy picked up in this mode flows to the
>dipole and then couples into the inside of the cable and back to your
>receiver. The balun blocks this path.
>
As I have 100 feet of coax cable, which was ungrounded originally,
I can definitely buy the idea that the coax shield was acting as a
significant antenna, and fed energy up to the dipole where it
re-radiated into differential mode and back down the coax to the
receiver.
If I unscrew the coax connector so that the center pin goes into the
radio, but the outer shield is disconnected, the noise level rises
dramatically. We're talking maybe 20 dB or more. Before the
grounding and balun, the noise level didn't change regardless of
whether the shield was part of the wiring. In fact, I had been in the
habit of touching only the center conductor or the shield to my
portable for listening in bed, and didn't notice a problem with using
only one connector over what I got from the R8.
But now, the results of using only the center pin are so much noisier
that I can't go back to doing this anymore. The shield really does
reduce noise via grounding. :-)
>> 16000 kHz VNG Llandilo Australia (hovering just above the noise)
>> 5025 Khz Radio Rebelda, Cuba (good signal)
>> 5077 kHz Caracol, Columbia (good signal)
>> 4919 Khz Radio Quito, Quito Equador (fair signal)
>> 4914.4Khz Radio Cora, Lima Peru (weak)
>> 6070Khz CRFX/CFRB (fair-to-weak signal)
>> 4890 Khz NBC Port Moresby, Papua New Guineas (fair signal before
>> sunrise)
>
>Nice log! Where are you? What time of day was it? How high was the antenna? I
The antenna is only about 30 feet high. This is all the supporting
trees were willing to give me. ;) As you mentioned earlier, it's
possible that the antenna is responding to vertically-polarized waves
at the lower frequencies anyhow?
I didn't take formal times, but basically I was listening off and on
from around 0400 and 1300 UT. Not exactly all night, but I woke up
several times during the night and tuned around to see what I could
hear with Passport blue pages in hand. I heard IDs of all but Radio
Cora and NBC (but I've heard NBC before and recognized the
programming.)
I'm in the San Fransisco Bay area, in the mountains above Santa Cruz
at an elevation of approximately 1700 feet. The nearest neighboring
house is easily 500 feet away, so the listening environment *should*
be very quiet. As there are two computers running Linux 24/7 at the
house, though, there was always computer hash. Receiver is a Drake R8
which I bought used from someone on this newsgroup. It's safe to say
that this is the first time I have been genuinely impressed by my R8,
which always seemed to be noisy and unable to resolve weak signals,
and often needed the preamp to get adequate audio. None of these
things are a problem anymore. Mark me impressed and pleased with the
results.
>ask these questions because I'm trying to get a feel for the tradeoff between
>vertical and horizontal polarization between 3-10 MHz. Theory suggests that
>horizontal polarization should be better at higher shortwave frequencies and
>vertical polarization should be better at mediumwave, but at the lower
>shortwave frequencies the question is interesting.
>
I can't comment on this. It does seem that a vertical would be better
for DXing on the lower bands, but I don't have the wherewithal to
support a large vertical or bury radials all over the yard.
>Please answer to the group, as there may be others interested in this.
Done. You know, experimentation is fun. I'm also thinking an
active loop would be a fun project.
-- Ross
Richie L. Allen
polarization of a signal - does it "stay" as transmitted or is it "skewed"?
Does a signal have multiple components after an ionospheric bounce, some
horizontal, some vertical?
Based on my observation in 2 meter weak signal work on the ham bands (low
end of 2 meter band, around 144.3 MHz) - vertical antennas pick up more
electrical noise than horizontal antennas. i.e. take the same 2 meter Yagi,
in vertical orientation it pick up more noise than in horizontal orientation
at the same location. I assume electrical noise from power lines,
appliances etc has a stronger "vertical" component. Does the same hold true
for frequencies below 30Mhz?
Thanks.
Richie
********************************
John Doty
>
> After two of three bounces off the ionosphere, what effect is there on
> polarization of a signal - does it "stay" as transmitted or is it "skewed"?
> Does a signal have multiple components after an ionospheric bounce, some
> horizontal, some vertical?
The ionosphere thoroughly scrambles the the transmitted polarization. Part of
the "fading" that is so annoying on shortwave is the effect of rapidly
changing polarization.
> Based on my observation in 2 meter weak signal work on the ham bands (low
> end of 2 meter band, around 144.3 MHz) - vertical antennas pick up more
> electrical noise than horizontal antennas. i.e. take the same 2 meter Yagi,
> in vertical orientation it pick up more noise than in horizontal orientation
> at the same location. I assume electrical noise from power lines,
> appliances etc has a stronger "vertical" component. Does the same hold true
> for frequencies below 30Mhz?
It's the same below 30 MHz: man-made noise is predominantly vertically
polarized. However, DX signals are also predominantly vertically polarized
near ground level. The proximity of the ground enhances vertically polarized
signals and attenuates horizontally polarized signals. The effect is very
strong if the antenna is less than 1/4 wavelength from ground, and weak if the
antenna is more than one wavelength from ground. The effect is strong for
signals coming in from low elevations (DX), and weaker for signals coming in
at higher elevations (nearby stations, regional weather, cosmic noise).
Thus, for antennas of different linear polarization within a wavelength of the
ground, the horizontally polarized antenna may be expected to have a better
(signal/man-made noise) ratio, but a worse (signal/natural noise) ratio.
What's best is difficult question. I'm therefore very interested to learn that
a dipole only 9 meters above the ground is capable of respectable performance
at 60 meters if it's set up to minimize coupling of conducted noise and
feedline pickup. That's real data!
John Doty
> The whole thing was motivated by your web page. I still wish to thank
> you for sharing the insights and methods so that others can experiment
> and benefit from your experience.
Well, thank Steve Byan. He encouraged me to keep writing this stuff, collected
the best of it, turned it to HTML, and put it on the BADX web site.
Woofle
>Thus, for antennas of different linear polarization within a wavelength of the
>ground, the horizontally polarized antenna may be expected to have a better
>(signal/man-made noise) ratio, but a worse (signal/natural noise) ratio.
>What's best is difficult question. I'm therefore very interested to learn that
>a dipole only 9 meters above the ground is capable of respectable performance
>at 60 meters if it's set up to minimize coupling of conducted noise and
>feedline pickup. That's real data!
Surprisingly, I read the following while looking over the Eavesdropper
installation manual:
"the Eavesdropper is designed to be suspended with the center of the
antenna about 25 feet above ground."
This is especially strange as the antenna goes down to 90m coverage,
(90, 75, 60, 49, 41, 31, 21,19, 16, 13, and 11m), where this would be
only about 1/10th wavelength above the ground!
One antenna configuration that has captured my interest for further
experimentation has been the inverted-V, in a multi-wire/multiband
configuration. However,, it *would* be interesting to compare a
non-resonant configuration -- especially vertically polarized -- on DX
signals in an A/B comparison. So far, I've seen nothing about how
tall to make a vertical, and how to feed it, when multiband operation
is desired. Do you have any pointers as to how high/how to feed?
Thanks for any insight!
-- Ross
John Doty
> Surprisingly, I read the following while looking over the Eavesdropper
> installation manual:
>
> "the Eavesdropper is designed to be suspended with the center of the
> antenna about 25 feet above ground."
>
> This is especially strange as the antenna goes down to 90m coverage,
> (90, 75, 60, 49, 41, 31, 21,19, 16, 13, and 11m), where this would be
> only about 1/10th wavelength above the ground!
The height affects the tuning a little, so they probably mean it resonates at
the intended frequencies when suspended there.
> One antenna configuration that has captured my interest for further
> experimentation has been the inverted-V, in a multi-wire/multiband
> configuration. However,, it *would* be interesting to compare a
> non-resonant configuration -- especially vertically polarized -- on DX
> signals in an A/B comparison. So far, I've seen nothing about how
> tall to make a vertical, and how to feed it, when multiband operation
> is desired. Do you have any pointers as to how high/how to feed?
Almost any unbalanced antenna will be predominantly vertically polarized if
it's less than a quarter wavelength high. For a nonresonant antenna, feeding
at ground level seems best theoretically. For a straight vertical, 30 feet
works well from longwave through shortwave. The stuff at
http://www.anarc.org/naswa/badx/antennas/ describes the inverted-L
configuration, but applies equally to verticals.
Now for a crazy suggestion: you've decoupled the common mode of your coax
downlead from your dipole with a balun. Why not couple to that mode, as a
vertical? Here's what to do:
Get an Amidon BN-43-7051 "binocular" core (you can order it at
http://www.bytemark.com/). Get a couple of feet of RG-174 coax, and wind three
turns on the core (a "turn" is up through one hole, down through the other).
This is your primary winding. The secondary is just a single turn of ordinary
wire. Install this transformer at the ground point for your coax downlead,
feeding the downlead signal through the RG-174, just as if it was a straight
piece of cable. Connect a separate coaxial feedline to the secondary,
grounding the shield at your ground points: this is the feed for your vertical.
The fancy core is needed here to get enough magnetizing inductance, since you
can't put many turns of coax (even skinny RG-174) on a transformer. Three
turns is enough for shortwave and mediumwave on this core, but longwave
performance may be poor.
The presence of the core has no effect on the differential mode signal inside
the cable. RG-174 is a bit too lossy for long runs, but the losses in two feet
are negligible.
Woofle
>The height affects the tuning a little, so they probably mean it resonates at
>the intended frequencies when suspended there.
Probably. It is doubtful that going higher could harm performance in
terms of radiation angle, but perhaps tuning would be affected.
As an update, within the SW broadcast bands this antenna is still
pulling in the same signals with remarkable consistency. For example,
I get the 5kW time signal VNG Australia on 16000 Khz readably every
time I've tried for it in the evenings.
Most of the 60m stuff is there to a lesser or greater extent
day-by-day. Caracol is always fairly strong.
Having a non-resonant antenna will be great, especially since my
receiver has an "A/B" antenna switch to make comparisons easy.
If nothing else, the non-resonant antenna will be better for utility
monitoring and possibly low-angle DX (I hope!).
>> One antenna configuration that has captured my interest for further
>> experimentation has been the inverted-V, in a multi-wire/multiband
>> configuration. However,, it *would* be interesting to compare a
>> non-resonant configuration -- especially vertically polarized -- on DX
>> signals in an A/B comparison. So far, I've seen nothing about how
>> tall to make a vertical, and how to feed it, when multiband operation
>> is desired. Do you have any pointers as to how high/how to feed?
>
>Almost any unbalanced antenna will be predominantly vertically polarized if
>it's less than a quarter wavelength high. For a nonresonant antenna, feeding
>at ground level seems best theoretically. For a straight vertical, 30 feet
>works well from longwave through shortwave. The stuff at
>http://www.anarc.org/naswa/badx/antennas/ describes the inverted-L
>configuration, but applies equally to verticals.
>
30 feet would be easy to do with a wire antenna, given a strong enough
slingshot... but I worry about proximity to trees. A rigid radiator
would be harder for me to manage, especially with the high winds which
can blow up here. I suppose one alternative would be to loft two rope
segments up different trees, and tie the inside ends together via an
insulator with wire connected to it, then pull the whole mess up to 30
feet, and feed the wire at ground level.
>Now for a crazy suggestion: you've decoupled the common mode of your coax
>downlead from your dipole with a balun. Why not couple to that mode, as a
>vertical? Here's what to do:
>
Is the purpose of this to couple the antenna to feedline magnetically
instead of electrically? I can't figure out why I wouldn't just use
an 9:1 MLB-type configuration for feeding.
That said, I've been meaning to learn how to wind toroids and other
stuff but have never gotten around to it. Which is to say I'm not
adverse to experimentation.
John Doty
> >Almost any unbalanced antenna will be predominantly vertically polarized if
> >it's less than a quarter wavelength high. For a nonresonant antenna, feeding
> >at ground level seems best theoretically. For a straight vertical, 30 feet
> >works well from longwave through shortwave. The stuff at
> >http://www.anarc.org/naswa/badx/antennas/ describes the inverted-L
> >configuration, but applies equally to verticals.
> >
>
> 30 feet would be easy to do with a wire antenna, given a strong enough
> slingshot... but I worry about proximity to trees. A rigid radiator
> would be harder for me to manage, especially with the high winds which
> can blow up here. I suppose one alternative would be to loft two rope
> segments up different trees, and tie the inside ends together via an
> insulator with wire connected to it, then pull the whole mess up to 30
> feet, and feed the wire at ground level.
Don't worry about proximity to trees. It makes very little difference. Just
use insulated hook-up wire. You don't need copper-plated steel wire for a
vertical, because it takes very little tension to hold it up.
> >Now for a crazy suggestion: you've decoupled the common mode of your coax
> >downlead from your dipole with a balun. Why not couple to that mode, as a
> >vertical? Here's what to do:
> >
>
> Is the purpose of this to couple the antenna to feedline magnetically
> instead of electrically? I can't figure out why I wouldn't just use
> an 9:1 MLB-type configuration for feeding.
This *is* a 9:1 impedance transformation from the common mode of the downlead
to the feedline for the vertical, but the trick is that it doesn't affect the
differential mode of the dipole's downlead, so the downlead can act both as a
transmission line for the dipole and as a vertical antenna at the same time.
From the transformer you run two feedlines to your receiver: the original
dipole feedline, plus a new line connected to the transformer's secondary.
These effectively feed independent, cross-polarized antennas.