yes, this is a common problem. try checking out the arrl's radio direction
finding book for more specific techniques. but a couple simple ones are
1. a small loop, use the sharp nulls off the ends of the loop instead of the
big wide lobes.
2. a doppler system that electronically rotates antennas to give direction.
there is a very simple one on my web site at:
http://www.k1ttt.net/technote/doppler.html
3. a pair of phased verticals set up to give a cardioid pattern, again use
the sharp null not the fat lobe.

"Jim" wrote in message
...
This isn't strictly a Ham question, but I hope you all can help me anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000
feet
from my transmitter, but the closer I get, the more inaccurate it becomes.

What kind of antenna design could I switch to when I get to close range
that
would have a narrower beam so I could pinpint my target? It would be nice
to have something smaller than my 35" x 21" yagi for close in work, but
the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim

Jim wrote:
This isn't strictly a Ham question, but I hope you all can help me anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000 feet
from my transmitter, but the closer I get, the more inaccurate it becomes.

What kind of antenna design could I switch to when I get to close range that
would have a narrower beam so I could pinpint my target? It would be nice
to have something smaller than my 35" x 21" yagi for close in work, but the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim



Jim,

Assuming your transmitters are NOT super well filtered I would build a
yagi for 3 times the frequency (498-501MHz) and listen on that frequency
when very close. Very few transmitters will be so clean as to not be
able to hear the 3rd harmonic. I DF and do running ARDF very often and
if you don't mind the small second antenna and you have a receiver that
can tune to the 3rd harmonic this will get you both the needed
attenuation for being close and the ability to pinpoint the source.

73,
Larry, W0QE

On Tue, 11 Oct 2005 01:03:34 -0400, "Jim" wrote:

This isn't strictly a Ham question, but I hope you all can help me anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000 feet
from my transmitter, but the closer I get, the more inaccurate it becomes.

What kind of antenna design could I switch to when I get to close range that
would have a narrower beam so I could pinpint my target? It would be nice
to have something smaller than my 35" x 21" yagi for close in work, but the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim


You already got two good suggestions from others. Here's a third.
Build a MOXON square, less gain than yagi but if built right the MOXON
has a very distinct readward null, you exploit the null. Another
small advantage is it's smaller than a 2 element yagi for the same
band.

Allison

Jim wrote:
This isn't strictly a Ham question, but I hope you all can help me anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000 feet
from my transmitter, but the closer I get, the more inaccurate it becomes.

What kind of antenna design could I switch to when I get to close range that
would have a narrower beam so I could pinpint my target? It would be nice
to have something smaller than my 35" x 21" yagi for close in work, but the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim


The doppler shift method mentioned by Dave has worked very well for me
but it only gives you a chioce between 2 directions. If you know the
kind of area your turtles are likely to be then it should work
otherwise you can use iyt in conjunction with you yagi antenna to get a
accurate direction. The doppler shift method (unlike directional
antennas) is not affected by signal strength. This is because it works
by using two small antennas and determining which one the radio wave
hits first (this type of device usually generates a tone which
dissapears when the wave hits both antennas at the same time). This
tells you with a good degree of accuracy that the transmitter is in
front of or behind you.

Hi Jim

Hams will quite often get involved in hidden transmitter hunts during
field day games. You might be surprised at the expertise out there.

Some notes/thoughts for you;

- I'd be surprised that the yagi would exhibit 48dB F/B and if it did it
would only be for a very narrow frequency range. This however isnt a
huge problem. Even 20dB is more than ample for hidden TX finding.

- Ensure that you arent cross polarized with the turtle antenna. It is
possible for polarisation to change as the signal reflects off objects
so reflected signal maybe stronger than the direct. This only gets to be
a huge issue at cross polarised (say) plus or minus 10 degrees or so so
just make sure you are in the ballpark.

- Something that you modeled probably didnt allow for a ground so close
and your body so near. What I am trying to get at is you may not be
getting anything like the performance you expected because of these
local detuning effects.

- I'd suspect that the problem might actually be more in the receiver
than antenna. As the signal gets stronger/closer the radio might be
running out of signal reporting range,. ie everything is just over a
certain value so it is treated as being at maximum. It is quite common
under these conditions to switch in an amount of attenuation to reduce
the signal to a more reportable figure. This usaully goes in the antenna
feedline.

- In addition to the above your receiver may not be very well shielded
such that signal actually bypasses the antenna/coax path and thus gives
you less peaks and nulls to go off. You can check this by placing a
dummy load (usually 50 ohms) in the antenna cable socket and checking
for the nearby signal.

- The human body is excellent for using as a directional attenuator. If
you had a simple omni antenna attached and only a few inches in front of
your body you'll get maximum attenuation (lowest signal) when you are
facing away from the target turtle.

- If you antenna design isnt symetrical and there is an imbalance of
currents in the coax you may get to receive a signal directly to the
coax as well as the antenna. A few ferrite beads at the antenna
feedpoint around the coax, coiling the coax in a 4" dia 5 turn loop or
using some other form of balun (coaxial or otherwise might help.
Obviously you can test this theory by playing with the coax orientation
as well as the antennas.

- Considering also using the null or side of the antenna. ie turn it at
right angles to the turtle and where you get a very sharp loss of signal
thats the direction (or 180 degrees out) the TX is.

- Trying to narrow the beamwidth probably isnt a useful exercise with
the size limitations you have. You stated 90 degrees (3db down)
beamwidth. With (say) a 12 element yagi, 20ft long the beamwidth is
around 38 degrees. Better but probably not useful for you.

Hope this is useful to you.

Oh and a question. Is the box turtle the one one that has the hinged
flap in front? I found one on a busy road and I picked it up and moved
it along in what I hope was the direction it was going. Had never seen
one before! (New to the US)

Cheers Bob W5/VK2YQA

Jim wrote:
This isn't strictly a Ham question, but I hope you all can help me anyway.

actually... for some reason i didn't put it on that web page, but you can
have a unidirectional null with that simple system. to do that you make the
antennas 1/4 wavelength apart and make one feed line 1/4 wave longer than
the other. in this way you get a cardioid pattern since the only direction
that won't have a phase shift is when the signal gets to the antenna with
the longer feedline first and the other one exactly 1/4 cycle later. just
be sure to take the coax velocity factor into account.

"Duncan" wrote in message
ups.com...

Jim wrote:
This isn't strictly a Ham question, but I hope you all can help me
anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000
feet
from my transmitter, but the closer I get, the more inaccurate it
becomes.

What kind of antenna design could I switch to when I get to close range
that
would have a narrower beam so I could pinpint my target? It would be
nice
to have something smaller than my 35" x 21" yagi for close in work, but
the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim


The doppler shift method mentioned by Dave has worked very well for me
but it only gives you a chioce between 2 directions. If you know the
kind of area your turtles are likely to be then it should work
otherwise you can use iyt in conjunction with you yagi antenna to get a
accurate direction. The doppler shift method (unlike directional
antennas) is not affected by signal strength. This is because it works
by using two small antennas and determining which one the radio wave
hits first (this type of device usually generates a tone which
dissapears when the wave hits both antennas at the same time). This
tells you with a good degree of accuracy that the transmitter is in
front of or behind you.

Duncan wrote:
Jim wrote:

This isn't strictly a Ham question, but I hope you all can help me anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000 feet
from my transmitter, but the closer I get, the more inaccurate it becomes.

What kind of antenna design could I switch to when I get to close range that
would have a narrower beam so I could pinpint my target? It would be nice
to have something smaller than my 35" x 21" yagi for close in work, but the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim



The doppler shift method mentioned by Dave has worked very well for me
but it only gives you a chioce between 2 directions. If you know the
kind of area your turtles are likely to be then it should work
otherwise you can use iyt in conjunction with you yagi antenna to get a
accurate direction. The doppler shift method (unlike directional
antennas) is not affected by signal strength. This is because it works
by using two small antennas and determining which one the radio wave
hits first (this type of device usually generates a tone which
dissapears when the wave hits both antennas at the same time). This
tells you with a good degree of accuracy that the transmitter is in
front of or behind you.


Let's not forget that the antenna on the turtle will not polarized in
any particular fashion so antennas such as dopplers may not show nulls
and may show peaks incorrectly. I would take a yagi that I can rotate
to match the polarization to the source any day over TDOA or doppler
type antennas. Also I would always favor hunting the peak versus the
null especially if any reflections are present. If you are having
problems with a full sized yagi is it due to equipment limitations or to
physical constaints such as vegetation?

73,
Larry, W0QE

Bob Bob wrote in
:

Hi Jim

Hams will quite often get involved in hidden transmitter hunts during
field day games. You might be surprised at the expertise out there.

Some notes/thoughts for you;


- I'd suspect that the problem might actually be more in the receiver
than antenna. As the signal gets stronger/closer the radio might be
running out of signal reporting range,. ie everything is just over a
certain value so it is treated as being at maximum. It is quite common
under these conditions to switch in an amount of attenuation to reduce
the signal to a more reportable figure. This usaully goes in the
antenna feedline.

- In addition to the above your receiver may not be very well shielded
such that signal actually bypasses the antenna/coax path and thus
gives you less peaks and nulls to go off. You can check this by
placing a dummy load (usually 50 ohms) in the antenna cable socket and
checking for the nearby signal.

Ahhh, you beat me to this.

Now, if the receiver IS well shielded enough, if you can insert a
switchable attenuator in the coax, you can reduce the signal strength
significantly...and, FM receivers get very non-linear below about 20dB
quieting, they quiet faster than the actual input. In otherwords, you
could get 3-4 dB quieting change for 1dB of singal change. What this will
do is make you antenna pattern SEEM narrower.

What I used to do was to change to an RF field strengtth meter with a
small cavity filter tuned to the frequency I was hunting when I got
really close.


- The human body is excellent for using as a directional attenuator.
If you had a simple omni antenna attached and only a few inches in
front of your body you'll get maximum attenuation (lowest signal) when
you are facing away from the target turtle.

YES! But, you need to keep you body completely symetrical, stick an elbow
out and you will distort the pattern. Also, when the signal gets large
enough, remove the antenna.


-Bruce

Jim wrote:
"What kind of antenna design could I switch to when I get to close range
that would have a narrower beam so I could pinpoint my target?"

J. Roy Smith, W6YA described what he called "the simplest of DF loops,
with no ambiguity of direction" in CQ magazine`s 1963 "Antenna Roundup".
Roy also called ot the "Nobaloop".

Roy said it was designed for locating signals on the 10-meter band but
has been used with reasonable success on the lower frequency bands. Roy
used the loop on 80 transmitter hunts winning first place 15 times when
many of his competitors were using identical loops.

The loop is about a 1-meter length of 1/4-in. dia.copper tubing (gas
line) bent into a neat circle. The two ends are flattened for about half
an inch in a plane perpendicular to the radius. Number 28 holes are
drilled through the flattened tubing about 1/4-inch from the ends. The
loop is attached to a coax receptacle such as Amphenol 83-IR or military
type SO 239. One loop end is attached to the outer conductor by a 6-32
screw through one of the four holes. The other loop end is placed over
the receptacle`s center conductor lug and soldered well.

The small loop gives a figure-8 response in the plane of the loop. The
attached coax distorts this pattern, adding its vertical response, like
a sense antenna. The resulting pattern is a cardiod.

Nearby vertical objects may need detuning to avoid interference with the
vertical loop.

The maximum lobe is about 30-degrees wide.

The null is about 5-degrees wide, according to the author.

I have not built it but it seems simple, small, and easy to duplicate.

Best regards, Richard Harrison, KB5WZI

Sorry I'm too tired to look up a better example, but
http://members.aol.com/BmgEngInc/Adcock.html
is still pretty good. It's essentially about like the
"phased verticals" the other fellow mentioned, in that
they both result in a "cardioid" pattern, i.e. a sharp
*notch* in the pattern, broadside to the array in this
case. That notch will be far sharper than any beamwidth
in gain that a Yagi can possibly have, and, a notch is
what you want, close-in, rather than overloading your
receiver. The one I used was built by another ham, and
IIRC the twin-lead between the two vertical elements was
some weird, 50 or 75 ohm twinlead, which is going to be
hard to find, and it was crossed, as hinted at in the lower
diagram in the URL above, but I think it probably just
connected to the coax to the rig at that point, rather than
any sort of "Combiner Box" like that shown. We used them
to DF a guy who was jamming a 2M repeater net every week.
We had cheap compasses, like you'd have on a boat,
fixed to the broomstick "boom" of the array, and it was
suprisingly accurate at to where all the lines intersected
over a period of time. Don't know what ever happened;
it was coming from inside a secure telco property, so we
couldn't do much "up close and personal" and FCC did
not show up as they were allegedly scheduled to do the
night I was involved.

Just surf around for "Adcock" and DF, antenna, whatever,
and I'm sure you'll find what you need. The Doppler antenna
the other fellow mentioned are spiffy, but I don't know how
inexpensively they can be done. Last time I saw one, it
was a dedicated, factory built unit... maybe these days
it's possible to "power" the logic with a laptop or something..
It'd be very simple to scale a 2m (144-148 MHz) antenna to
your band...

Good luck..

HTH,
73,
Sluggo

On Tue, 11 Oct 2005 01:03:34 -0400, "Jim" wrote:

This isn't strictly a Ham question, but I hope you all can help me anyway.

I am using small transmitters in the 166-167 mhz range in some Box Turtle
research I am doing. My RDF antenna is a 3 element Yagi designed via
Yagicad 4.1 which works pretty well. It has 48db front/back and about 90
degrees beamwidth in the H pattern.

This works well for initial locating......usually starting 1500 to 2000 feet
from my transmitter, but the closer I get, the more inaccurate it becomes.

What kind of antenna design could I switch to when I get to close range that
would have a narrower beam so I could pinpint my target? It would be nice
to have something smaller than my 35" x 21" yagi for close in work, but the
beam width is the primary concern.

Yagicad doesn't let me design solely on beam width (at least I haven't
figured out how) so is there another way to go on this??

Thanks
Jim