Hello,

I just finished building this one:

http://www.drslick.org/Temp1/yagi929.jpg

It's quite unusual, in that there are no directors and two
reflectors.

Here's the H-plane plot (the pattern looking down on the antenna
from above, when the elements are vertical) in the yagi optimizer
program that i used:

http://www.drslick.org/Temp1/yagi929.jpg


It's an interesting method that my friend used to get the
pattern we wanted. We couldn't use the optimization algorithms,
because these are designed to maximize the frontal lobe (can become
quite narrow in span), whereas we wanted something with a 180 degree
pattern.
Notice it's about only about -2dB at 90 and 270 degrees. Also
note that the outer edge of the pattern is about 4.51 dBi, so the dBd
(decibals over a dipole) will be somewhere around +2.4dBd or so.
This isn't that much gain, we realized, but the constraining
factor was to make the yagi as compact as possible, so as to keep the
landlord quiet. So we ended up with 3 elements minimum, in order to
achieve the pattern and get close to 50 ohms.

This antenna was designed for use when you are on the edge of
your city, and you don't wanna waste ERP into the mountains or
whatnot.

As you can see in the Jpeg, it's built with HomeDepot stuff: 2"
pvc and 5/16" treaded stock rod, which comes in lengths of 3 feet,
with nuts used to attach lugs that are soldered to an SO-239, and for
attaching to the PVC itself. Buy the connecting "link" nuts (they are
longer than regular nuts), and you can put these on the ends to tune
the lengths for best SWR.

Here's the basic plans:

Position (inches) 1/4 wavelength
(inch)x two

Driven element: 0 29.4"
Reflector 1: 12" 38.01"
Mast: 48" try for non-resonant
length
Reflector 2: 58" 32"


So you will obviously have to make a two times the 1/4
wavelength that you see, to make a dipole. The DE has a space between
the elements, but the reflector sections can be attached inside the
PVC.

Supposedly, the mast has no effect at a distance of 48" from
the DE, but when i simulated this in the YO, the length of the mast
DID matter, so it's advised that one make the mast at a non-resonant
length, which for 92.9 megs would be avoiding a multiple of 5 feet.

I didn't get a chance to measure the F/B ratio, nor do i really
have the proper equipment to do so, but it's certainly directional, as
a field test has proven.

Questions and comments appreciated.


Dr. Slick

(not a real Doctor)

Dr. Slick wrote:
Here's the H-plane plot

Nope, it's not but an H-plane plot would be interesting.
--
73, Cecil, W5DXP

Interesting photo but NO PATTERN

Dr. Slick wrote:

Hello,

I just finished building this one:

http://www.drslick.org/Temp1/yagi929.jpg

It's quite unusual, in that there are no directors and two
reflectors.

Here's the H-plane plot (the pattern looking down on the antenna
from above, when the elements are vertical) in the yagi optimizer
program that i used:

http://www.drslick.org/Temp1/yagi929.jpg


It's an interesting method that my friend used to get the
pattern we wanted. We couldn't use the optimization algorithms,
because these are designed to maximize the frontal lobe (can become
quite narrow in span), whereas we wanted something with a 180 degree
pattern.
Notice it's about only about -2dB at 90 and 270 degrees. Also
note that the outer edge of the pattern is about 4.51 dBi, so the dBd
(decibals over a dipole) will be somewhere around +2.4dBd or so.
This isn't that much gain, we realized, but the constraining
factor was to make the yagi as compact as possible, so as to keep the
landlord quiet. So we ended up with 3 elements minimum, in order to
achieve the pattern and get close to 50 ohms.

This antenna was designed for use when you are on the edge of
your city, and you don't wanna waste ERP into the mountains or
whatnot.

As you can see in the Jpeg, it's built with HomeDepot stuff: 2"
pvc and 5/16" treaded stock rod, which comes in lengths of 3 feet,
with nuts used to attach lugs that are soldered to an SO-239, and for
attaching to the PVC itself. Buy the connecting "link" nuts (they are
longer than regular nuts), and you can put these on the ends to tune
the lengths for best SWR.

Here's the basic plans:

Position (inches) 1/4 wavelength
(inch)x two

Driven element: 0 29.4"
Reflector 1: 12" 38.01"
Mast: 48" try for non-resonant
length
Reflector 2: 58" 32"


So you will obviously have to make a two times the 1/4
wavelength that you see, to make a dipole. The DE has a space between
the elements, but the reflector sections can be attached inside the
PVC.

Supposedly, the mast has no effect at a distance of 48" from
the DE, but when i simulated this in the YO, the length of the mast
DID matter, so it's advised that one make the mast at a non-resonant
length, which for 92.9 megs would be avoiding a multiple of 5 feet.

I didn't get a chance to measure the F/B ratio, nor do i really
have the proper equipment to do so, but it's certainly directional, as
a field test has proven.

Questions and comments appreciated.


Dr. Slick

(not a real Doctor)

Dave Shrader wrote in message news:CA44c.6318$C51.52257@attbi_s52...
Interesting photo but NO PATTERN


Opps! Here is the H-plane plot:

http://www.drslick.org/Temp1/yagiplot.jpg


It was an interesting usage of the program, in that we used a
"dummy" reflector that was very far away from the DE, and then used
two directors as reflectors (notice on the plot that the lobe is
actually pointed in the 180 degree direction).


Taking a look at the photo again:

http://www.drslick.org/Temp1/yagi929.jpg

You will notice the 4 turns of 4" diameter coils in the RG-213
coax, just before the feedpoint (SO-239), using a plastic tie-wrap.
You can use a balun instead, but i have never done this before,
perhaps one of you out there has something i could try in the future.
It would be a 1:1 in this case.
You'll also notice that i didn't just leave the coax hanging off
the feed point, as this might ruin the pattern and impedance. I tried
to keep the coax at 90 degrees to the elements by tie-wrapping it to
the PVC.

It would be nice to borrow an anechoic chamber with the proper
field-strength meter, and see how closely the actual unit matches the
YO program in the H-plane.

Great fun, and it should be able to handle up to 500 watts or so,
but i don't recommend using anything more than 1 mW or so... hehehe!

Questions and comments appreciated.


Slick

If you put the array on a rotator and put a receive antenna with some
sensitive output indicator beyond the near field, you can plot relative
field, and thus accurate F/B by adjusting the power to the antenna for same
received F/S at, say, 10º increments. Nothing else should change, so the
relative field strength in the various directions should relate to each
other. Ground reflections and absorption, and all the other artifacts will
be constant unless you change something else.

--
Crazy George
Remove N O and S P A M imbedded in return address
"Dr. Slick" wrote in message
om...
Dave Shrader wrote in message
news:CA44c.6318$C51.52257@attbi_s52...
Interesting photo but NO PATTERN


Opps! Here is the H-plane plot:

http://www.drslick.org/Temp1/yagiplot.jpg


It was an interesting usage of the program, in that we used a
"dummy" reflector that was very far away from the DE, and then used
two directors as reflectors (notice on the plot that the lobe is
actually pointed in the 180 degree direction).


Taking a look at the photo again:

http://www.drslick.org/Temp1/yagi929.jpg

You will notice the 4 turns of 4" diameter coils in the RG-213
coax, just before the feedpoint (SO-239), using a plastic tie-wrap.
You can use a balun instead, but i have never done this before,
perhaps one of you out there has something i could try in the future.
It would be a 1:1 in this case.
You'll also notice that i didn't just leave the coax hanging off
the feed point, as this might ruin the pattern and impedance. I tried
to keep the coax at 90 degrees to the elements by tie-wrapping it to
the PVC.

It would be nice to borrow an anechoic chamber with the proper
field-strength meter, and see how closely the actual unit matches the
YO program in the H-plane.

Great fun, and it should be able to handle up to 500 watts or so,
but i don't recommend using anything more than 1 mW or so... hehehe!

Questions and comments appreciated.


Slick

If you put the array on a rotator and put a receive antenna with some
sensitive output indicator beyond the near field, you can plot relative
field, and thus accurate F/B by adjusting the power to the antenna for same
received F/S at, say, 10º increments. Nothing else should change, so the
relative field strength in the various directions should relate to each
other. Ground reflections and absorption, and all the other artifacts will
be constant unless you change something else.


ok, that sounds like a good idea, with the rotator, keeping everything else
the same. This site:

http://radioproshop.com/antennaconcepts.htm

Says that the near-field is defined by Rmin=(2*D**2)/wavelength, where
D is the largest dimension of the antenna. So if we roughly say one wavelength
is 10 feet, and the largest dimension is the dipole, then we get Rmin=5 feet,
or about a 1/2 wavelength. This sounds a bit too close to me, perhaps the
far-field is better defined as, like, ten wavelengths or so?


Also, i don't really have a transmitter that I can vary the power that
much...maybe that's something i should make or get. Something variable at
least by 20dB or so.


Slick

If the second reflector is actually doing anything, which I doubt, you
should be using a non metallic mast, and not bringing the coax down between
the reflectors.

Tam/WB2TT

"Dr. Slick" wrote in message
om...
Hello,

I just finished building this one:

http://www.drslick.org/Temp1/yagi929.jpg

It's quite unusual, in that there are no directors and two
reflectors.

Here's the H-plane plot (the pattern looking down on the antenna
from above, when the elements are vertical) in the yagi optimizer
program that i used:

http://www.drslick.org/Temp1/yagi929.jpg


It's an interesting method that my friend used to get the
pattern we wanted. We couldn't use the optimization algorithms,
because these are designed to maximize the frontal lobe (can become
quite narrow in span), whereas we wanted something with a 180 degree
pattern.
Notice it's about only about -2dB at 90 and 270 degrees. Also
note that the outer edge of the pattern is about 4.51 dBi, so the dBd
(decibals over a dipole) will be somewhere around +2.4dBd or so.
This isn't that much gain, we realized, but the constraining
factor was to make the yagi as compact as possible, so as to keep the
landlord quiet. So we ended up with 3 elements minimum, in order to
achieve the pattern and get close to 50 ohms.

This antenna was designed for use when you are on the edge of
your city, and you don't wanna waste ERP into the mountains or
whatnot.

As you can see in the Jpeg, it's built with HomeDepot stuff: 2"
pvc and 5/16" treaded stock rod, which comes in lengths of 3 feet,
with nuts used to attach lugs that are soldered to an SO-239, and for
attaching to the PVC itself. Buy the connecting "link" nuts (they are
longer than regular nuts), and you can put these on the ends to tune
the lengths for best SWR.

Here's the basic plans:

Position (inches) 1/4 wavelength
(inch)x two

Driven element: 0 29.4"
Reflector 1: 12" 38.01"
Mast: 48" try for non-resonant
length
Reflector 2: 58" 32"


So you will obviously have to make a two times the 1/4
wavelength that you see, to make a dipole. The DE has a space between
the elements, but the reflector sections can be attached inside the
PVC.

Supposedly, the mast has no effect at a distance of 48" from
the DE, but when i simulated this in the YO, the length of the mast
DID matter, so it's advised that one make the mast at a non-resonant
length, which for 92.9 megs would be avoiding a multiple of 5 feet.

I didn't get a chance to measure the F/B ratio, nor do i really
have the proper equipment to do so, but it's certainly directional, as
a field test has proven.

Questions and comments appreciated.


Dr. Slick

(not a real Doctor)

OK, one alternative is to use a known accurate (1dB) step attenuator in the
receive line to get the relative numbers. Always go for the same strength
indication on the receiver. Start at the azimuth with lowest indication,
and add attenuation as it rotates.

--
Crazy George
Remove N O and S P A M imbedded in return address

"Tam/WB2TT" wrote in message ...
If the second reflector is actually doing anything, which I doubt, you
should be using a non metallic mast, and not bringing the coax down between
the reflectors.

Tam/WB2TT


Actually, if you believe the yagi optimizer program, the second
reflector is indeed not doing TOO much, but it does affect the back
shape of the pattern, and improves the impedance bit, and also
increases the dBd of the outer edge of the pattern by tiny bit too.

And i did simulate having the mast at 48 inches from the DE, which
i was advised was the distance where the mast would have the least
affect. Changing the mast length in the YO program does affect the
pattern shape though, so i tried to make the mast a non-resonant
length.

At any rate, we could probably survive without the second
reflector, you are most likely correct, but maybe for the next design.

It first, i considered having the "T" section of PVC pipe move
outwards at a 90 degree angle, and then attach another section of 2"
PVC that would be sticking out, so that the coax would be totally
perpendicular to all the elements. But, this would be a bit unwieldy.



Slick

"Radio913" wrote in message
...
If you put the array on a rotator and put a receive antenna with some
sensitive output indicator beyond the near field, you can plot relative
field,

SNIP


Also, i don't really have a transmitter that I can vary the power that
much...maybe that's something i should make or get. Something variable at
least by 20dB or so.


Slick


You don't need to pump power to measure antenna patterns. A simple signal
generator with a milliwatt or so will work fine. If you want to be really
serious about designing antennas, then you need to start acquiring serious
tools. See eBay; start bidding on a sig gen, some pads (attenuators), a
rotor, a field strength meter.

Ed