Help with 900Mhz turnstile....
 

I would like to build a 908Mhz turnstile.

How does one normally go about testing and adjusting an
antenna at that high of frequency?


For a lower frequency I can hook up a MFJ antenna analizer or an SWR
meter,.

I want to build a 900Mhz antenna with a hemispherical pattern.
It is going on an RC plane, so a normal vertical style antenna will
be problematic as it will have a null straight down.

So I though a turnstile antenna would be perfect.


Any advise on designing such a device?


Paul Kl7JG

On Mon, 14 Jul 2003 20:41:53 -0700, wrote:

I would like to build a 908Mhz turnstile.

How does one normally go about testing and adjusting an
antenna at that high of frequency?


For a lower frequency I can hook up a MFJ antenna analizer or an SWR
meter,.

I want to build a 900Mhz antenna with a hemispherical pattern.
It is going on an RC plane, so a normal vertical style antenna will
be problematic as it will have a null straight down.

So I though a turnstile antenna would be perfect.


Any advise on designing such a device?


Paul Kl7JG


Hi Paul,

1. build a 90.8MHz version and scale it down;
2. build an Inverted-F
see:
http://www.qsl.net/kb7qhc/antenna/In...%20F/index.htm

73's
Richard Clark, KB7QHC

On Tue, 15 Jul 2003 11:56:58 GMT, Dale Parfitt wrote:


Is it my browser or are the images on that page next to useless? I could
not figure out the antenna's configuration, or read the graph ordinates?
Text was less helpful.
Is there another site that discusses this antenna Richard?

Dale

Hi Dale,

Graphics straight from EZNEC which has served this group for years.
Select any graphic and simply ask the browser to view that page. If
you are unfamiliar with the drawing conventions I would suggest
getting the free version and becoming adept. The SWR drawings show 40
data points quite clearly between 440 and 460 (every 0.5MHz) no more
resolution is required, there are no surprise resonances hidden
between data points.

This is a very simple antenna, the description of the main radiator:
"This is a low, long antenna roughly quarter wavelength from the
ground point, up to the bend, out to the end. The drive point was
found by trial and error."
is quite explicit but I already admit may be confusing, thus the
attending illustration. There is only one ground, there is only one
bend, there is only one feed. The name describes it as well:
"Inverted F" The letter F certainly is descriptive. The only
embellishment is that it is constructed with a sheet rather than a
wire, but both sheet and wire models' SWR are also shown and
discussed:
"Because of the sheet metal form, this qualifies as a thick
radiator. The SWR curve on the left confirms this when compared to
the SWR curve below."
"This is the SWR curve for a one wire model of the design modeled
here."

The design couldn't be simpler for those already adept at building
grounded verticals with a gamma feed - and without the cap of the
gamma feed. The vertical is simply folded over at the feed.

73's
Richard Clark, KB7QHC

I'm not antenna adept, so in trying to read between the lines I have a few questions

Here is my ASCII art diagram....


XMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
M F
M F
M F
M F
M F
M F
M F
M F
GGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGG GGGGGGGGGGGGG



M= metal strip.
X= corner where metal strip is bent.
F = feed
G = ground plane.


How is the feed setup?
If I am driving it with coax the ground braid ends at the ground plane and the center conductor goes
up to the FM junction point?


The strip is 1/4 wavelength off of the ground plane, but how long is it from the bend to the end?
How wide is it?

What frequency did you build it for?

How did it work?

Paul KL7JG
















guy, so On Tue, 15 Jul 2003 05:06:54 GMT, Richard Clark wrote:

On Mon, 14 Jul 2003 20:41:53 -0700, wrote:

I would like to build a 908Mhz turnstile.

How does one normally go about testing and adjusting an
antenna at that high of frequency?


For a lower frequency I can hook up a MFJ antenna analizer or an SWR
meter,.

I want to build a 900Mhz antenna with a hemispherical pattern.
It is going on an RC plane, so a normal vertical style antenna will
be problematic as it will have a null straight down.

So I though a turnstile antenna would be perfect.


Any advise on designing such a device?


Paul Kl7JG


Hi Paul,

1. build a 90.8MHz version and scale it down;
2. build an Inverted-F
see:
http://www.qsl.net/kb7qhc/antenna/In...%20F/index.htm

73's
Richard Clark, KB7QHC

On Tue, 15 Jul 2003 09:25:07 -0700, wrote:

I'm not antenna adept, so in trying to read between the lines I have a few questions

Here is my ASCII art diagram....


XMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
M F
M F
M F
M F
M F
M F
M F
M F
GGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGG GGGGGGGGGGGGGG

yup


M= metal strip.
X= corner where metal strip is bent.
F = feed
G = ground plane.


How is the feed setup?
If I am driving it with coax the ground braid ends at the ground plane and the center conductor goes
up to the FM junction point?

perfect



The strip is 1/4 wavelength off of the ground plane, but how long is it from the bend to the end?
How wide is it?

What frequency did you build it for?

How did it work?

Paul KL7JG


Hi Paul,

Total radiator length (M) is ¼ wavelength (approx).

Use the drawing (viewed as its own page) to print a copy and scale all
dimensions by wavelength. Do not presume any correction factors,
simply use one Meter to be one Meter (not 0.95 Meter) at one Meter
wavelength. I deliberately left the segments visible to aid in
scaling. You should be able to perceive that the structure folds at
roughly 0.04 wavelength, which is also the height above the ground
plane. The foreshortened width of the sheet metal is also on the
scale of 0.04 wavelength.

For your application at 900MHz be aware that even thin wire (as in the
feed) becomes significantly thick enough to change resonance and feed
resistance through slight geometrical variations. This is true for
any construction. That is why I provided the caution that it was
chosen (or discovered) by trial and error. Given the 10MHz wide 2:1
BW, discovery covers a lot of turf. By purposely using thick
components, this becomes less an issue. At these very short
wavelengths, it is a rather simple matter to build deliberately wide
resonant (low Q) structures.

73's
Richard Clark, KB7QHC

Use the drawing (viewed as its own page)
Maybe I'm not getting it, but you seem to imply that there are things on your page to click
on to open in a larger window????

When I load the webpage there is not "Larger" link and the pictures on the webpage are too small to read.

Paul (KL7JG)



On Tue, 15 Jul 2003 17:06:25 GMT, Richard Clark wrote:

On Tue, 15 Jul 2003 09:25:07 -0700, wrote:

I'm not antenna adept, so in trying to read between the lines I have a few questions

Here is my ASCII art diagram....


XMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMMM
M F
M F
M F
M F
M F
M F
M F
M F
GGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGGG GGGGGGGGGGGGGGG

yup


M= metal strip.
X= corner where metal strip is bent.
F = feed
G = ground plane.


How is the feed setup?
If I am driving it with coax the ground braid ends at the ground plane and the center conductor goes
up to the FM junction point?

perfect



The strip is 1/4 wavelength off of the ground plane, but how long is it from the bend to the end?
How wide is it?

What frequency did you build it for?

How did it work?

Paul KL7JG


Hi Paul,

Total radiator length (M) is ¼ wavelength (approx).

Use the drawing (viewed as its own page) to print a copy and scale all
dimensions by wavelength. Do not presume any correction factors,
simply use one Meter to be one Meter (not 0.95 Meter) at one Meter
wavelength. I deliberately left the segments visible to aid in
scaling. You should be able to perceive that the structure folds at
roughly 0.04 wavelength, which is also the height above the ground
plane. The foreshortened width of the sheet metal is also on the
scale of 0.04 wavelength.

For your application at 900MHz be aware that even thin wire (as in the
feed) becomes significantly thick enough to change resonance and feed
resistance through slight geometrical variations. This is true for
any construction. That is why I provided the caution that it was
chosen (or discovered) by trial and error. Given the 10MHz wide 2:1
BW, discovery covers a lot of turf. By purposely using thick
components, this becomes less an issue. At these very short
wavelengths, it is a rather simple matter to build deliberately wide
resonant (low Q) structures.

73's
Richard Clark, KB7QHC

On Tue, 15 Jul 2003 13:16:29 -0700, wrote:

Use the drawing (viewed as its own page)
Maybe I'm not getting it, but you seem to imply that there are things on your page to click
on to open in a larger window????

When I load the webpage there is not "Larger" link and the pictures on the webpage are too small to read.

Paul (KL7JG)


Hi Paul,

Right click the picture and select "View Picture" option. This will
display the picture in its 640x480 native size (the picture is only
reduced on the page, not at the server).

73's
Richard Clark, KB7QHC

A while back, QST had an article on building a GPS turnstile antenna. The
article might be useful. Check with the ARRL for a reprint.

Randy ka4nma

On Wed, 16 Jul 2003 22:27:12 -0700, wrote:

Richard,
Could you please post your NEC source file for this antenna?

I get the same patttern, but the SWR / impedance is all out of wack.


Paul (Kl7JG)

Hi Paul,

This is first for me, find attached the EZ file for one of several
models I worked on. Let me know how useful it is (it may be the wrong
one, but it is certainly very close to what I published).

73's
Richard Clark, KB7QHC

On Thu, 17 Jul 2003 05:55:36 GMT, Richard Clark
wrote:

Hmmmm,

Not quite what I expected. I will do this the old fashion way,
e-mail.

73's
Richard Clark, KB7QHC