In alt.internet.wireless DaveC wrote:
Working with 2.4 GHz 802.11b wireless signal. Using USB adapter (one of those
thumb-sized products) and an old 18-inch satellite TV dish.

With a circular, symmetrical parabolic dish, I presume that the focus is
directly above the center. I've located the tiny antenna within the adapter
as close to this point as i can.

Oops. It's not circular or symmetric.
I've heard that it is a portion of a parabola, rather than a parabola in
itself, as if the oringal mounting point of the LNB were mounted in the
predicted focal point of a larger piece of metal that did form a parabola.

The proper focal point is wherever the antenna element was on the original
LNB. The frequency doesn't matter. That's still the focal point.
The next problem is knowing exactly where the antenna is in your USB
widget, so that you can put it in the same spot.

( Odd thing about this dish -- this one is a DishTV brand -- is that in the
original design for satellite reception, the driven element [ the "can", the
input to the LMB ] was located low on the dish. In other words, the focus
used in this design was off-center, about 30 degrees below the center axis of
the dish. This would mean that aiming would be off-axis, too, yes? What is
the reasoning behind this design? )

I think it's 22 degrees, but it is certainly "off". The old Primestar
dishes often looked like they were pointing at the ground, instead of into
the heavens.

I tried rotating the dish on its axis to account for polarity mismatch (
would this really be an issue? ). No joy.

My DLink DWL-122 USB adapter improves noticeably when the orientation is
correct. I tried the face facing the WAP, horizontal and vertical, as
well as edge on, horizontal and vertical. I didn't try end-on.
For my setup the face on - horizontal was best.

The wireless adapter can receive signals well, generally, w/o a dish, but
because of my location, reception of the desired signal is not great.

I put my USB widget 1" in front of a 9" pie tin. The 9" pie tin isn't
really quite large enough. 1" was discovered empirically, and appears to
be good for about 6db, according to Netstumbler. It is also far more
stable with the pie tin, where without it the signal seems to wander.

http://www.usbwifi.orcon.net.nz

His wok is a parabola. That doesn't count for you.

I was going to use a reflector like Trevor's BiQuad to feed the USB to a
Dish-Dish, but my son cleaned out his garage and threw the dish away about
a week before I asked for the dish ;-(.

I think the reflector behind the USB widget, painting a parabola, should be
a good combination. And the right focal point for the odd-shaped Dish-dish
is important.

http://www.trevormarshall.com/biquad.htm
http://martybugs.net/wireless/biquad/
http://63.142.46.238/buck/CantennaUSB.html

And my favorite site:
http://www.nodomainname.co.uk/cantenna/cantenna.htm
Where David uses a full sized USB adapter, pulling the antenna loose and
sticking it through the side of a can.

I might put my mini-usb inside a can.

--
---
Clarence A Dold - Hidden Valley (Lake County) CA USA 38.8-122.5

18 inch is much too small a diameter @ 2.4 Ghz for gain or directionality.
24 inch is "almost" workable, but it is easier to use a corner or yagi.
You can use a omni type generic antenna, and chicken wire, or hardware cloth
as reflector on a corner antenna.

Check your equations or gain, diameter with wavelength/frequency on the
parabolic.
If you aren't getting 10 dB out of it, it is not worth it. (you can get an
easy 6 from a simple corner reflector)
Illumination from the feed is important, should be 10 dB down at the edges.
Most likely it is not being fed at the right angle either, related to your
F/D ratio.

Note: lots of lame parabolic on the 802.11b market, too. (like a parabolic
on terrestrial TV)

Antennas are Fun! This is the place to post your questions.


"DaveC" wrote in message
al.net...
Working with 2.4 GHz 802.11b wireless signal. Using USB adapter (one of
those
thumb-sized products) and an old 18-inch satellite TV dish.

Calculating dish's focus point as f = ( d^^2 ) / ( 16 * c )
f = focus point above center of dish
d = diameter of dish
c = depth of dish at center

With a circular, symmetrical parabolic dish, I presume that the focus is
directly above the center. I've located the tiny antenna within the
adapter
as close to this point as i can.

( Odd thing about this dish -- this one is a DishTV brand -- is that in
the
original design for satellite reception, the driven element [ the "can",
the
input to the LMB ] was located low on the dish. In other words, the focus
used in this design was off-center, about 30 degrees below the center axis
of
the dish. This would mean that aiming would be off-axis, too, yes? What is
the reasoning behind this design? )

Setting up on the roof, finding the wireless access point's signal, I
moved
the wireless adapter around the focus point a little to maximize signal
strength (I use the word "maximize" loosely...). I then aimed the dish
around
in the general direction of the access point, looking for a leap in signal
strength (using a signal strength utility program to verify my aim ).

I find 2 things:
1. It doesn't really matter where I point the dish. The signal varies a
little bit when I approach the direction of the access point, but no
"leaps
and bounds" in the signal strength between being "dead on" and way off.

2. I pick up signals that aren't even in the direction of the access
point.

I tried rotating the dish on its axis to account for polarity mismatch (
would this really be an issue? ). No joy.

The wireless adapter can receive signals well, generally, w/o a dish, but
because of my location, reception of the desired signal is not great.

I'm aware that reflections can be strong from nearby objects, so that
could
account for some pickup of signals in directions other than from their
origin. Also, I suspect that the shallowness of the dish is a contributor.
But the lack of directionality of the dish, in general, has me stumped.
Have
I miscalculated the focus? I understand that the ratio of
focus-to-diameter
of the dish is important; for 2.4 GHz, is best in the 0.25-0.55 range
(this
one is 0.69). Is this contributing to my problem?

I used this web page as a reference:

http://www.usbwifi.orcon.net.nz

Any help would be greatly appreciated.

If this is off-topic for any NG here, please let me know what is a more
appropriate forum. Are there mailing lists for such topics?

Thanks,
--
DaveC

This is an invalid return address
Please reply in the news group