On Wed, 15 Apr 2015 21:05:10 -0700, Jeff Liebermann
wrote:
Oops. I "send" before I was done. Continuing.
So... now I need to buy or make antennas for them. I'm interested in
ideas and suggestions.
Hint: 900 MHz radios don't like high VSWR antennas. I suggest you
optimize the antenna for the 902 MHz TX area.
I've seen a number of DIY designs on the net:
Allow me to take some pot shots at these:
- Simple ground-plane antennas using an N connector and a few wires
soldered on are easy to make, and quite predictable. Unfortunately
the wires sticking out in several directions makes them not all
that "go-kit-friendly". I could figure out a design using (e.g.)
banana plugs and jacks to connect the elements, so the antenna
could be easily "unplugged and knocked down" for storage, but that
seems like a lot of fiddling.
Y'er right about that. Worse, the banana jacks and such tend to make
the length of the elements rather variable, which is probably ok for a
simple antenna, but not so great if you want to build something
better, like a double skirted ground plane. I would save the ground
plane for the antenna of last resort.
- Several people have published designs for 33 cm J-poles, often with
a collinear structure for higher gain. Fairly predictable but
need some tuning during construction. Go-kit-friendly if installed
in a fiberglass or PVC radome tube.
A PVC radome will detune the antenna, often in an unpredictable
manner. Also, before you build something inside a plastic or
fiberglass pipe, put a piece of the stuff in the microwave oven and
see if it gets hot. If it does, it's an RF absorber.
I have a bad attitude about J-poles. I'll let someone else proclaim
what a wonderful antenna they can be. I haven't had much luck with
them.
- Simple sleeve dipoles (fold back the coax braid, trim for SWR,
heat-shrink) are easy, cheap, go-kit-friendly, and should be fairly
robust.
The simple coaxial antennas have a common problem. There needs to be
a gap between the outer sleeve, and the coax cable braid. If you look
at commercial coaxial antennas, the sleeve ground diameter is huge
compared to the center mounting pipe. Just peeling back the braid and
burying it under some shrink tubing is kinda marginal.
Yagis would also be useful but aren't as go-kit-friendly due to their
size; Kent Britain's "cheap Yagi" design is easy and the ones I've
made for other bands have worked well.
The problem with mounting vertically polarized Yagi antennas on a pole
is that the coax cable gets in the way of the pattern. If you use a
metal pipe for mounting, that too gets in the way.
So, any other suggestions for simple-yet-effective DIY antennas for
this band?
Of course. I wouldn't have taken pot shots at the other ideas without
having a favorite solution available. It's called an AMOS or Franklin
antenna. Here's some I've done for 1090 MHz:
http://802.11junk.com/jeffl/antennas/AMOS-5-1090MHz/
http://802.11junk.com/jeffl/antennas/AMOS-7/
I build the 2.4Ghz variety inside ABS rain gutter downspout pipe.
Mo
http://www.qsl.net/yu1aw/Misc/vhf_ant.htm
http://www.brest-wireless.net/wiki/materiel:amos
While 900 MHz results in a longer and larger antenna, it's not that
big. The same methods popular with sector antennas on 2.4/5GHz can be
used.
Note the vertical radiation pattern of the 1090 Mhz antennas above.
It's totally horizontal because the antenna is vertically symmetrical.
That means you're not sending as much RF into the ground or into the
air as you would with an end fed antenna, which tends to have uptilt
problems. Since the back of the AMOS antenna is a strip of sheet
metal or wire mesh, the coax cable can be fed behind the strip without
trashing the antenna pattern (as in a yagi or vertical dipole). Since
it's fed with a balun, the coax cable doesn't radiate.
You can also get very creative with the construction and still end up
with a decent antenna. I've made them on 2.4Ghz out of a KD pine 2x4,
some #12 hose wire, and assembled with a staple gun. For shielding, a
strip of aluminum duct tape trimmed to the proper diameter. I'm also
working on one that's made out of copper stained glass tape glued to
the outside of an inflatable vinyl tube. Use your imagination.
Be prepared to have some method of measuring VSWR versus frequency on
900 Mhz. Leave your Bird Wattmeter at home and look into a return
loss bridge, RF sweep generator, DC amp, and a scope:
https://www.google.com/search?q=return+loss+bridge
Or, just build one:
http://www.wb.commufa.jp/ja2djh/html/e_rlb.html
Basically, what an RLB give you is a display of the VSWR versus
frequency without any indication of whether the mismatch is inductive
or capacitive. It's not a VNA (vector network analyzer) but for this
project, you don't need one.
If you want, I can grind out the numbers for most any configuration
you want for 900 MHz.
Are there commercial antennas good and cheap enough that I
should just buy a bunch? Good mobile antennas?
Of course. Most commercial mobile wire antennas cut for the 800-900
Mhz band can be trimmed to 902 Mhz. I have a mess of Motorola NMO
mount 800/900 Mhz antennas that should work. Here's one:
http://802.11junk.com/jeffl/antennas/Misc/slides/Motorola-850mhz-NMO.html
I could probably send you a few if you want them. To use for portable
operation, put some kind of ground plane under it. The downside is
that you'll end up with some uptilt, which is great for talking to
airplanes, but not very good for talking to the horizon.
Low gain fiberglass antennas are usually marked 902-928 MHz. However,
once the gain goes above about 6dBi, they end up cut for specific
frequencies. Since they're sealed, they can't be retuned. Yagi's are
ok, because they can usually be retuned.
You can also find 900 Mhz patch and panel antennas. These do not
handle high power very well, but if you have the 15 watt flavor of
TK-981, it should be ok. The catch is that they're directional, which
might be a good thing. I use one on my Motorola GTX 900 MHz radio in
my palatial office.
http://www.microcom.us/mt262006trhak.html
Note the 9dBi gain. Also note that this one is circular polarized,
which produces an automatic -3dB polarization mismatch loss. CP is
common for RFID tag readers.
Good luck.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060
http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558