I use 12 cores ( 2 cables 'liberated' when they re wired LANs at work). This
gives 5 data lines, plus return, for selection of up to 32 antennas, a pair
of wires for turning on the remote power relay (these get the polarity
reversed on tx), a pair of wires for controlling tuner motors, and a pair
for an indicating meter for the tuner position.
I would be chary about trying to do it with RF: I have enough problems with
wires. To avoid galvanic connection, I isolate things with opto isolators
where possible, and use ferrite chokes on all the leads in and out:
nevertheless, it's not completely EMC proof. I found it is well worthwhile
using the reversed polarity power up on tx to inhibit the switching of all
but the tuning motors, and they're relay controlled. I'd also avoid trying
to mix data signals returns and power returns. The common impedance causes
trouble. Been there, done that, while trying to use the coax outer has led
to greater difficulties in keeping the RF out. Some bands can be worse than
others, too, depending on strange resonances in the leads.
If there isn't power available at the remote point, you'll also need to run
some current down the lines, too. That can be a problem because of voltage
drop unless you use fairly substantial cables. Bias Tees are all very well,
but for higher powers (above 100 watts), you need substantial capacitors to
carry the RF current.
One point I would make as the result of experience with remote antenna
switching: build in plenty of test features, so that at the remote point,
you can power it up and give it 'local instruction' as to function. Local
metering is useful, and lots of LEDs to indicate the status of various
switches etc. To allow flexibility, any decoding of the input signal for
antenna selection should either be in software or firmware - I use 2716
EEPROM, but that's because they were being thrown out at the firm where the
XYL worked at the time...........
The biggest problem may be finding a suitable multiway switch at HF, rather
than a relay matrix. I use a 'standard' rotary switch wafer, about 1-3/4
inches across (the NSF style HC) with paralleled front and back contacts.
This sits very close to a ground plane on a PCB, and 50 ohm strip lines made
from strips of PCB material of the correct width are glued to the base
ground plane. This gives a pretty good SWR up through 10 metres. The switch
wafer itself is driven by a Ledex. These aren't easy to find in surplus or
at hamfests, though. There is another approach, which is to use the usual
switch index mechanism, a motor and a loose clutch to the push the switch
round until it clicks. A suitable oriented microswitch can be used for
depowering the motor, and slave wafer for positioning.
Ian can, of course, come over and examine.........
73
Peter G3RZP
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|