Cable lengths sometimes are critical and sometimes not, depending what you
are trying to accomplish with the particular setup. In the case of
interconnecting cables between notch cavities a critical length cable will
provide more notch depth because it unloads the next cavity.

In the case of using pass band cavities to make a combiner into a common
junction the cable lengths are critical even though everything is running at
50 ohms. The off frequency skirt of the cavity provides a short at some
frequency and that short needs to be transformed to a high impedance at the
junction so as not to load the other devices.

 

DUPLEXER TUNING: 

In this case proper duplexer tuning should make little if any difference in
the interference problem noted. As long as there is no desense from his own
transmitter the duplexer is doing what it can. A bp-br duplexer has little
off frequency rejection (pass band rejection) compared to a pass cavity. The
bp-br duplexer does provide a little pass band help but not as much as some
people like to believe. The frequencies between the transmit and receive
frequencies have the most rejection from pass band effects but outside of
them there is little. 

As was stated he is getting interference from a public safety transmitter so
it would fall outside the more protected part of the duplexer. An additional
pass filter on the receiver would indeed provide more protection but 45 db
sounds like a lot to expect from a pass filter unless the frequency is a
long ways away. As Skip noted a notch cavity at that point may well provide
more protection.

 

More to the point of cable length, a pass filter actually looks like a notch
filter at the unwanted frequency i.e. a low impedance at that frequency. So
ideally if you want the most rejection a cable that provides a half wave
length at the unwanted frequency will reflect that low impedance provided by
the cavity skirt to the next port in the system at that frequency.

In this case the difference will probably not be noticed for the trouble
involved.

 

With the public safety antenna right beside the repeater antenna there can
be fundamental overload of the receiver so some form of additional
protection may help.

 

73

Gary  K4FMX

 

  _____  

From: [email protected]
[mailto:[EMAIL PROTECTED] On Behalf Of Ken Arck
Sent: Thursday, July 26, 2007 9:40 AM
To: [email protected]
Subject: Re: [Repeater-Builder] Coax length between added cavity and
duplexer

 

Hi Nate

Firstly, in the case of your additional bandpass cavity - if cavities are
properly tuned to 50 ohms, the length of 50 ohm coax between them doesn't
make one bit of difference. The problem most people have with proper
cavity/duplexer tuning is that they don't maintain a 50 ohm load on ALL
ports when they tune 'em. So when they're placed into service, the port
impedances are different and the tuning of the cavity/duplexer changes.
Which is why you should NEVER EVER tune either without at least a 3 dB 50
ohm pad on each of the ports. From your description, I'm willing to bet you
didn't use pads :-)

My other comment (and in my humble opinion) is that if you're using a
properly tuned BpBr duplexer on a low power repeater (the 720 is 25 watts,
yes?) and you need additional cavities in the receive side, you've got
bigger issue than simply needing an additional cavity. You didn't specify
what the inteference is but have you done an intermod study of the site? 

IMHO, the additional cavity is a waste of time and effort until you identify
the source of the interference (and made sure your duplexer is first
properly tuned)

Ken


At 05:09 AM 7/26/2007, you wrote:




I'm helping a club in the next county get their repeater working
better. A couple of weeks ago they brought the RF unit (TKR-720) over
and we (KC0MLS, K0BYK, and myself) checked it out. The PA transistor
required soldering and after that everything checked out well.

Next we checked out the duplexer, a Wacom BpBr set. Lacking a tracking
generator, we used our ancient IFR-1200 and a reprogrammed Spectra
mobile radio and tuned the pass filters for best SINAD and the notch
filters for the poorest SINAD for their respective frequencies.

After they put everything back on site, it all works well except that
the local public safety is getting into the receiver intermintently. 
My first thought was intermod, but the various programs don't turn up a
match for the receiver's frequency.

A week ago we were able to visit the site and tightened several loose
connectors on the other hardware in the site. Since then the
interference does seem to be less but is still present on occasion.

Observations of the site revealed that the public safety and the club's
repeater antennas (DB-224 style, unsure of exact models) both share the
top of the tower and are broadside to each other and are maybe four
feet apart at most. So now our thinking is that the problem may be
receiver overload. 

We set up a spare Celwave bandpass cavity that has about 2 dB of
insertion loss and offers about 45 dB of insertion loss at the public
safety's transmitter frequency. My question is whether the coax length
is critical between the RX port of the Wacom duplexer and the input
port of the Celwave cavity? I plan to send along a length of RG-393
(double shielded teflon coax) with the cavity. As far as I know, it is
a random length. Should I cut it to something closer to 1/2
wavelength? 3/4 WL?

Thanks!

73, de Nate >>

-- 
Wireless | Amateur Radio Station N0NB | Successfully Microsoft
Amateur radio exams; ham radio; Linux info @ | free since January 1998.
http://www.qsl.net/n0nb/ | "Debian, the choice of
My Kawasaki KZ-650 SR @ | a GNU generation!"
http://www.networksplus.net/n0nb/ | http://www.debian.org
<http://www.debian.org/> 

 


----------------------------------------------------------------------------
--
President and CTO - Arcom Communications
Makers of repeater controllers and accessories.
 <http://www.arcomcontrollers.com/> http://www.arcomcontrollers.com/
Authorized Dealers for Kenwood and Telewave and
we offer complete repeater packages!
AH6LE/R - IRLP Node 3000
http://www.irlp.net <http://www.irlp.net/>  
"We don't just make 'em. We use 'em!"

Reply via email to