[OZAPRS] Cavity info

[Ray Wells]

Tony,

How do you plan to implement this;
1. all systems sharing a common antenna
2. separate antennas.
3. what frequencies are used for the existing system. Wider frequency 
spacing relaxes filtering requirements.
4  is the the existing repeater duplexed or using separate antennas
5. what physical spacing between antennas
6. what protection will the existing cavities provide both to and from 
the aprs system.It will provide some.

A typical 4" cavity would be approximately 21" long, with the centre 
element being about 1" in diameter and adjustable in length to be a 
quarter wavelength at resonance. Those dimension are from an old TCA 
cavity. Typically, only the end section of the centre rod is adjustable, 
with finger stock being used to ensure reliable contact with the fixed 
portion. Being that its near the (high Z) end of the centre rod, 
currents are not high. The size of the coupling loops determines such 
things as insertion loss (and hence Q), and operating bandwidth. RFS 
cavities generally have rotateable coupling loops calibrated from 0.5dB 
to 2db insertion loss.

Surface conductivity is paramount for reliable performance, particularly 
the connection between the end plate and the outer tube, with silver 
plating being commonly used on mating surfaces. Temperature stability of 
the central tuning rod is also crucial. Commercial designs usually use 
Invar because of its low temperature coefficient. The importance of this 
would be mandated by temperature stability in the hut. The RFS cavities, 
both 4" and 6" are made from aluminium tubing and the aluminium end 
plate is pressed into the tube.

Many (insert more manys!) years ago AWA were producing cavities for at 
least 70-85 and 148-174MHz from copper. They were about 5" in diameter 
and were a slightly shortened design (with lower Q). They were tuned by 
capacitance. The main portion of the centre rod was fixed in length and 
a short adjustable length was fitted close to the open end of that rod. 
I never saw any performance figures on those cavities but one would 
reasonably expect performance to be worse than for a "conventional" design.

Somewhere in my book collection I have an old ARRL publication, FM and 
Repeaters. IIRC there is a design in that book. If you can give me a day 
or two I'll find the book and copy some pages for you. I'll also take 
some pics of the innards of a TCA cavity.

Depending on what's already on site, you may get away with something as 
simple as a single bandpass cavity, or a cavity hanging off a Tee as a 
band reject filter, in the antenna lead of the aprs system. Antenna 
isolation can be a useful tool with the isolation between two dipoles, 
one mounted directly above the other and with their ends almost 
touching, being about 30dB. Greater spacing increases that figure. Such 
things are readily measured with just a receiver and a calibrated signal 
generator. Measure the rx sensitivty. Then connect the rx to one antenna 
and the sig gen to the other. The isolation is the difference in rx 
sensitivity. With 600khz frequency separation approximately 100dB of 
isolation is required between tx and rx to avoid desense issues (depends 
of the tx and rx used). The greater the frequency separation the lower 
the isolation that is required. Isolation can come from a filter alone, 
or filters and antenna separation.

Ray vk2tv



Tony King wrote:
> Hi All,
> I am currently in the process of finalising an APRS repeater for 
> coexistence with the Yarra Valley Amateur radio groups 2m repeater and 
> am at the point where I need to machine up a cavity or two.
>
> Can anyone point me to a suitable amateur design for a 2m unit as ex 
> commercial ones are a bit thin on the ground here at the moment.
>
> Actually any design formulas would also be appreciated then I could 
> design my own.
>                                                                                         regards Tony VK3API
> _______________________________________________
> OZAPRS mailing list
> OZAPRS at aprs.net.au
> http://lists.aprs.net.au/mailman/listinfo/ozaprs
>
>