Re: A Home Brew 224 MHz Repeater Project.
> n...@... wrote:
> Thanks for posting this, Skipp. It's always fun to see
> another angle on repeater building. Threaded are my $0.02:
Thank you Bob, it would be great to see other examples of how
people build repeaters and why they use certain construction
techniques and equipment.
In this example, someone requested a repeater be quickly placed
in service before a sanction application deadline passed, not
knowing the final repeater package had not even been constructed.
Many of us build repeaters using different techniques based
on what is supplied or available at the time of construction.
I thought about a surplus commercial radio conversion... but
the delay in having crystals cut made the frequency synthesized
Hamtronics RF Decks much more attractive.
> Any idea what Hamtronics chose for the loop bandwidth of the
> PLL (I assume it's not a direct DDS synth.)? If very low, it
> could have very good phase noise characteristics. Not that
> that's important on 220, but could be good for a 2 meter
> version @ 15 kHz spacing.
A very good question with a non-technical reply. Rather than
get into the gritty details of the PLL operation before the
purchase... I will say that I jumped in and bought the Hamtronics
modules on trusting faith from my decades of using Jerry's
(of Hamtronics) kits and pre-made units.
I've seen a relatively small number of reported horror stories
related to the synthesized Hamtronics transmitter and receiver
strips (module). The reported problems I saw posted were related
to PLL Stability in relation to mechanical and physical chassis
movement.
I'm fairly sure the Hamtronics PLL circuits have been modified
(tweaked) from the original design, but where, when and what I
have not compared in detail.
I was not disappointed with choosing the Hamtronics Equipment,
the modules seem to function very well. The PLL performance
is noticeable in very detailed measurements and the end result
is nothing bad anyone is going to be able to hear or detect in
normal operation. Nothing objectionable or unexpected in regards
to the previously mentioned PLL mechanical stability issue reared
its ugly head.
The R302-6 Receiver Manual with diagrams is available for free
download at the Hamtronics web site. I had concerns about the
receivers front end performance based on the number of (or lack
of) tune circuits helical resonators. I placed the repeater on
a site antenna combiner system and it's working very, very well.
The receiver is red hot and the repeaters performance is much
better than expected.
> > New old stock ("NOS") Com Spec TS-32 CTCSS (PL) tone boards
> > are very nice "Ebay sleeper deals".
> Yeah, but I don't like them. They do decode reliably, but
> the encoders have harmonics that need to be externally filtered
> if fed to a phase modulator.
I had not ever noticed the harmonics and related them to a
problem... but I'll fire one up and check it out. Easy enough
to deal with using a low pass filter. Some exciter (transmitter)
circuits run external sourced CTCSS through a basic filter,
but not all...
> They also don't respond to reverse burst. Again, not a big
> deal if we're talking 220 & a direct FM TX.
While reverse burst and muted CTCSS during the repeater tail/hang
time is a wonderful thing... I took the simple and less time
consuming path of wiring the constant on CTCSS encoder direct
to the transmitter strip (module). CTCSS is on (encoding) anytime
the transmitter is up (on the air) and that for me is just fine
versus not having any Tx CTCSS at all. I felt it's better to
at least provide for CTCSS enabled receiver operation.
One option would be to shunt mute the CTCSS line based on the
receiver COR/COS or CTCSS detect line... but that might be a
future project if the need was really there (required).
> My favorite is the Sigtone C1116. Unfortunately they've
> been out of production for a few years, but fortunately for
> me I made a bulk purchase at a local swap meet many years
> ago & still have a couple of them left today. They decode &
> release faster than the TS-32 & have a better HPF
> (see below).
Now you let the cat out of the bag... we'll all be sharking
Sigtone equipment at the flea markets and of course Dayton
next week. Of course the TS-32 is out of production but any
similar device would fit the bill. I believe the TS-64 or a
similar animal is still being made if someone wanted to go
retail (buy one new).
> The problem with the TS-32 tone filter is that it has poor
> transient response, "ringing" around 400 Hz. To me, a system
> that uses one in the audio path sounds "boxy".
I have seen the above posted comment before, but have not
experienced it myself. I will say the filter performance does
change with source and termination impedance but again for
me using standard techniques I've not experience any ringing
issues and "so far" the repeaters overall audio reports are
great.
> What's worse is that if more than one is used in
> a linked system, the audio at the other end will quickly
> degrade as that 400 Hz region begins to dominate the
> frequency response.
Once again I have not experienced that problem but I will
also say I use notch (versus low pass) filters in linked
systems.
> > Don't panic, the TS-32 also provides a separate tone
> > generation "encoder" section for your transmitter CTCSS
> > requirement. A shielded audio quality wire is routed out
> > of the receiver box to the transmitter board at the
> > proper CTCSS connection point.
> I assume this is done using feed-through capacitors - you
> don't want to plumb any wires straight through the case.
In the past I've actually routed the shielded CTCSS endode
audio source wire through chassis and cabinet/box holes
and in this case I used the feed through-capacitor method.
The holes in the direct wired version was already there so
I used it. Through hole direct wiring can make servicing
by box swapping a bit more difficult. In very short wire
lengths I have not experienced a problem with stray RF or
Ground Loops bringing a gremlin on board.
> >The Hamtronics Receiver COR/COS output is active high, which I
> >don't like one bit.
>
> When I first started building repeaters, I used active low
> CTCSS as well. The main reason was that it was convenient
> to use as a cheap way to key the TX when a controller wasn't
> available back in those "lean" college days. When I started
> using G.E. radios with their active high CAS & RUS outputs,
> I switched to active high as my standard.
I like active low logic for a number of reasons and personal
preference. In a situation where the controlling device loses
power, there is a potential for the transmitter to key up.
It has happened to myself and others during equipment failures
and service events. If a repeater controller locks up (fails),
loses power or similar... I wouldn't normally then have an
uncontrolled signal on the air. All in an effort not to become
famous...
> > So I simply routed the receivers COS/COR Output logic
> > line through a 120 ohm resistor to the gate of a 2N7000
> > FET
> That works. If you don't have a 2N7000 available, any
> decent NPN BJT (2N2222, 2N3904) will do too - just use a
> higher value resistor on the base like 4.7k. In fact,
> you can use a bigger resistance on the 2N7000 as well
> since the gate is high impedance. I mention this because
> around here BJTs are easily obtained whereas I had to
> order 2N7000s via mail.
There are actually cases (in very high RF environments)
where using a bipolar transistor makes much more sense
(easier to reliably maintain safe/positive control of...)
... and I normally would not route a control wire to the
gate of a 2N7000 without a few more parts parked in place
at the fet gate. But the Hamtronics receiver COR/COS logic
line deals fairly well with the fet gate switching tasks
and protection I need to ensure happen. So the 120 ohm
series gate resistor is enough in this specific example.
If anyone wanted to talk about properly protecting the
2n7000 gate lead in a more generic industrial application...
it's a topic we can visit as time allows.
The distance from the receiver pc board to the fet gate
is only a few inches inside a normally well shielded box.
> >DB-9 Connectors are very popular with repeater builder types. I
> >for personal preference and experience tend to move away from
> >using them in for this type of repeater chassis through hole
> >connections. I have returned to using through hold feed-through
> >capacitors
>
> So what do you use for a connector? In my latest RX boxing
> project I decided to put a DB9 "doghouse" on the box. The
> doghouse is a cheap plastic box since shielding isn't necessary.
> I plan to install the feedthroughs in a manner similar to what
> you describe, but then mount a DB9 & Anderson PowerPole on the
> doghouse so all the connections are "connectorized". I may
> even add some switching circuitry inside the doghouse so this
> receiver can be grafted into a existing system using a
> single-port controller (2 RXs on one port).
For this project... I only used the feed-through capacitors
for everything. The wires at the rear of the controller get
into the controller via a DB-9 plug but nothing on the chassis
or RF deck. I stopped using small Anderson Power Pole connectors
because of all the grief they caused me on the commercial
radio side of my life. A lot of people like and use them but
I don't trust or use them anymore after a few 10 hour days
sourced back to intermittent small power-pole connectors.
> >and while the value is not ultra critical, you don't
> >want the capacitance value large enough to impact the information
> >passing through. I found and used surplus 100pf (pico farad)
> >feed-through capacitors although I'm sure higher values will work.
> To make sure that the shunt capacitance blocked any RF, I
> added some series resistance to form an RC. On lines like
> the RX audio, adding 470 ohms in series with a source that
> was already several k-ohms didn't affect the audio
> at all. Same for COS & CTCSS decode logic outputs if the
> controller's input impedance is much higher than the series
> resistance used. Also serves to protect those outputs if
> you accidentally short them to 12 V or ground.
Unless I'm sure there's going to be a lot of shear RF at the
repeaters location... I take the simple and easy route. Even
with a broadcast station nearby (but not in the same value)
I didn't feel the advanced filtering techniques were required
and so far, so good.
> >In my opinion the 100pf value is enough to bypass troublesome stray
> >RF, but not greatly impact or modify the audio information passing
> >through to the external controller.
> However, it might not be enough shunt C for 2 & 6 meter RF.
> At 2 meters, the reactance is 11 ohms. Who knows what the
> impedance of the hanging wire is at RF? But add the 470 ohm
> series R with the 100 pF & you get over 32
> dB of attenuation.
> Bob NO6B
Not a bad idea if it's an easy implementation... I chose the
100pf feed-through capacitor value because I had them handy
and cheap. In a 1981 constructed Hamtronics 6 meter repeater
I still have working... I used 220pf disc caps because I couldn't
easily find feed-through. Later on I remember saving and removing
the feed-through plates from GE Master Pro receivers and
transmitter strips.
Being creative is very good thing...
cheers,
skipp
