Saturday, July 21, 2012

Soundcard Interface for APRS/Packet - Part 2

Well, I finally got the PTT part of the soundcard interface working.  This was fun, and sometimes frustrating, but very educational.  I am lacking in my understanding of component level electronics engineering.  As a network engineer, I would be comfortable designing and implementing an enterprise network for a new hospital (that's a different story), but understanding why a transistor works the way it does is beyond me at this point.  So I kind of fumbled my way through this.  One thing that is very important in network engineering is attention to detail, and solid planning.  For some reason I didn't carry this over to this project, and it bit

Let me state, for the record, plan your work, and work your plan.  Don't just wing it.

Okay, on that note, on to the project.  Once again, was a great source of information on getting this part working.  I used this schematic from their site (click on the image to see more details):

R1 = Resistor, e.g. 1K2, to reduce voltage on the IC pin 1
IC =  Integrated Circuit; this sketch shows an IC, such as a 4N33
        or PS2603 Optocoupler, which uses a Darlington pair transistor.
        (Note: to identify pin #1, look for a small embossed circle on
         the top of the IC above pin 1; or looking into the notch in one
         side of the IC with the pins down, pin #1 is to the right of the notch.)
D1 = Diode, e.g. IN4001, would shunt any potential reverse voltage
        that might damage the sensitive diode/emitter in the IC.
        (Note: the band printed on the diode marks the cathode end,
        which attaches to the Serial Port/IC Pin 1 line in the sketch above.
        The opposite/anode end attaches to Ground.)

In the last post about this project, I built the receive and transmit cables that connect the output of the radio to the input on the soundcard, and the output on the soundcard to the input on the radio.  This worked well for listening to APRS and packet conversations, but I couldn't join in without manually keying the mic, which works okay if you're doing PSK31 or other HF digital modes, but isn't as easy with packet or APRS.  With this addition, a signal is sent via the PC's comm port to key the PTT circuit on the radio.  This allows the radio to be keyed via software instead of manually.  This allows things like APRS beacons to work normally.

This particular design uses an optocoupler to isolate the radio and the PC, while still allowing communications to flow.  This, along with the audio transformers used in the last post, helps to eliminate ground loops and voltage differentials, which are a source of hum in your transmission audio.

While building my version, I found that I didn't have the required 1.2K Ohm resistor, so I substituted a 1K and a 220 Ohm resistor in series.  This took a little additional space, which was at a premium in the tiny project box that I used, but ended up working well.  As you'll recall, in the last post, I decided to build this in a project box, instead of as a set of cables.  I think that in the long run my way will be more durable since there will be less movement of the internal solder joints due to cables wiggling around.  Time will tell.

A couple of issues that I ran into:

  1. On Windows XP (need to upgrade my shack PC), the serial mouse driver keeps the RTS pin HIGH, which causes the mic to key if nothing has taken ownership of the comm port that you are using (AGWPE, Digipan, etc) to bring it LOW.
  2. The pinout of the IC that I used was different than the one pictured in the schematic, don't blindly trust any schematic, do your homework first.  It saves solder (and frustration)
  3. Don't use too small of a project box, it is easier to do things if you have the room to do them in.

A couple of lessons that I learned:
  1. When if doubt, check solder joints.
  2. Follow the voltage, step through the circuit with a multimeter and make sure that you are seeing what you expect to see.
  3. Amperage kills...  This applies to ICs and other components as well as people.
  4. Be sure that the part is installed correctly BEFORE soldering...  

My Parts List:
  • 1 x 4N33 Optoisolator
  • 1 x 1K Ohm resistor
  • 1 x 220 Ohm resistor
  • 1 x IN4001 diode
  • 2 x 1/8" mono audio jacks
  • 1 x 6-pin mic connector to interface with my TR-7730
Time to Completion:
  • This part took me about eight hours due to lack of understanding and being too trigger happy with the soldering gun.  Next time should go smoother as hopefully I have learned my lesson on this.
Here are a couple of pics:

The IC and diode on the bottom right, and the two resistors on the bottom are all that make up this part of this circuit.  The other two resistors and the two transformers are for the rx and tx circuits.  You can also see the two audio jacks that I used to connect the PC and Radio.

This shows the completed and closed project box.  Again, while I at first thought that smaller would be better, after building it, I should have used a slightly larger box.  The next one that I build will be a bit bigger.

These are the cables that I built to connect the computer to the project box, and then on to the radio.  The gray DB9 connector attaches to the comm port for RTS based PTT.  The other end of that cable is an 1/8" mono connector that attaches to the project box.  The 6-pin cable actually has two pig-tales, one 1/8" mono for audio, and another 1/8" mono for the PTT signal.

And here is a shot of the APRS terminal showing me sending and receiving traffic.  I need a better antenna to hit more than one or two stations directly, but thankfully, one of the stations that I can hit is a digipeater/IGate.

All in all this project was a lot of fun and extremely educational.  It is also a LOT cheaper than buying another SignaLink.  I think that I have around $20 invested in this, plus my time.


Richard, KK4JDO

No comments:

Post a Comment