Build a packet modem from scratch.
Austen Harris VK4TN c/o Gold Coast Amateur Radio Society, Inc,
P.O. Box 588 Southport QLD 4215
PSK packet? What the hell is PSK packet, you might ask? Well, for a sian, PSK stands for Phase Shifl Kc>ing. PSK has cjuiie a lot going for it. especially when you want to run highspeed traffic on a reliable path. 12.00 baud on Hi isn't a problem at all. Faster baud rates are quite possible.
Our first introduction to PSK packet came about when the QDG (Queensland Digital Group) asked us to consider set-ling up a PSK link to VK7 as a message forward station. As VK4W1G BBS had been set up as a club BBS within our club rooms, it was thought that we would be able io drau upon our resources to make the link a going thing.
A subcommittee was formed to investigate the possibiluv ot such a setup, A crystal-locked transceiver was lent l o the club b> Ed VK4JEN for our tests on 14 MHz. A TNC. modified for PSK use, was sent down from QDG for our test.
OK. what have we got 1 we asked ourselves, Not knowing a ihiny about PSK packet we had a lot of questions in ask. The thing that stood out the most was the fact ithat not Loo many others reaJlv understood PSK either We were faced with a sleep learning curve» io say the least! Then along came Doug VK4ZDR. He had been instrumental in setting up the PSK radio links for NASA during the '60s and 46 73 Amateur Radio Today • November 1997
70s. and he is a member of our club. He is also a eo-sysop for our BUS VK4WIG.
Our resources had grown at a great rate. All we had to do was put h together—when it all wont sour. Wc did put the setup together after we we re able lo find out what frequency lo set ihe rig on. Peter VK4XPD gol hold of a crystal that would let lis work the frequency and tune the rig lo it. The commands for the TNC were ftnall> sorted out. and after a couple of weeks we finally connected to VK7BBS. We thouuhl wc had it made. A few more tests were done and then disaster ...the TNC died I
What io do now? wc asked ourselves. We didn't have another TNC modified for PSK use. What do we have to do to modify another TNC? Then we came across the G3RUH design for a PSK modem vised for satellite use in conjunction with a TNC, which turns out In be the basic setup in the TNC1 we had.Acop\ of (he G3RUH modem ^etup was given to all concerned to evaluate whether we could use ii with any other TNC. Our options were diminishing at this stage.
As VK4WIG was running with Baycom-style modems at the lime, we considered that there wasn't a great need to persist with a TNC-based PSK
setup Now we had our work to do, to design a Baycom-style PSK modem that would run into an FBB based BBS along with the other modems already in use.
The basic concept for the design revolved around the G3RUH design, but utilizing the principles as H it were a Baycom modem feeding the computer The original design simply replaced the PCM3105 chip pins as in a Baycom modern setup, A PCB was laid out to include the basic G3RUH design lor his terrestrial u^c and feed into a Baycom board. Initial tests were done on 2m between our two local stations, VK4AOC and VK4TN. A H er we sorted out the filter restrictions thai the rigs imposed upon us, it worked out well.
The next step took a while to finalize. As we had found out quite a while ago.
Photo 1.. Top vie* shows futndy sire of PSK Modem.
Photo 1.. Top vie* shows futndy sire of PSK Modem.
Fig. I ♦ Schematic diagram of the Baycom-style Hi- PSK Modem.
the original Baycorn modem fed the computer via a 74HC04 chip, which meani a psetido RS-232 level was presented to the serial port. This caused problems willi reliable operation. We are at present driving our serial port via MAX232 chips which delher a standard RS-232 level — no more problems!
The next problem to overcome was in provide a stable 1500 Hz TT1. level signal to feed the TX modulator a* described in G3RUH s article. He had suggested that a frequency somewhere between 1400 \\i and 1600 Hz would do the job. Our selection seemed lo work out quite well, but il drilled. That problem was solved by the introduction ol a low-drift capacitor in the circuit. A ,01 \iF J polyester with a 120 ppm/deg C was our choice. All our tests on the oscillator have shown i\ high decree ol stability, running weeks ai a
We also required a timeout circuit to he incorporated with the design and ihe two spare inverters within ihe MAX232 were pressed into operation for that. *I X and RX LEDs were also required. The TX LED wasn't a pmhlem ui all. bui the RX LED required a bit of thought. It finally turned out to be the "Lock" indicator l.ED in G3RUH's setup.
A prototype PCB was laid out and built. Tests carried out provided instant success with connections to VK7, VK2, VK5 and VK4 stations.
The circuit can he broken down into five parts consisting of:
1. The TX modulator
2. The RX demodulator
3. The timeout and PTT circuit
4. The RX LED circuit
5. The power supply.
73 Amateur Radio Today * November 1997 47
IC RAIL BYPASS cAPACrroRS
Was this article helpful?