[aprssig] CNSP-18 now 30m HF APRS transmitter

Chris Moulding chrism at crosscountrywireless.net
Wed Dec 5 03:23:14 EST 2012

About 4 years ago I made a 10W transmitter for 30m with AX.25 frequency 
shift keying of a crystal oscillator.

It worked really well.

I'd forgotten about it until I spotted your posts. I'll take look at it 
again. It would be relatively easy to add it to an APRS TNC Digi Tracker 
to make a self-contained HF APRS tracker with it's own GPS receiver.


Chris, G4HYG

Chris Moulding
Cross Country Wireless
7 Thirlmere Grove, Bolton, Lancs, BL4 0QB, UK
Tel/fax:     +44(0)1204 410626
Mobile:      +44(0)7752 391908
Website:     http://www.crosscountrywireless.net
Yahoo group: http://uk.groups.yahoo.com/group/cross_country_wireless/

On 05/12/2012 06:21, Stephen H. Smith wrote:
> On 12/5/2012 12:12 AM, Bob Poortinga wrote:
>> Lynn KJ4ERJ writes:
>>> If anyone knows of a cheap SSB HF 30m transmitter, or better yet
>>> transceiver, I'm interested!
>> I wonder if a Norcal 40A can be adapted to 30M.
>> <http://www.fix.net/~jparker/wilderness/nc40a.htm>
>> 73 de
> I've just been carrying on an off-list discussion on this very issue.
> After playing around with the various APRS Messenger digimodes on 30
> meters, I think there is a way to do this easily.  You don't need a
> full-blown 30M SSB transmitter with a separate TNC converting data to
> audio tones.
> Classic AX.25 packet, GMSK and MFSK16 are all single-tone-at-a-time FSK
> modes. Two throughts I had were to either have a DSP chip directly
> generate the RF signal at 10 MHz, given that DSPs can operate on signals
> at 50 MHz and beyond.
> These three modes are constant-power FSK (i.e. like key-down CW) between
> two RF frequencies 200 Hz apart in the case of FSK, several tones in the
> case of GMSK, or between 16 discrete RF freqs at only a 2-3 Hz rate over
> a spread of about 500 Hz in the case of MFSK16.
> Since they are not a complex two-tone simultaneous signal like PSK63:
> a) you aren't concerned with the shape of the waveform, just it's
> frequency.  It shouldn't be much of a challenge for a DSP to directly
> synthesize a simple sine wave (or reasonable approximation thereof) at
> 10 Mhz.
> b) with a single tone at a time involved, there is no issue with
> amplifier linearity or intermodulation distortion. You stick a couple of
> MOSFET power amp stages directly onto the output pin of the DSP to jack
> the output up to half a watt or a watt or so. Saturate them and drive
> them into highly-efficient class-C operation.
> After my experiences on multiple cross-country road trips running both
> classic AX.25 300-baud FSK and the APRS Messenger PSK63-GMSK-MFSK16
> modes, I would vote for MFSK16.   Any of the digimodes is vastly
> superior to classic 300-baud FSK for weak signal work, especially in the
> presence of multipath and noise, thanks to their low symbol rates, but
> MFSK16 is waaay beyond any of the others in the signal-to-noise ratio
> required to work successfully.  It will provide solid copy on signals so
> weak, you can't even hear them in the speaker.
> MKSK16 has about a 10 dB advantage over PSK63 in t;erms of the minimum
> SNR required.  You can cut the transmit power in half (losing 3 dB on
> TX) and still be 5-6 dB ahead at the RX end.  And the TX can use a
> saturated efficient class-C PA which PSK63 can't!
> An alternative approsch we discussed would be to hang a varactor diode
> across a simple crystal oscillator running directly at 10.149.70 MHz.
> I.e the basic crystal-controlled "rockhound"-type 1-watt QRP CW
> transmitter.
> A single-chip low-end microcontroller could easily output a 0/5 volt DC
> logic-level waveform to apply across the varactor (rather than audio
> tones) to directly FSK the crystal on classic AX.25. Or an output
> voltage at 16 stepped values could be generated by the on-chip a-to-d to
> modulate the varactor for MFSK16 operation. One could even use a table
> to generate 16 UNevenly spaced voltages to compensate for the
> non-linearity of voltage vs freq in the varactor/crystal system.   Such
> a device would have the cheap, essentially-disposable, minimalism of a
> TinyTrack III.
> The microcontroller and QRP crystal transmitter could be combined on one
> small board to yield the 30-meter equivalent of a MicroTrack AIO.
> ------------------------------------------------------------------------
> --
> Stephen H. Smith    wa8lmf (at) aol.com
> Skype:        WA8LMF
> Home Page: http://wa8lmf.net
> High Performance Sound Systems for Soundcard Apps
> http://wa8lmf.net/ham/imic.htm
> http://wa8lmf.net/ham/uca202.htm
> Vista & Win7 Install Issues for UI-View and Precision Mapping
> http://wa8lmf.net/aprs/UIview_Notes.htm#VistaWin7
> "APRS 101"  Explanation of APRS Path Selection & Digipeating
> http://wa8lmf.net/DigiPaths
> .
> _______________________________________________
> aprssig mailing list
> aprssig at tapr.org
> https://www.tapr.org/cgi-bin/mailman/listinfo/aprssig

More information about the aprssig mailing list