After getting my ham radio license late last year and not having much luck operating from my dorm room, I decided that I might still be able to do some shortwave work from here.
While listening to 4Z1UG’s QSO Today podcast with Murray Greenman, ZL1BPU, Murray mentioned how QRSS MEPTs - manned experimental propagation transmitters - can be run successfully, even with subpar antenna setups or where it is outright impossible to put up big HF antennas.
I did some research and eventually came across the Knights of QRSS website, which has some good starting info on what QRSS is and how to get started.
The idea of QRSS is to use a low power transmitter (usually less than 200mW) to transmit your callsign in very slow morse code (QRS in a CW QSO means send more slowly, hence QRSS meaning very slow). There are other ham radio operators around the world running receivers (so called “grabbers”) and software to generate a waterfall image. Those images are then uploaded to the web for other operators to inspect and see if they can find their callsign. Because the location of transmitter and receiver are known, propagation conditions can be examined.
The optimal length of one unit or “dit” is deemed to be 6 seconds, which allows the callsign to be deciphered from the waterfall display, even in noisy HF conditions.
After learning a bit about QRSS I set out to build such a transmitter myself. I opted for FSK-CW, a mode where the Morse characters are not sent using On-Off keying (OOK) but by slightly shifting the carrier frequency (only 4 to 6 Hz). There are other modulation schemes in use as well, inclduing “standard” slow Morse (OOK) and DFCW, where dits and dahs are the same length but denoted by two different frequencies.
The transmitter that I have built up on the bench over the weekend consists of the following components:
- ATtiny2313 as the microcontroller in charge of timing and DDS frequency control
- AD9850 DDS module to generate the carrier
- SM0VPO 500mW class-B amplifier followed by a 5th order Butterworth LPF
Transmitter output power was originally 400mW into 50 Ohms, apparently already considered high power by QRSS standards. It has been suggested to me to reduce that output power to 200mW or less, which should still yield satisfying results.
I have reduced the drive from the DDS by placing a 47 Ohm resistor from the DDS output to ground right before the AC coupling capacitor into the first stage. Final transmitter output power is about 150mW.
The transmitter will feed a magnetic loop antenna (the only feasible option in my situation). Eventually I will put this transmitter into a diecast aluminium box (Hammond 1550Z106), once I have the output power adjusted.