Note: This simple transceiver will radiate some signal on the RX frequency. It should be used responsibly where other 80m stations are active locally. It is also highly recommended that a low pass filter is used to reduce harmonics, especially in places like Europe.
A 12 part QRP transceiver
Developed from a brilliant idea by Miguel, PY2OHH to whom I give my thanks. Do visit Miguel's inspirational website for more excellent circuit ideas.
It is crystal controlled on both RX and TX. The xtal controlled regenerative (or is it direct conversion?) RX seems to be moderately sensitive. Unlike my FETer, this rig is full break-in and needs no multi-way TX/RX switching. I have already seen and heard the TX part on the on-line Twente University SDR which is 150 miles away from here in Holland. It has received 80m ham CW stations from all over Western Europe in the first few days of use.
The rig uses just two 2N3904 transistors and a handful of standard parts, so can be built for the price of one bottle of cheap wine. It is just a developing "rat's nest" on a piece of copper right now, but it may get put into a tiny box soon.
I did try adding a small capacitor to ground from the xtal that would be shorted out by the key. The idea was to add some offset between RX and TX, but this was not too successful. I need to try something else to simply give an 800Hz shift.
Since drawing the above schematic, and following a solid 2-way QSO with M0BXT 2 miles away using the uu80b micro-transceiver on 3.560MHz CW, I went across to his shack to measure the performance figures on a decent calibrated signal generator and spectrum analyser. These are the measured results:
Conclusions (so far)
Overall, not bad performance for something this ultra simple. However, it needs more TX/RX offset and needs a low pass filter for use in the UK and Europe. As it stands, the RX is also too insensitive for serious use as it would not hear weaker QRP stations on the same frequency. Nonetheless, it will allow some QSOs over very decent distances to stations >10uV in level. It would be best used to reply to stations who are strong enough to hear. Calling CQ with the rig could be an issue if there are signals too weak to copy that you might TX on top of. I suspect that some attention to the emitter circuit L and C values may improve RX sensitivity a little. There is also some AM breakthrough from strong 75m broadcast stations at times but this has not been a major issue in use.