Over the years I have built a lot of gear from simple test equipment to an eight band CW transceiver using reclaimed parts. Technology moves on and I now have three commercial SDR rigs. The performance of modern transceivers is amazing but I feel something has been lost; it’s all done in software which feels a bit cold, clinical even.
A few years back I followed a series of articles in RadCom written by Eamon Skelton, EI9GQ, where he described how he built a transceiver mainly from discrete components. I bought the book a couple of years ago and now feel I want to make my version of his project.
Why now? It is mainly the availability of cheap test equipment like spectrum analysers, VNAs, GPS locked frequency meters and wide band noise sources which work well with SDR receivers like the RSP1A.
Not long ago all we had were grid dip oscillators, drifting sig gens and frequency counters with dubious accuracy.
I have made a start and almost fallen at the first hurdle! Instead of making a 9Mhz, 2.4Khz wide SSB filter I have bought one from Spectrum Communications. That will save buying lots of 9Mhz crystals and spending hours matching them to get a good shape.
I have already built a VFO based on a synthesiser chip. There are some things it is not worth trying to replicate and a drifting VFO is top of the list. I did build a free running 8 Mhz VFO in the early 1970s that was multiplied by 18 to drive an old Pye valve TX on 2M. It took hours to sort out the drift but eventually it was rock solid. The key was an Oxley Tempatrimmer. The fixed vanes of the capacitor were fixed to a bimetalic strip so as as the temperature varied the capacitance changed. They are impossible to find these days.
Watch this space as I will post progress reports as each stage is completed. The first thing to do is tidy up the workshop, I may be some time…