After having finished my last ssb transceiver project I was to a certain degree dissatisfied with the mechanical size of the rig. Even though it performed very well and I did some nice DX with it (including cracking a pile up for a Jordanian station) I wanted to find out, if I could shrink the mechanical dimensions of the device to another significant degree. A pocket-size transceiver with somewhat around 5 watts PEP output was the goal that I wanted to achieve.
First I searched the web for attractive circuitry. I quickly came across the „Kajman Transceiver“ made by SQ7JHM, OM Jurka (Link to schematic). At the first glance the circuit seemed extraordinary simple. But the problem was: The device has been designed for operation on the 80-meter-band. So some extensions and improvements had to be made to make it perform well on 20.
First, the original recveiver seemed to be a problem. The layout is very simple. Above all, there is no rf preamp. This might be OK for the high noise levels on 80 and the relatively high field strength that signals have got there. But for 14 MHz this performed, as I had expected, very poorly as I could easily find out when breadboarding the receiver strip and having it run with a simple VFO. And, in addition, the receiver did not lack only sensitivity, the noise figure was also very poor.
So there were two possibilities to cure this problem: add an rf preamp to the receivers’s front end (to improve noise figure) and maybe add an if amp stage between the two NE612-mixers to improve overall gain.
After having put a dual-gate mosfet rf preamp stage to the receiver it suddenly opened up its ears. Performence now was great compared to that before the modification had been made. With a simple dipole I could listen to VKs ans ZLs in the morning and to state side (even West Coast) in the evening. Being very satisfied with the outcome of this simple measure and strictly sticking to simplicity of cicuitry and limited space I went away from adding an if amplifier. Maybe I’ll give it a try later.
The transmitter had also been modified after a short intensive thinking. I re-constructed my proven 3 stage design with single-ended preamp and driver and two 2SC1306s in push-pull mode as the final. That should deliver 5 watts easily. And, not to drive your tension too high, it finally did.
As Collpits, Hartley and Co. VFO-times are over, I took another AD9835 DDS chip out of my shelf, grabbed into the microcontroller box and an AT Mega 8 microcontroller found its way into the circuit. To have the TTSOP-16-package of the AD9835 on a veroboard I bought a very compact breakout-board for the AD 9835 made by artekit in Italy. Very neat, very compact, very nice. And cheap.
So, enough the talking, here’s the hard stuff, folks! The circuit (not yet finished, but OK as a preview):
The crucial thing is the arrangement of the 2 NE612-mixers involving 2 relays close to them and the 9-MHz-SSB-filter. The relays reverse the signal way between receiving and transmitting. For this purpose I used small print relays made by Finder company.
That’s all so far, the next days I will put in some photos of this neat little 20-Meter-Band-SSB-Transceiver to give more details to the interested reader.
73 the Peter (DK7IH)