Converting the 80M Receiver
Dale Hunt, WB6BYU
I converted one of my receivers to 40m to track some powerline noise here locally, though since it was temporary I didn't take the time to optimize the performance.
There are only two frequency-sensitive circuits in the radio: the VFO and the loop antenna itself. Basically, all you have to do is to change these to resonate on 40m.
The VFO is pretty straightforward. You can either reduce the coil or capacitors by a factor of 4. My thought would be to use a 1 or 2uH coil in place of L4 then add capacitance at C30 as needed to get the right tuning range. A tunable coil will make things much easier, but with some trial and error you should be able to get the VFO on frequency. (Since the VFO is on the operating frequency, you can simply hook a counter to the output of C9 or listen for the signal on a receiver.)
In theory, the coil could be retuned by removing C13 and readjusting C14 for resonance, but that didn't work very well on the one I tried. The problem was that the distributed capacitance in L1 made the loop self-resonant at or below 40m. One solution is to reduce the size of the loop: normally this will reduce the sharpness of the bearing and the signal strength, but in your application you probably can get by with a 4" diameter coil. Another approach is to create the coil in such a way as to minimize the capacitance between the wires (which tend to bunch together in places.) One approach might be to use high voltage test prod wire for the loop, since the thicker insulation (compared to the wire diameter) will reduce the capacitance. I also have a prototype loop wound on a wooden frame rather than inside the pipe: this allows more spacing between the turns. Anything else you can do to keep the wires more separated will help in the same way: for example, taping the wires to the outside of a thinner piece of tubing that can then be put inside a larger one and bent into a loop. (The coax can go through the hole in the center tubing.) Stuffing bits of styrofoam in the tube to keep wires separated is another possibility.
I can't give you a lot of explicit dimensions for doing this, since there are so many variables. But I found it works quite well to couple a dip meter to the antenna coil to find the resonant frequency. Or just run C14 through its range and see if the background noise peaks on the desired frequency.
You will have problems getting good bearings in a building due to coupling to the nearby electrical wiring. Don't bother trying to adjust the sense antenna - it won't work indoors very well. Usually what happens is that the signal you are looking for gets picked up in the electrical wiring, and the loop ends up giving a null when it is at right angles to the wires irrespective of the direction of the signal. However, I've had good results simply using the receiver as a compact hand-held sniffer. It has a good wide-range RF gain control rather than AGC, which makes it easier to simply wander around the building and find where the signal is strongest. Then move it around the walls and especially any exposed wiring or panels. You may be able to probe each cable in a trough if you can separate them a few inches apart.
Reducing the loop size generally isn't a good idea for outdoor DF, but shouldn't hurt for indoor sniffing. Going any higher than 40m gets tricky due to the distributed capacitance of the loop: the best solution is to reduce the number of turns on L1, but that changes the turns ratio with L2 and it is difficult to use a half-turn shielded loop.
The ultimate solution probably is to use an FET for Q1 so L2 sees a high impedance that won't load down the Q of the antenna. That will require some changes to the biasing in the front end, so will be more of a project.
Oscillator stability, feedback, etc. will also be more of a problem on the higher bands, but for tracking noise (rather than narrow-band signals) it should still work.