After having used the BITX40 for a while as a receiver the drift of the supplied VCO started to anoy me. Then when I made my first QSO with it, a QSO with my friend Gerald PA3GEG, Gerald told me that the drift was so massive that he had to use his RIT all the time. At the end of the QSO he told me that he has recorded the QSO and would send it to me by mail. Having recieved that recording, a new #1 (moving the dynamic mic amp to #2) was on my todo list : a DDS for my BITX40.
Getting started with it
Having buld a DDS VFO for my FT-301, I based the BITX40 DDS on that design and source code. And after a hour stripping the sourcecode, I used my prototype shield of the FT-301 DDS to test it. And it worked as expected.
So after that it was time to rebuild it on a breadboard with the parts I would use in the final design.
Video source : https://www.instagram.com/p/BPAJrCZh6XY/
Building the final DDS (and adding drama)
After building the DDS and testing it, all seemd fine and I started to build the final DDS assembly.
Final test before make a large hole in the case.
Image source : https://www.instagram.com/p/BPDWpHbB4kb/
After installing it, checking the wires the moment was there. Power on, display init and a screaming high tone almost made me deaf. Quickly switching down the power.
At first I thought I made some wireing errors (ground loop) but dubbel checking excluded that one. Then I dis-assembled the DDS module and tested it lose. Still the screaming tone and by excedent I disconnected the MAX7219 display. . . and the tone was gone.
One thing I didn't expected was that the MAX7219 displaydriver would be a nasty noise generator, but some google later I found out that the MAX7219 does has that effect on sensitive analoge electronics. Big bummer because I wanted to have a nice LED display in my BITX40.
The question remained : why didn't it make that noise when I tested it on my breadboards. The only difference was . . . two seperate power supplies. Big bummer nr 2. So to make it work I had to find a way to cleanout the noise from the power wires.
After a few hours experimenting I found out that the "noise killer" that workt for this setup was a setup of 1 doide, 2 inductors of 1mH, 2 capacitors of 3300uF and a piece of RG174 as power cord to the DDS from the "noise killer" and some manhatten style PCB work.
TODO : INSERT PHOTO / SCHEMATIC HERE
After that the noise was gone and even the static didn't have any strange tones in it anymore.
Some photo's from the final assembly.
After having worked a few evenings with the BITX40 with a DDS, I found out that the selection of the frequency could be better. Although the displays shows 1 Hz, the minimal stap is 5 Hz, which in most of the times is just to small (most of us can't hear a offbeat of 20Hz). So why display the 1 Hz digit and use such a small staps anyway? I changed the source code so the display is only showing 10 Hz and up and changed the steps to a comfortable 10/50/100/500/1k/5k/10k. I don't think I would be using the 5k and 10K step much, but for now I'm leaving it in the source code.
Another thing I added is a VFO a/b switch allowing me to change between 2 frequencies quickly and allowing me to add split in the near future (using a On-Off-On switch and finding a good way to switch on PTT).
I uploaded the source to my GitHub and can be found here : https://github.com/ph2lb/BITX40-led-vfo
Continue reading on : Page 4 - mods
Update : As of late december 2016 the BitX radio module has a DDS shipped with it, allowing more frequency range (entire 40m band) and less drift as the original VFO. The only difference, the price went up a little (now $59 still incl shipping). So now for a little extra, more cool stuff to experiment with :-)
Update 2017-12-27 : On a Bitx40 facebook poste, Ricardo Martins pointed out that he also had the same problem with the MAX7219. He documented his findings on his blog. A absolute must read : http://speakyssb.blogspot.nl/2017/09/max7219-8-digit-display-module.html