This is a quick update post. The road I took with this little project is slowly getting to a dead end due to what it seems to be a lack of available crystals for 144.8 (like 48.26 or 28.96 MHz, for using the third or fifth harmonic). I believe that, currently, all APRS modules are done with some sort of DDS, which should be much easier to do. Due to my lack of expertise in this area (in the end is a hobby) there might also be some other methods to get — by analogic means — the carrier for APRS (144.8 MHz). I will dig further, but for now it seems that the DDS path is much more convenient. Although this looks like a dead end I don’t mind, I’ve learned a lot. Including how to use my tools. I made some minor modifications today by adding a rudimentary input for modulation. It works, but the modulation is very weak (or the output power is too low). Boosting (+ 6 dB) the input through a preamp does not help. On the contrary, some nasty artefacts occur. Back to the drawing board.
This should be the starting point for improvements. Now the prototype looks like this:
The latest “look” of the APRS TX proto board. I extended the board with an additional strip of PCB (left side) and glued some green ABS strips beneath the entire stuff, for easy fixing and stability in both horizontal and vertical position.
I glued 🙂 an extra PCB board (for the preamp part) and stick some ABS strips beneath the rest of the board, for a better prop and grip. These also helped me fixing the board in a vice, in vertical position for an easier connection of various cables and attaching the power supply (via two crocodiles). The latest version of the schematic is this (I will probably get rid of the preamp and add an additional RF amplification stage):
Prototype revision 3.0; this version includes a preamp for modulation and a simple way to modulate the 144.0 MHz carrier.
Some comments about the filter. I did some simulation today and the simulated profile seems to fit the findings in the real world:
Amplitude vs. frequency response for the 9-pole 2m output bandpass filter. Simulation done for an AC step response between 50 MHz and 380 MHz, 1,000 steps / octave (.ac oct 1k 50Meg 380Meg).
The simulated response is better than the calculated (-50 dB @ 108 MHz simulated vs. -45.3 dB calculated etc), however the actual measurements I took (see here) are with one order or magnitude poorer. However, next step will be to create a proper PCB (etched) and get rid of this dead bug layout. I suspect that many artefacts and poorer performance than expected are also due to this method of laying out the components.
Not much to add for now. The main problem I face is to find an appropriate crystal to get 144.8 MHz carrier for APRS. It seems there is no such crystal as 48.26 (third harmonic) or 28.96 MHz (to use the fifth harmonic). DDS approach will be a parallel development for sure, I still did not gave up to my original idea of using only discrete, analogue components. That’s it for today.
DDS for APRS Tracker? May be a interesting experiment but the DDS is power hungry opposites to a PLL.
Maybe a compact solution like this http://www.radiometrix.com/content/hx1 is better. GL & 73 de YO3HJV
Hi, Adrian. Thank you for your comment and interest in this. Actually a mixed DDS/PLL one–chip solution based — for example — on Analog Devices’ ADF7021. With a 12 mA quiescent (15 mA Rx/ 23 mA Tx) and 0.1 μA leakage on idle, that would do the trick even for the most power–aware applications. However, this entire project is an experimental prototype. I was curious to explore the hurdles one might experience while trying to achieve an APRS beacon taking an analog approach (or as much analog as possible). I already face big problems trying to achieve 144.8 MHz from quartz oscillators. The most convenient way would be to stick to a Radiometrix 144.8 MHz module. But although the goal is an APRS beacon, the road should be as experimental as possible. Thus the musings. 🙂
Warmest regards,
AP