Any of you who play with Arduinos or R-Pi etc will be aware of the vast number of add-on modules available for experimentation. Amongst these are little UHF radio data link modules. Now, I used to use very high quality versions of these many years ago made by Radiometrix, but today you can pick up a Tx/Rx pair of these cheap imports for around a quid!
The transmitter of these is a little over half inch square, with just three connections for supply, ground and data, one for an antenna wire, and nothing more than a SAW (Surface Acoustic Wave) oscillator on the board! The idea being to simply use the data stream to key the device on and off, what is known as OOK (On/Off Keying, easy this radio lark init?)
But, thinks I, I want to send audio! Can it be done with these modules?
Well, sooo many people will tell you no, it cant. Or they will tell you you have to digitize the audio, or PWM the device. But is this really so? I mean, its really nothing more than a crystal oscillator, and we've all made simple 'bug' transmitters that are nothing more than an oscillator!
So I played with some ideas. Directly driving the data pin with audio didnt work. Neither did feeding the audio in via a transformer in the supply line. So I asked the wonderful folks who inhabit the G-QRP club forum, and was directed to Youtube for video of Peter Parker, VK3YE, experiments in a similar vein -
Peter has made these modules do it! In a later video he shows a DC bias system that improves the results. So I copied this to see if it would work,
I used whatever was already lying on my overcrowded, untidy bench - namely a 47k pot and a 100uF electrolytic, plus various wires and leads. 100mV at 1kHz audio in from the test set, 10V DC supply,
And bugger me with a fish fork! It actually bloody works!
In this test, I seem to get considerable, reasonably clean FM deviation. On AM there is mod there but its not very deep. Varying the supply voltage does vary the frequency, and the modulation level. But it quite clearly works!
Next job is to actually listen to it, see how it sounds in a receiver. From then, the next stage will be to experiment with the Rx module to see if audio can be recovered from it. I suspect the LM358 IC on board is being used to 'shape' the output to reasonable logic levels, so might have to play around with this somewhat.