Skip to first unread message

NZ0I

unread,
Mar 28, 2019, 10:01:00 PM3/28/19
to Receiver Development Platform
It recently dawned on me that there might be a simple way to add FM tone modulation to the ARDF transmitter design: by simply toggling back-and-forth at an audio rate between two VHF frequencies 10 kHz apart. A narrow-band FM detector would, I believe, demodulate such a signal as a square-wave audio tone. 

My first thought was to accomplish this fully in software. But to have the software reprogram the Si5351 400 times a second (for a 200 Hz audio tone) is probably beyond the capability of the ATMega328 processor and the I2C bus. But the same thing could be accomplished in hardware with the simple addition of an analog SPDT switch. The Si5351 can be programmed to output the two VHF signals continuously, then the processor can toggle the analog switch between the two signals at 400 Hz or even faster if desired. 

The main concern that I have about this approach is the RF signal bandwidth. It seems that the sudden transition between the two VHF signals will likely result in discontinuities at the switch transitions. Whereas the traditional approach of FM-modulating an oscillator by pulling its frequency up-and-down keeps the RF output free of hard-transition "glitches", switching between two separate oscillator outputs might result in discontinuous transitions that create FM "splatter".

But perhaps such splatter could be reduced by decreasing the separation between the two VHF frequencies? Or perhaps, since both VHF clock signals are clocked from the same time base, the discontinuities can be minimized by selecting two VHF frequencies with a difference that is an integer multiple of the audio-modulating switching frequency?

I don't know what the answer is - but if anyone has any thoughts on the matter I'd be interested to hear them. Otherwise, the easiest way to find out if this might work is to give it a try. I think that the RF Explorer will be capable of displaying the resulting FM-modulated waveform to determine its characteristics. The experiment will need to wait until the next transmitter board spin, or when there is time to hack the hardware together.

73,
Charles
NZ0I
Reply all
Reply to author
Forward
0 new messages