Bo Deadly
Well-known member
Is it feasible to send data over audio so that some audio device can be digitally controlled in-situ?
For example, let's say you have a remote device sending / receiving 4 channels of audio over CAT6 but you want to be able to control things like gain, low-cut and so on without loosing a channel. So maybe you send an AM signal with 50kHz carrier and a circuit like the following:
This circuit is just the transmitter but I think the receiver would be pretty much the same.
V1 is the audio source at 20kHz. V2 is a 50kHz carrier. L1 C1 are tuned to be high Z at 50kHz. C2 is high Z at audio frequencies. A balanced version using common mode chokes wired anti-parallel might work even better.
The FFT shows that the carrier signal is thoroughly isolated from the audio.
The AM would just be bit-banged out of a PIC to make pulses representing the data. I figure a lowly PIC can reliably read an input every 100us so that's 5 cycles of 50kHz to represent a digital 1 and 100us of nothing to represent a 0? So that should be able to transmit about 10kbps minus some kind of initial start key.
What do you think? Is line capacitance going to foil me?
For example, let's say you have a remote device sending / receiving 4 channels of audio over CAT6 but you want to be able to control things like gain, low-cut and so on without loosing a channel. So maybe you send an AM signal with 50kHz carrier and a circuit like the following:
This circuit is just the transmitter but I think the receiver would be pretty much the same.
V1 is the audio source at 20kHz. V2 is a 50kHz carrier. L1 C1 are tuned to be high Z at 50kHz. C2 is high Z at audio frequencies. A balanced version using common mode chokes wired anti-parallel might work even better.
The FFT shows that the carrier signal is thoroughly isolated from the audio.
The AM would just be bit-banged out of a PIC to make pulses representing the data. I figure a lowly PIC can reliably read an input every 100us so that's 5 cycles of 50kHz to represent a digital 1 and 100us of nothing to represent a 0? So that should be able to transmit about 10kbps minus some kind of initial start key.
What do you think? Is line capacitance going to foil me?