I'm trying to design a proximity detector for a PIC-based robot (I just need to know if something gets close, not how close it actually is). I've bought a pair of matched ultrasonic transducers from Maplin, but I'm having trouble with the circuit (I don't have access to a 'scope). I want to build something from scratch rather than but 'module'.
I've come across this design to use as a basis for the reciever:
I understand that Q3 and Q4 provide amplification and that D4, R14 and C5 provide 'rectification / peak following' (not sure of the correct term) for the ~40kHZ waveform.
Curiously, D3 has no function in the said circuit, because it's permanently reverse biased. If you put a capacitor in series with R13, it would have a purpose.
As a general remark, it's a very simple proximity detector, because it only measures intensity rather than transit time of ultrasonic sound.
Indeed, I'm aware of the limitations of the circuit, I don't need accurate (? - a relative statement due to the speed of sound through air) measurements, just an indication of whether something is in the way or not.
Would it be simpler to do this with a modulated IR source? Wouldn't I need additional circuitry in this case, to provide the (narrow) band-pass of the signal frequency (whereas with the ultrasonic transducer this is essentially inherent due to the resonance affect)? Which frequencies for the IR would people recommend (and more importantly why)?