This is intended as a general
repository for circuit notes about transmitters, receivers,
and ancillary circuits that are useful in ultrasonic
systems such as TAPS. Maybe you don't know it but
retirement is a full-time job and I don't have near the
spare time I used to have when I was working!
Designing self-contained acoustic systems requires one to deal with two major problem areas: power consumption and noise.
Since a self-contained system must run on batteries, extravagant use of battery capacity leads to short deployment life. This constrains one to use switching-type power amplifiers, which then require low-pass filters as part of the matching network to the transducer. Class A power amplifiers draw too much current when idle to even be considered. See the page on TRANSMITTERS.
Echoes from zooplankton, in particular, are quite weak. Echo voltages can be in the micro-volt range. This requires considerable gain in the receiver and very careful attention to sources of noise. If you are required to combine the transmitter and receiver on a single card, the various oscillators required can cause noise to travel on the groundplane into sensitive parts of the receiver, swamping the weakest echoes. Since I always needed to measure the echo amplitudes precisely, it turned out to be a better design to have both + and - supplies for the receiver. Problem was, I had a single battery to supply power. I have had some success with voltage inverters, although these can be a vexatious source of noise themselves. Careful layout of ground planes, some shielding, and a few inductors in the power leads seem to make things work pretty well. The best receiver I ever built had a voltage inverter section in it.
I have been intrigued with the notion of a simple receiver (bandpass filter, impedance matching, maybe a little gain) driving a fast ADC into a DSP chip. All of the processing, including some filtering, could be accomplished in the DSP. These are devilishly hard to program and I haven't yet made a serious effort to learn. Sigh ... someday.