Hy Stanley,

from your scope plot, which I repeat here



you can find every detail needed to design a successfully working circuit:

1. Find out why the base line of the plot isn't totally narrow. You can very clearly see noise on this base line, which is superimposed to your signal. Play a bit with the deflection settings of your scope to find out what this noise is. Then find out how you can mask it!
Very probably a bandpass filter can be used to filter out anything which hasn't the same frequency as the transducer frequency. But take care, to me it looks like there are some older ultrasonic echoes superimposed to your actual echoe. So, you might have to make the time spacings between the ultrasonic bursts wider?

2. Have a look at exact frequency of ultrasonic signal. This will tell you the center frequency of your bandpass filter.

3. Have a look, how much the transducer signal changes in frequency from burst to burst. Take care, temperature changes can make your ultrasonic frequency undergo changes.
This frequency range tells you how high the damping of your bandpass should be. I guess a Q of about 5 is adequate. If Q = 5 does not suffice to suppress noise enough, then two or more identical bandpass filters can be put in series.

4. Vary the distance of what is reflecting the ultrasonic burst. Look at the scope how much the signal amplitude will vary. I guess it will be a range of about 1 : 1/200, or so. Highly depends on the range of distance you will allow to be detected.
This tells you the dynamic range of your signal and tells you where to set the threshold of your Schmitt-trigger.

Also:

(a) An AGC amplifier, which automatically controls the gain might not be a cure to your problem. Why? Just because the echoes are only lasting for a short period. Too short for most AGCs? More, if your signal is hidden in the noise, then an AGC amplifier will only detect the noise and nothing is gained. First of all, you must adequately pick out the signal from noise.

(b) A Lock-in amplifier which is a phase sensitive detector will not work, because the phase of the echoed ultrasonic burst isn't predictable and isn't constant. Think about what happens if the reflector of ultrasonic burts wanders a bit: This will also change the phase of echoed ultrasonic bursts.


Kai