Hello

I am currently working on an object localization project. One of the main parts of the project is a system that will measure the Time Of Flight (TOF) of audible frequency pulses of around 4Khz that will be transmitted by low cost buzzers (just like the ones inside computers). The audio signal will be detected by an electronic circuit, producing an external interrupt for a microcontroller that will eventually measure the TOF. The maximum operating range will be at 6-10 meters and resolution about 2-5 cm

The problem is that i am quite new to electronics and i havent quite figured out the circuit that will amplify the audio signal from a mic, filter it using a low loss bandpass filter and then produce a transition from high to low or the opposite when 4Khz pulses are detected. Could anyone help me with this circuit? Also, could someone help me with the component selection (buzzers , mics etc) in order for the system to be as fast and sensitive as possible?

Thank you in advance
Kampianakis Lefteris

P.S I am seeing quite a few posts about TOF measurement using IR LEDs could anyone explain this technique?

 

I will say that using a 4khz acoustic signal to measure TOF less than it's wavelength (~8.6 cm) will be a neat trick. I think you need to do some background study on sonar.

 

Zero loss bandpass filters are called "active filters". They are made with operational amplifiers.

This is entirely achievable, but all you have right now is an idea, and a faulty one at that.
I see 4 khz as 8.38 cm in Earth atmosphere. If you want to measure to 5 cm, you should be thinking in terms of a wavelength shorter than 5 cm. 7kHz or higher.

 

I will say that using a 4khz acoustic signal to measure TOF less than it's wavelength (~8.6 cm) will be a neat trick. I think you need to do some background study on sonar.

That's if they emit sinewaves.

You could pulse the buzzers with a very short pulse (low duty squarewave) and receive the pulses, resolution could be much shorter than the frequency. Think of it more like multiple sonic peaks then as a 4kHz "frequency".

 

Ok, its clear to me that i should better use a higher frequency than 4Khz. This is easily achievable by using connected to the output of an mcu with which i cann control the frequency of the pulses. Then, i can find the minimum length by calculating the wave length. Done with the easy part

Now, regarding the audio pulse detector circuit : i have been experimenting with some audio amps and active filters and i have came up with a circuit that when audio pulses of a certain frequency range are "heard" the circuit "drops" the output, from 3 to 0 Volts.

However this is an extremely complex (and quite expensive) circuit for my needs (schematic is attached). Do you, as more experienced, have any proposals as regards to the circuit? I also understand that trade-offs will appear regarding, cost, sensitivity, power consumption , complexity, etc. The quantities that i am trying to minimize first are the complexity, then the cost.

Thank you in advance