Connecting ultrasound sensor to computer

I am carrying out an experiment at school to measure the ultrasounds produced by plants when the xylem vessels cavitate due to water stress. I have a sensor that picks up sounds of approximately 40kHz. However, I now need a way to connect this sensor to a computer to record the sound patterns it picks up. The school technicians say that a USB interface is the best way to do this, combined with software to process the information from the sensor. Also, it must be able to deal with high frequency and low voltage, which makes it more specific and harder to find, so we are having trouble locating the equipment. I was wondering if it would be possible to use a Genuino product as the interface. If anyone has any information about this or know of somewhere else I could obtain an interface and appropriate software, that would be enormously appreciated. As a side note, I really have no idea what I'm doing, so any help at all would be fantastic. Thank you.

How big a signal does your transducer generate? you will probably need an amplifier and some form of detector…

regards

Allan

Following on from that, have you a link to a data sheet which describes the characteristics of the ultrasonic sensor ?

http://www.murata.com/en-eu/products/productdetail?partno=MA40S4R

http://www.murata.com/en-eu/products/productdata/8797589274654/MASPOPRE.pdf?1450668610000

That should be a link to the page with the sensor on it, and also a link to the data sheet about the sensor.

Also, what measure of how big the signal is do you need? I think the sounds being measured are between 30 and 50 dB and the voltage is very low, at no more than a few millivolts. If any other measurements are needed, I can try to get them.

If you follow the link here, you'll find someone has hacked a device (an ultrasonic motion detector) which contains a similar ultrasonic detector to yours. There are circuit diagrams, including an amplifier stage, which may be useful to you. http://electronics.stackexchange.com/questions/34447/can-anyone-identify-the-microcontroller-on-an-hc-sr04

Without knowing anything about "xylem vessels cavitation" hence the sound spectrum of interest, but as someone with a strong background in digital signal processing, I'm not sure Arduino is the right approach. My inclination would be to find a broadband microphone with response into the ultrasonic range and run it into a PC sound card* capable of 192 kHz sampling rate.

At least the following potential issues come to mind with the approach as described:

1) The sensor you've selected is narrow band, that is it will detect sounds only near 40 kHz, which may or may not be all that your interested in.

2) It will require a pre-amplifer prior to to analog to digital conversion.

3) The techniques to getting sampling rates sufficient to capture a 40 kHz signal on the low end Arduino processors are beyond beginner level twiddling.

Edit: * - after writing this post it occurs to me that a generic PC sound card probably has a low pass filter that attenuates ultrasonic frequencies. That would have to be modified or defeated somehow . . .

For what it's worth, a Google search on "ultrasonic microphone" takes me to web pages of people who record bats, birds, and such at ultrasonic frequencies. Some of them are using relatively inexpensive home-brew equipment.

It seems like you need something like a bat detector... you can buy them.

But if you want to do it yourself...

Having looked at the transducer spec...

the poure analog approach.....

I think you'd need about 60dB of gain ( 2 reasonably good opamps), a double balanced multiplier ( 2 4-quadrant mutipliers , 2 quadrature networks, summing amplifier)

and a variable frequency oscillator.

I designed a decent bat detector some years ago, which recorded bat calls as .wav files on an SD card. This used a different approach : a zero crossing detector, driven by the bat noise, divider chain and sample and hold followed by a 12-bit A/D. Hence the bat sound was recoded at 16th or 1/32 it's actual frequency, but the envelope of the waveform was retained...

It wasn't entirely trivial..

regards

Allan