6
$\begingroup$

Has anyone experience with using low-cost hydrophones, for example the Acquarian Audio H2a, and use a calibration measurements, for example by comparing to a Brüel Kjaer 8106, to get quantitative sound pressure measurements? Can such frequency dependent calibration be applied to recordings already made by many of the whale watchers? How would such a calibration vary from one individual hydrophone to the next and how stable are they over time? How would such a procedure affect the noise level and the dynamic range?

$\endgroup$

1 Answer 1

4
$\begingroup$

The answer perhaps depends on how accurate you want your measurements to be. Do you want them accurate to .1dB, 1dB, a few dB ? If you can accurately measure the frequency response, then in principle it's possible to account for that in the analysis - though you may spend so many hours sorting it out, that you regret not buying a better hydrophone to start with.

Another very important thing to consider is the calibration of the recording equipment. This particular hydrophone seems designed to plug into the mic input of any recorder. it's therefore likely that that recorder had some kind of gain control on it to adjust the sensitivity. As well as knowing the hydrophone calibration, you'll need to know exactly how the recorder was set up and probably calibrate the recorder yourself with a signal generator and oscilloscope. With historical data, the exact settings may have been lost, in which case the data will not be usable for calibrated measurements.

$\endgroup$
1
  • $\begingroup$ The precision or accuracy is actually what I am interested to know. A +/- 1 dB precision would be great, but a +/- 3dB would also work. I probably need to measure the precision and the long-term stability if no one has already experience with this. I will not regret the effort for a 32 channel array, the cost saving is substantial. Unless someone points out kind of a showstopper that I am overlooking. Yes, the consumer electronic recorders often do not save the gain. But at least we could use the calibration data for a whitening filter to get the spectral shape correct. $\endgroup$ Commented Jul 29, 2022 at 16:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.