Minimum detectable pressure

One doesn't, unless you also know how many volts output is 'minimum detectable'. It gets complicated, with the noise-versus-frequency distribution and the dwell-time at a given frequency of input, if your detector can be tuned for a small frequency range.

Hey, thanks. I really couldn't see how a number relative to 1V/uPa could be the whole answer. So is -175dB re 1V/uPa just a dumb way to quote the volts per Pa a sensor puts out? So like if I have y re 1V/uPa = 20log(x/x0) then x = 10^(y/20) (1V/uPa). So a -200 dB sensor is less sensitive than a -174 dB sensor given the same frequency band and electronic noise levels?

Yes, that is correct. It's a useful number in the sense that it helps a designer with attenuator/preamp calculations, and a 'dumb way' in the sense that it is hard to relate to a pressure-in/voltage-out measurement.

Since this is important to me, let me be certain. The ratio of the detector sensitivity is in units of V/uPa. When divided by an arbitrary 1V/uPa reference the number drops from -174dB for the first detector to -200 dB for the second meaning the second detectors sensitivity or Volts/uPa is a smaller smaller number than the first detector. Given a fixed acoustic pressure the first detector will produce a larger output voltage than the second. How else can this be interpreted?

Will this help?

Yes, it does. -200dB re 1V/uPa is indeed less sensitive than -175dB. IMO quoting as dB re 1V/uPa does well to obfuscate a simple and straightforward concept.

Hi Paul,

The quantity you are discussing is responsivity. Responsivity tells how many volts the sensor outputs per uPa. For the sensors you listed, -174 corresponds to an output of 2 nV/uPa, and -206 corresponds to an output of 50 pV/uPa. The first sensor will output about a 40x larger signal than the second sensor for a given input signal.

Sensitivity is related to the smallest signal you can detect with some level of certainty and depends on the responsivity, the quantity of noise present and the time available for the measurement.

The details of a specific measurement scenario are important, and the above explanation is a gross understatement and oversimplification, but hopefully is not too misleading.

Surely it is +6dB. In your example the first transducer has sensitivity

Or have I got it all wrong given the date?

piglet

Clearly a sane definition

So when a manufacturer list the sensitivity of a hydrophone (just the sensor) as -174 dB re 1V/uPa he's likely talking about resposivity?

Yes, that's my best guess of what they mean. For some certainty I suggest you take a look at the definitions of precision, accuracy, responsivity and sensitivity. Look at what they mean mathematically. Then you'll have less chance of a misunderstanding based on semantics and thus of making a bad assumption.

It's much easier to fix bad assumptions at the beginning of a project than at the end. ;)

Best Regards,

ChesterW

I don't know about hydrophones, but I think that I know something about radio receivers.

For VHF and up, the sensitivity is limited by the electronic device thermal noise density at -174 dBm/Hz noise floor for 3 dB NF at room temperature (300 K). At lower frequencies, the sensitivity is limited by the band noise.

So you really have to figure out, is your system limited by (audio) band noise or amplifier white noise. Whichever is worse, will limit your ability to extract the real signal.

Yes, John, but for the same reference sound pressure, the output of the second one is 2V which is twice as much for the same reference pressure level. Yes?