Acoustic pressure from dB scale

[Amy_D]

Hi all.

I'm nearly there, after a lot of figuring out I now know how the soundcard and microphone interact to give a certain voltage for a certain sound pressure. Now I need to know (fingers crossed it's possible) how to work out what voltage/dB audacity is recording from the dB scale, so I can work back to the actual pressure I am measuring.

I know that the maximum input level (presumably when it "clips") for the soundcard is +2 dBu. I am recording with the soundcard TRIM on max, which gives 44 dBu of amplification, so when this is on I assume the maximum input is actually -42 dBu (in "real life").

So, can I just check I have this right please. If on the dB scale in audacity the dB level is -36 (as used in the example by Galeandrews) then the actual level is 10^(-36/20) = 0.016 of the maximum scale. So, in my case the maximum scale should be -42 dBu (explained above), which is (10^(-42/20))*0.775) volts = 0.006156044 volts. Does this mean that the actual voltage for this sample at -36dB in audacity is 0.00615 v * 0.016 = 0.000096 v? From this I can work back (using the sensitivity of the mic) to actual acoustic pressure.

Anyway, I hope I have explained how I *think* it works well enough for someone to let me know if I have it right or not!

It would be very much appreciated as it is for a scientific project and at the moment I have only "relitave" amplitudes, and to get more meaning from the results I really need to know the pressure. Thanks very much in advance.

Amy

[kozikowski]

<<<which gives 44 dBu of amplification>>>

44dB, not 44dBu. dB is a relative term. My voice is 6 dB louder than yours, wherever they started out. dBu, dBv, dBm, etc all have specific starting places or references. For example, 0dBm is achieved by an audio signal that generates one miliwatt into 600 ohms. The other popular one is dBFS. 0dBFS is where a digital audio system clips.

You can't use Audacity as an audio voltmeter to measure analog signals without a calibrated A to D converter in the middle. To get the calibration, you need reference to a standard audio signal or a standard audio voltmeter.

I'm going to read that again, but on first reading, I think you're trying to derive your way into a calibrated system based on circumstantial evidence.

Koz

[steve]

at the moment I have only "relitave" amplitudes, and to get more meaning from the results I really need to know the pressure.

From Audacity you can only get relative levels, unless you have a reference.
If you have a reference measurement, for example from an SPL meter or using a pistonphone, then with the appropriate mathematics you can work out the rest.

(this is a horribly complicated subject)