# Sound attenuation

1. Sep 24, 2007

### fisicasound

Sound attenuation

for my physics project I decided to investigate sound attenuation effect, for my equipment I’m using a speaker with a signal generator and at the other end a microphone with an oscilloscope...the sound generator and receiver were placed inside a tube which was well insulated to prevent the apparatus resonance(but i still get a resonance effect and I don’t know if it will affect my results)...also the apparatus is designed so I could insert different materials and different thicknesses in between the receiver and speaker

I was wondering if anyone could answer the following questions.

1. to show my results in dB i must find the sound intensity in W/m^-2...is it the same as the peak to peak voltage readings in my oscilloscope

2. What type of materials should I be looking for that can be easily accessible?
I was thinking of polystyrene, also what properties in a material make it good for sound insulation?

3. To plot graphs I was thinking of using the equation lnI= lnIo - μx where μ is the materials coefficient of attenuation and x is the thickness of the insulation....for the graph I was advised to vary x keeping the frequency constant but I don’t really know what frequency I will chose since humans have can hear a really big range of frequencies.. will the second graph for the same material vary μ as I change frequency

if u have any suggestions for any other experiment suggesting sound attenuation ill be glad to hear it

Thanks

2. Sep 24, 2007

### CaptainZappo

3. Sep 24, 2007

### fleem

If you are only measuring attenuation, then you only need to know the ratio of power lost, and therefore only need to compare amplitude with attenuator to amplitude without attenuator. Don't forget to square the amplitude to get power.

If you really want to know absolute sound amplitude (and square it to get power), then you need to find out the relationship between the P-P you see on the scope, and the particular microphone you are using along with the gain of the amplifier, if any. It could vary wildly from mic to mic, and also with frequency. It will certainly be a constant proportion, as long as you still see a clean sign wave (no saturation, distortion). You might have to measure it yourself for a particular microphone, but then you'd need a source of sound with known output power and radiation pattern. Maybe there are some ideas on the Internet for making such a source.

If you want to even roughly measure a coefficient of attenuation of a material, the gradient from air to material must be gradual over at least a wavelength so that impedance is matched. For example, the material should be formed such that its surface (both sides) is comprised of cones with a height greater than a wavelength, and separation between the cones notably less than a wavelength. Otherwise there'll be reflections. For practical reasons, this means you'll want to test with a short wavelength, if its allowed. Of course, if you just want to demonstrate the system then you can have any shape and thickness you want--but expect reflections. In calculations, use the length of the material from the top of cones on one end to the bottom of cones on the other--that is, include only the cones on one end to calculate the effective length.

Another option is to use a material who's impedance is close to that of air (typically something very light and soft, like cotton or a fluffy fabric stuffed in the tube), and use a long cylinder of it (many waves). That way you'll get a measurable attenuation with little reflection.