frequency response of air

[mikelm]

Does anyone know the frequency response of air and formulas that describe it ?

I was thinking about how much attenuation at different frequencies and different distances from a sound source

mike

 

[sreten]

Farfield nearly all sound sources will follow the inverse square law,
this dominates any attenuation or frequency dependent losses.

sreten.

 

[mikelm]

OK sure, but I wondered within this, wether some frequecies that are attenuated more that others due to the charactoristics of the air.

By way of example, I get the feeling that square wave leading and perhaps trailing edges will be softened with distance.

mike

 

[sreten]

Its the different speed of sound through air at different densities
that effectively kills high frequencies through phase interference
and frequency spreading.

A sharp lightening transient will be spread by the different
speeds of sound in the various atmospheric layers, as it
spreads its frequency components wil continually lower,
giving only the rumble effect at long distances.

sreten.

 

[Christer]

I also think the humidity of the air matters, or at least I get the
impression that higher frequencies are more attenauted when
it is very humid.

 

[Sch3mat1c]

I think he wants to know about ultrasonics. Like, are they attenuated any more than otherwise?

Speaking of which, something I've always wondered is just how high a frequency can be transmitted acoustically - I know over 100kHz exists!

Tim

 

[Christer]

Speaking of which, something I've always wondered is just how high a frequency can be transmitted acoustically - I know over 100kHz exists!

Unless I misremember the figures 300kHz obviously works well
http://www.diyaudio.com/forums/showthread.php?s=&threadid=6309&highlight=

 

[mikelm]

I think he wants to know about ultrasonics. Like, are they attenuated any more than otherwise?

Not exclusively,

Ideally I would like to see frequency response graphs

10hz - 100khz

at 1 meter, 2m, 3m, 4m, 5m.

what happens to a square wave leading edge would also be interesting.

 

[Bill Fitzpatrick]

what happens to a square wave leading edge would also be interesting.

Finding the transducer for this would be quite a challenge!