Air will produce significant THD when compressed by more than 5%?

[lowerFE]

I was reading this book called "High Performance Loudspeakers" by Martin Colloms, and he mentioned something interesting. He claimed that when air is compressed by more than around 5%, the air itself will start causing a lot of distortion of the sound.

 

Does anyone know why that's the case? The 5% figure doesn't really hold up when looking at some of the highest dB/liter subs Josh has measured. For example, the highest dB/liter subwoofer is probably the MTX 9515-44 in the 1.25 cu ft box. It hit 118dB at 40Hz, and according to WinISD it takes 44mm of excursion and 21,000W to produce this level of output (how is Josh getting that kind of output with only a 10,000W amp?). The displacement at 44mm is 3.8L. The enclosure is 36L, so the air compression ratio is 3.8L/36L = 10.56%.

 

I know from the Xmax investigation thread that the calculated Xmax values don't make sense compared to the output measured. What if we make a lenient assumption that the excursion from the woofer is always at or below Xmax. Then in that case, the MTX 9515-44 is still at 6.25% air compression, but the distortion numbers around 40Hz (where output is still Xmax limited) are still very low.

 

I remember reading that when you reduce the enclosure volume of a sealed speaker, the distortion (I believe it was 2nd order?) caused by the non linearity of air compression will go up in a predictable way and I remember there was even a formula to predict this. I'm thinking this has to be related. If this is true, the formula must be a exponential function where there is a point where the chart just takes off, and that's the value of air compression percentage where air will contribute significant distortion.

 

What do you guys think?

[MemX]

I'm not sure I can reason (in my head, anyway) why compressed air might sound distorted - surely if the air molecules are closer together, they will just transmit sound more quickly? 

 

Unless compressing air creates some sort of turbulence that affects the transmission of sound through the medium?

 

 

Extension of that logic would seem to suggest that static air at a lower pressure will reduce distortion?

 

And if that was the case, surely we'd have seen 'audiophile' vacuum chambers that come with oxygen masks so you can listen to the 'ultra-clean' music inside it and still stay alive?   haha

[Funk Audio]

The main issue is that the air compresses asymmetrically, meaning that for a given displacement change on the inward stroke the air compresses producing a pressure on the cone. The same displacement amount on the outward stroke produces rarefaction applying pressure via atmosphere on the other side, but the amount of this is different than on the inward stroke. Different drivers will be able to overcome this by differing amounts, and at different frequencies more and less, depending on their resonance, BL curve( also varying with frequency, and drive level)

[MemX]

The main issue is that the air compresses asymmetrically, meaning that for a given displacement change on the inward stroke the air compresses producing a pressure on the cone. The same displacement amount on the outward stroke produces rarefaction applying pressure via atmosphere on the other side, but the amount of this is different than on the inward stroke. Different drivers will be able to overcome this by differing amounts, and at different frequencies more and less, depending on their resonance, BL curve( also varying with frequency, and drive level)

 

Am I right in thinking this assymmetricity (is that a word?   EDIT:  Wait, I think it's assymmetry!) can be fixed or at least partly overcome by the use of a setup similar to Bosso's, with the plates opposite the face of the drivers to create an increased pressure against the outward stroke?