[TronII]

I've been looking at graphs for THD for different headphones on Innerfidelity and have noticed a pattern that exists regardless of driver type: nearly all headphones have harmonic distortion in bass frequencies that reach audible percentages; why is this?

 

[TronII]

I'd like to add that I know why it's higher in bass frequencies, but I don't get why it's that much higher.

 

[castleofargh]

the highly technical answer is that at high frequencies you surprise the air by coming at it fast with the membrane and running away in the opposite direction before the air has time to realize what happened and fight back. while with low frequencies, the push is slow and the escape is slow so the air can turn around, call out some friends and punch the driver in the face.

if you're not entirely satisfied with that answer about air displacement at low frequency(I really can't imagine why ^_^), there is always this http://www.aes.org/aeshc/pdf/how.the.aes.began/white_problems-of-low-frequency-reproduction.pdf
and I'd add possible ambient noise where low freqs are often the loudest because it's harder to isolate properly from them when measuring stuff(on Innerfidelity the measurements were THD+noise). and another occasional suspect could be the amplifier itself.

 

[bigshot]

It's worse with speakers because of excursion, but it doesn't matter that much because bass is a fat indeterminate sound compared to treble anyway. Dynamic punch is the thing that matters most in bass from a practical point of view. I always suspect that when people talk about "tight bass" they are really talking about tight lower mids. That's where the definition of the "pluck" is.

 

[71 dB]

To have the same SPL at an octave lower frequency you need to increase amplitude of the diaphragm 4-fold. Two octaves lower the amplitude is 16-fold and so on. The suspension of the diaphragm isn't linear forever and starts to generate distortion when the amplitude exceeds the more or less linear part. That's why the distortion raises fast at bass with audio transducers and as bigshot correctly points out, speakers have even more distortion.

Tight bass is related to damping, how fast the sound dies off when the signal stops. Systems with low damping have large group delay around resonance frequencies (of the cone/diaphragm) causing the bass appear coming delayed to the signal and also less controlled. Tight bass is a result of damped resonances = controlled sound = short group delay.

The good news is our hearing is quite unsensitive to distortion at low frequencies. Something like 1 % THD below 100 Hz is nothing to worry about, but around 1-2 kHz that same THD would be quite audible because the first harmonics hit the most sensitive range of hearing (3-4 kHz).

 

[TronII]

Thank you all for the responses!

 

[TheSonicTruth]

It's worse with speakers because of excursion, but it doesn't matter that much because bass is a fat indeterminate sound compared to treble anyway. Dynamic punch is the thing that matters most in bass from a practical point of view. I always suspect that when people talk about "tight bass" they are really talking about tight lower mids. That's where the definition of the "pluck" is.

Correct. A well-damped kick drum(stuffed with blankets for example) is most present in the lower mids, plus a mild 'click' from 3 to 6kHz. A good thing to point out is that with an undamped kick, or parade bass, most of that thunderous tail is distortion. Not exactly desirable when tracking a rock quartet in studio, lol!