2. speakers
  3. 1091420566

resonance and sealed enclosures

I was wondering: suppose you have a smallish sealed subwoofer with a resonance frequency of 40 Hz (obviously not up the quality of the things usually talked about here). This is right at the frequency of the low E on a bass. My concern is that the reponse at 40 Hz will be exagerrated. The resonance peak on a sealed sub isn't important when it's 20 Hz or below. Will the 40 Hz resonance really matter?

I am thinking about using this sub for computer speaker BTW.
daniel_newhouse5e69
danielnewhouse6adc
2 posts
 
Sean, what do you think of Miller and Kreisel's push/pull design?
sean
6,231 posts
 
Push / Pull lowers distortion, increases power handling, faster transient response, lower impedance peak at resonance, improved power transfer characteristics, etc... If properly implimented, it can be quite beneficial. I'm using a dipolar push-pull arrangement for my own personal subs. The dipolar pattern reduces standing waves, which further improves the linearity of in-room response. Problem is, you've got to use twice as many drivers, resulting in a lower impedance, much higher cost to build and the need for a much sturdier amp. Sean
>
danielnewhouse6adc
2 posts
 
...I thought push/pull by nature required 2 subs. Are you using 4 subs?
drew_eckhardt
394 posts
 
A sealed speaker's response at resonance (Fb) depends on how the system is damped (Qtc).

A sealed speaker acts as a second order high-pass filter. Ultimately it will roll off at 12dB/octave with Fb determining where the roll-off starts and Qtc the shape of the transition. With Qtc <= 1/sqrt(2) (about .707) there is no peak.

A Qtc of 1/sqrt(2) is "maximally flat." It provides the slowest approach to the ultimate second-order roll-off without having a peak. The speaker will be 3dB down at Fb. Perceptually you get a bit of bloom.

A Qtc of .5 is "critically damped." The speaker will be 6dB down at Fb and have an F3 point half an octave higher (60Hz for Fb=40Hz). Such a speaker does not ring. Perceptually the bass is very tight.

Frequency response for various Qs can be seen here:
http://www.diyaudio.com/forums/attachment.php?s=&postid=88030

Note that small deviations in output are _much_ more noticeable in the bass area because of the equal loudness curve spacing.

Fb and Qtc also affect phase shift as a function of frequency (group delay). While you may not need a lower Fb for music, a 20Hz Fb gets you half the group delay at 40Hz as a 40Hz Fb. Low group delay is perceived as fast bass.
More to discover