corrugated enclosure internal walls

[gmed]

is there an exact science to corrugated enclosure walls to minimize standing waves? just make them random? Can I alternate 3 different patterns? Im planning another translam, but this time a floor standing with the Mid chamber corrugated.

[JonMarsh]

The potential for standing waves depends on having internal dimensions that are long enough in relation to the reproduced wavelengths that the possibility for internal modal points to form; i.e., to get standing waves.

Lets take the example of a three way system with cone midrange, perhaps like the one I'm working on right now, or like a commercial system I've been looking at, the Avalon Indra. The LF to MF crossover is at 300 Hz. The wavelength of 300 Hz is 3.76 feet, approximately. The longest internal dimension is about 3 feet, so one could argue it's getting close. If you do a two way in such a cabinet, say, like the M8ta, then the LF transducer handles an even wider range, of course.

My point in this, (yes, there is one :W) is that in all my years of doing this, with a well designed conventional crossover, good wall stiffness and damping of the internal enclosure with suitable acoustical materials, I've never seen evidence of backwave standing wave phenomena influencing the frequency response, EXCEPT for those times when the driver is partly masked for the immediate rear wave coming off the back of the driver, in which case you see irregularities in the response in roughly the 800-1200 Hz area, usually some kind of peaking or even a cliff like phenomena associated with peaking.

What FAR more influences the subjective speaker performance, in my opinion, is the driver baffle design and polar radiation into the room (total power response), and room placement and treatments, to avoid early reflections or strong indirect path reflections (from the adjacent boundaries, like walls), and to position the LF portion of the speaker at golden mean ratios to adjacent boundaries tuned so that the boundary lift starts to occur where the anechoic LF response is rolling off.

A lot of other people have investigated this, including the idea of enclosure shape, and for the most part the box is acting as a pressure container, not a wavelength resonator as say, a tuned microwave transmission line. So, my suggestion is to focus on other topics that may bring measurable and audible results.

Conduct your own study if you feel compelled to, don't take my word. Get a good instrumentation microphone, like the ACO Pacific 7012 or B&K 4133 like I use, which can handle relatively high SPL's in nearfield to drivers; the way to investigate might be comparing measurements on true nearly infinite baffle with good rear wave damping against enclosure measurements at the same mic position. If you use a low cost electret microphone, be on the look out for overload of the capsule with nearfield measurements; the polarizing voltage is low, and amplifier headroom not that great, compared with the instrumentation microphones with 200V and relatively high voltage electronics. And let us know what you find out!

[davey_m]

gmed, it's a popular myth that non-parallel walls will eliminate cavity resonances. Parallel walls give you predictable cavity resonance modes. Cavities with complex shapes generally just make the cavity modes hard to predict, because they are complex.

The exception is acoustically small enclosures - where cavities are small and cavity modes don't exist inside the bandwidth covered by the driver. But it is usually not a practical proposition.

[Dennis H]

Bottom line, you need to glue some foam or fiberglass or whatever to the walls to effectively deal with resonances. And that's going to be a lot easier to do if the walls are smooth.