How to remove internal standing waves from speaker cabinets?
Last edited: Nov. 21, 2016
The most popular speaker cabinet design is the box shape (rectangular prism, cuboid). Unfortunately, due to the parallel sides these enclosure shapes suffer from internal standing waves. Some part of the standing waves may propagate through the speaker diaphragm, while some may cause excessive panel vibration. For high-fidelity sound reproduction the loudspeaker should not alter the sound and standing waves have to be eliminated. The amplitude of the standing waves can be reduced by choosing right box dimensions, by the proper position of the speaker, and by sound absorbing materials.
This article focuses on full range loudspeaker systems. In subwoofers due to the reduced frequency range the standing waves are much less of a problem.
Typical wrong placement of open cell sound absorbing foam in a speaker cabinet
How are standing waves created? In an empty cabinet most of the backward radiated sound is reflected from the cabinet walls, while the rest is absorbed or transmitted through the panels (of course this phenomena is frequency dependent). Because an untreated panel is a sound reflecting surface, it seems as if the mirror image of the speaker would appear in the same distance, but in the opposite direction from the wall. If there are two or more parallel sides, then it's not hard to find out what will happen... The mirror image of the image appears, the mirror image of image's image appears and so on... Standing waves are complex interference patterns caused by the original sound source and its' mirror images (reflected waves) at certain frequencies.
The internal standing wave resonances of a speaker cabinet can be calculated with Boxnotes. This is a Windows software and it takes into account the speaker's position (other calculators don't).
I was eager to know the importance of the amount and correct placement of sound absorbing materials. Since I was building a small two-way speaker that time, I did not have to build a separate test box. The inner dimensions of the enclosure that I used for the measurements: height is 34 cm, width is 19 cm and depth is 16 cm (at the loudspeaker the depth is 18cm, as it can be seen in the figure). The speaker is a 4 inch full range driver (with a very shallow cone), the frequency response is fairly linear up to 10 kHz, except there is a small "valley" at 2 kHz. The microphone distance is 10 cm (4 inch) and the measurement was taken on axis.
As with the majority of loudspeakers the strongest standing wave is formed between the speaker and the rear side of the cabinet. The wavelength of the fundamental is the double of the distance between the speaker and the rear wall, the harmonics are integer multiples of the fundamental. From both the measurement and the calculation the frequency of the fundamental is about 950 Hz.
Acoustic foam and polyester fiber fill ('wad')
And finally here you are the CSD (cumulative spectral decay or waterfall) diagrams of the impulse response measurements. I've measured the empty cabinet, the box filled with open cell sound absorbing foam, the box filled with polyester fiber fill (common wad lining) and open cell foam attached only to the side walls.
Empty enclosure: as a result of reflections from the rear wall strong standing waves build up, the decay of these resonances are the diagonal-parallel “mountain ranges”
Speaker cabinet filled with open cell polyester foam
Speaker cabinet filled with polyester fiber fill
Open cell foam in 5 cm (2 inch) thickness on the rear wall
It is obvious that sound absorbing material plays a big role in damping standing wave resonances. In order that the damping may be effective the damping material has to be placed in the middle of the enclosure where the air motion is the largest (maximum velocity). This means that the cabinet has to be filled almost completely with polyfill or other sound absorbing material, only the space around the speaker and the vent has to be left free. A common bad solution like sticking a very thin layer open cell foam on the rear wall has got minimal effect, so it is useless...