Eric Weitzman, May 18, 2006
I recently became aware of a theory that says dipole speakers are incapable of producing deep bass in small rooms. Essentially, the theory asserts that below the resonant frequency defined by a room's lowest modal frequency corresponding to its longest dimension, the dipole cannot pressurize the room and so it cannot produce any bass below this frequency.
For example, in the article "Room Response with Monopole, Dipole, and Cardioid Woofers, Part II: Examination of the in room response" at http://www.musicanddesign.com/roomgain2.html, John Kreskovsky writes that:
[A]s the frequency drops below the fundamental the sound pressure becomes more and more a function [of] the ability of the woofer to pressurize the room, and since a dipole is incapable of pressurization, the response of dipole woofers drops off precipitously below the fundamental, regardless of the free field response. If a dipole woofer system with a corner frequency of 20 Hz is placed in a room with fundamental frequency of 40 Hz, the response will drop off below 40 Hz. Very little can be done about this situation. ... In effect, where as [sic] monopole and cardioid woofers are augmented by room pressurization, the dipole response is nulled. For this reason, dipoles [sic] woofers do not tend to become bloated or boomy when placed in smaller rooms, but they are incapable of producing low bass response in such rooms as well.
Intuitively, this theory seemed flawed to me. A dipole cannot pressurize a room at any frequency because the part of the room that is displaced in front of the speaker is exactly cancelled out the the displacement behind the speaker. If this prevents the dipole from producing sound below any arbitrary frequency, why should the dipole be able to produce sound at any frequency at all?
I don't have knowledge or ability to derive a counter argument from fundamentals of acoustics, nor the insight to find any problems in the choice of accepted theory or the derivations that purport to prove this theory. But, I can test the theory in practice. The results of my experiment testing the theory follow.
A dipole speaker's bass response will drop precipitously below the fundamental resonant frequency in a small room.
A dipole speaker was placed in a small room. The room's dimensions are 3.75' by 5.5' by 7.5'. The room's construction is drywall on wooden studs. Approximately 75% of the wall surface is covered with porcelain tile, the rest being either painted or wallpapered. The floor is tile on concrete with a 0.25' step separating the floor into two equal areas. A covered immovable porcelain water-bearing cistern is located in the center of the higher of the two floor areas. The two sliding glass doors that divided the room were removed for the experiment. A small single-pane window, a standard hollow core door, and a small ventilation fan provide pathways for sound leakage.
The speaker was placed so its axis was aligned with the long side of the room. That is, the axis was aligned with the 5.5' dimension and the plane of the baffle was aligned with the 3.75' dimension. The speaker was place approximately 1.5' from the walls on either side of it and 2' from the wall in front.
Two types of test tones were played. The first test tone was a frequency sweep from 200Hz to 20Hz. The sweep is slow enough that the tones would energize room modes in average sized listening rooms whose RT60 is about 500ms. The RT60 of the test room is probably much larger than 500ms due to the highly reflective surfaces and lack of absorbing materials, so the sweep may not fully energize all room resonances. The second set of tones were shaped tone bursts at the following frequencies: 200, 180, 160, 140, 120, 110, 100, 90, 80, 70, 60, 55, 50, 45, 40, 35, 30, 25 and 20Hz. Details regarding these test tones can be found at http://www.linkwitzlab.com/burst-cd.htm.
After the speaker was connected to the appropriate electronic equipment, the primary investigator entered the room with a Radio Shack model 33-2050 analog SPL meter, a pad of paper, and a writing utensil. An assistant outside the room monitored the CD player as the sweeps were played and announced the time in a loud voice as each second passed. The assistant also controlled the CD so it played the tone burst tracks as requested by the investigator.
The SPL meter was set to the "fast" position, C-weighting. It was moved to various positions above and around the porcelain cistern, and in front of the speaker. No systematic difference in SPL readings were noted at any particular location. The measurements reported here were made at a position that was most comfortable to the investigator, about 1' from the wall behind the speaker and 3.5' above the floor.
During the sweeps, the investigator wrote down the SPL level beside the corresponding second on the pad. It took several repetitions of the sweep to get a reading at each second mark since the range dial of the meter had to be adjusted frequently. The frequency was determined from the elapsed time into the sweep playback.
The tone bursts tracks contain a voice-over announcing the frequency of each ensuing tone burst. The investigator wrote down the maximum SPL read during each tone burst adjacent to the frequency on the pad of paper.
The following chart shows the measured sound pressure levels at each frequency and the levels after correcting for the inaccuracies of the meter. The amplitude response for the two stimuli is very similar. The measurements were corrected for the known low-frequency under-reporting of the meter using the following corrections: 100Hz/0db, 60Hz/+1db, 40Hz/+2db, 30Hz/+4db, 20Hz/+9db. Both raw data and corrected data are shown.
The longest dimension in the small room where the test was performed is 7.5'. Since the first (lowest) resonance along an axis is energized by a frequency whose wavelength is twice the length of the axis, the room's lowest modal wavelength is 15'. This corresponds to a frequency, or fundamental resonance, of 75Hz. If the hypothesis is correct, the SPL should drop off precipitously below 75Hz, and it would be measurable. (Other room modes are calculated at 102, 127, 150(2), 168, 182(2), 197, and 205Hz.)
There is significant SPL below the room's fundamental resonance at 75Hz. The level is fairly constant from about 70 Hz on down. The levels below resonance were around the same as those between 120-160 Hz. None of the predicted drop off was measured. The hypothesis that dipole bass output drops precipitously below a room's fundamental resonance is false.
Copyright © 2006, Eric Weitzman. All rights reserved. [eric at acm dot org]