A study of diffusion in reverberation chambers provided with special devices

Abstract

With a specially designed, experimental test procedure some common and important ways of achieving sound field diffusion have been investigated. The test procedure is based on a measure which quantitatively estimates the degree of diffusion. When comparing the outcome of the experiments with applicable theories and experimental results reported in the literature, the measure has proved to behave logically. The use of different loudspeaker arrangements, additional absorbents, hanging panels and movable reflectors have been studied. In summary, the use of absorbents has affected the diffusion negatively at high frequencies; stationary hanging panels have improved the diffusion of an empty hardwalled chamber only in a small chamber and at 1250 Hz; some specific loudspeaker arrangements have given considerably better values, and extraordinary but statistically expectable improvements have been attained by a specially constructed movable boundary system. The movable boundary system is based on the same idea as a rotating diffuser, but the reflector locations are discretely altered and the one-sided reflective area is comparatively large (58 m 2). A procedure describing how to search for effective combinations of boundary configurations and evaluate the effectiveness of the associated room averaging is outlined. In applying the procedure to a 125 Hz situation, a suitable set of five boundary configurations was found. Upon performing the associated room averaging, the spatial variance of the mean square pressure was found to have decreased by a factor of 0·14–0·44 and the diffusivity values decreased to a level typical for the empty chamber at 500 Hz. The various treatments investigated have also proved to affect significantly the outcome of some standard measurements. This fact indicates that these matters should be considered in the development of future standards.