Reverberation-room qualification for determination of sound power of sources of discrete-frequency sound

Abstract

In practice, many sources of sound such as motors, pumps, fans, blowers, and transformers radiate sound having discrete frequency components. The reverberation room is a particularly convenient environment in which to determine the sound power produced by such sources provided that the measurement uncertainty can be reduced to an acceptable value. If the measurement uncertainty lies within standardized limits, the room is said to “qualify” for measurements on sources of discrete-frequency sound. The causes of the various uncertainties in the determination of sound power have been considered. It is shown that, in effect, the discrete-frequency room qualification procedure allows the measurement uncertainty to be estimated for a monopole-type sound source. The problems involved with discrete-frequency and swept-frequency qualification procedures are discussed, and experimental qualification data for several rooms with and without vanes and other accessories are presented. Methods of generating the frequencies necessary for the room qualification procedure are examined. From both theoretical and experimental studies it can be concluded that a major factor affecting the qualification of the reverberation room is the uncertainty in the measurement of the mean-square sound pressure throughout the room. This factor can be minimized by increasing the number of independent samples that one uses to estimate the mean-square pressure in the room, either by using a sufficiently long continuous microphone traverse or by using multiple microphones. At low frequencies (say, below 500 Hz in a 280-m3 reverberation room) the variability of the sound-power output with source position (change in radiation impedance) is also a major source of the uncertainty of the sound-power estimate if the sampling of the sound field has been accomplished adequately.