A Critical Study of the Precision of Measurement of Absorption Coefficients by Reverberation Methods

Abstract

This paper is the result of a careful comparative study of the coefficients obtained by various methods used for the measurement of absorption coefficients employing the equations of the reverberation theory. Both organ pipes and an electro-dynamic loud speaker were used as sources of sound. With the latter, both steady tones and flutter tones were employed. Absorbing power was measured by determining the rate of decay from measurements of the reverberation time for known relative values of the steady state intensity, using the loud speaker source. The results indicate that a loud speaker operated at a fixed value of the input current sets up a steady state intensity that is constant, independently of the absorbing power of the room. The measurements with the organ pipes confirmed the assumption that the latter acts as a source of constant acoustical power output, setting up a steady state intensity that is inversely proportional to the total absorbing power of the room. Comparison of the results using steady and flutter tones shows that with a frequency range of a semitone above and below the mean frequency the measured values agree within the limits of experimental error. As between the organ pipe and the loud speaker, the former shows consistently higher values on four materials tested at the tone 512 vibrations per second, although the difference is only slightly greater than the experimental error. At other frequencies there is no consistent difference. The experiments showed that although total absorbing power scan be measured with a precision of the order of 2% or 3% yet errors of this magnitude, if cumulative, may result in variations of about 10% in the value of the absorption coefficient, computed from the difference in absorbing powers of the reverberation chamber with and without the test sample present. Comparison of results with result on identical samples tested in other laboratories show that at the frequency 512, the agreement between different laboratories is within the limits of accuracy of the methods so far developed.