Logo PTB

Uncertainty for the measurement of sound absorption in reverberation rooms


The quality of life experienced in rooms depends significantly on their acoustic properties. Acoustics has to fit the purpose of the room. Meeting rooms must enable the comprehension of spoken language, open-plan offices undisturbed work. Measures for appropriate room acoustics need the sound absorption of room surfaces and interior objects as a main input quantity.

Sound absorption is determined according to ISO 354 by reverberation time measurements in highly reflecting rooms with a volume of about 200 m³, so called reverberation rooms. The reduction of the reverberation time by installing the test specimen into the reverberation room can directly be converted to sound absorption. The measurement of absorption is prone to large uncertainties which cannot be described by a model today. The main reason is the diffuse sound field which is required for the measurement and which is characterised by a uniform sound incidence from all directions at all room points. Even if this can be ensured in the empty reverberation room, the diffusivity will be considerably disturbed by installing the test specimen. The currently valid ISO 354 of 2003 therefore includes the statement: "The reproducibility of the measurement of the absorption coefficient is currently under investigation". Today's test reports on measured sound absorption thus do not contain information on measurement uncertainty.

Nevertheless, measurement uncertainties are required for the accreditation of testing institutes or for the comparison of product properties. To gain realistic uncertainty estimates, a multitude of round robin tests was extensively analysed. Therefrom, an expression of the reproducibility standard deviation in one-third octave bands between 63 Hz and 6,3 kHz as a function of frequency and absorption coefficient was achieved (Figure 1). These values are the best estimate of measurement uncertainty available today.

Uncertainties for the weighted sound absorption coefficient and for the equivalent absorption area were estimated from the round robin data, too. It is planned to standardise these values in a future ISO 12999-2 to make them available for use in test reports.

Figure 1: Reproducibility standard deviation σR for the measurement of the absorption coefficients αs in reverberation rooms as a function of frequency f


[1] Volker Wittstock: Determination of measurement uncertainties in building acoustics by interlaboratory tests - Part 2: Sound absorption measured in reverberation rooms. Acta Acustica united with Acustica, in print


Volker Wittstock, FB 1.7, AG 1.72, e-mail: Opens window for sending emailvolker.wittstock(at)ptb.de