LAUSR

Abstract The measurement of the sound absorption coefficient is clearly described in the ISO 10534 series of standards as well as in the ISO 354:2003 standard. However, the methods described in these standards are intended for laboratory testing under certain controlled conditions. There are situations where measurements have to be carried out in situ, as the material to be characterized cannot be transferred to the laboratory. To overcome this issue, the impedance gun, a device comprising a small spherical loudspeaker and a PU probe, can be very useful. However, several questions regarding the accuracy of this method remain open. The purpose of this paper is to determine, quantitatively and statistically, the accuracy of this in situ measurement system. To this end, an experiment was carried out in which the sound absorption of a sample of rock wool was measured. The sound absorption coefficient measurement of the specimen was carried out in several environments, at different specimen – impedance gun distances and using different calculation models (i.e. mirror source model (M), plane wave model (P) and Q-term model (Q)). In this way, the effect of each of these factors on the accuracy of the measurement could be evaluated. In addition, the results of these in situ measurements were compared with those of an impedance tube intercomparison for the same material sample. The evaluation of accuracy was performed by adapting the method described in the ISO 5725 series of standards, regarding the precision of measurements in terms of repeatability and reproducibility, to this case study. The main findings were that significant differences were found between the different calculation models and, therefore, cannot be considered equivalent. In addition, some calculation models were more sensitive to the influence of other factors such as specimen - impedance gun distance and the measuring environment. Finally, it was found that in situ measurement methods provide comparable results to impedance tube measurements for frequency bands above 500 Hz.