Aplicação de decomposição em ondas planas na medição da absorção in situ e isotropia

Abstract

The measurement of the sound absorption coefficient is an important process for room acoustics, as the results they provide are used by engineers and architects for projects, simulations, specifications and such. In this context, the measured values must be coherent with the behaviour displayed by the material in real application conditions. The current ISO 354 standard that reigns over diffuse incidence absorption coefficient measurement relies on the Sabine equation, which has as a premise the existence of a diffuse sound field in the reverberation chamber. The diffuseness condition of the sound field is influenced by factors such as the enclosure’s shape and size and the presence of diffusing elements and can be challenging to determine exactly. A number of techniques were developed to achieve a more trustworthy quantification of sound absorption. This work aimed to employ the process known as plane wave decomposition to measure sound absorption of an absorbing surface in situ and quantify the diffuseness inside a reverberation room under different absorption conditions. The in situ absorption measurement presented good results, comparable to values calculated from the flow resistivity measured in a laboratory, with an error that appears to be predominantly systematic. Moreover, the sound field reconstruction quality was assessed for different amounts of sampling in the array. In general, samplings with more points present reconstructions with more minor errors. The diffuseness of the reverberation chamber was quantified for four different absorption configurations, including the empty enclosure. The diffuseness indicator presents higher values for the empty chamber than for the cases with absorption but does not display any significant variation between the cases with different absorber distributions. However, the effect of the distribution of the absorber is noted in the energy decay curves and on the absorption values calculated by the Sabine formula. Aside from that, the reconstruction of the sound field was analysed. The reconstructions for positions inside the initial random sampling volume displayed smaller errors than those outside the volume. Moreover, the differences in using the 1-norm or the 2-norm in the isotropy calculation were also observed, with the 2-norm values displaying bigger values due to the rotational invariance of this norm.