Non-destructive reverberant testing of natural fibrous samples in a diffused acoustic field environment

Abstract

A hexahedron reverberation box was utilized to measure the reverberation time of samples in a diffuse acoustic field. The aim of this study is to design a mathematical model that accurately represents the correlation between frequency and absorption coefficient for analysis. The same was obtained by aligning the noise absorption coefficients (NACs) determined through an impedance tube and calculated based on the reverberation time. The analysis indicated that jute fiber exhibits superior performance compared to other natural fibers, such as rock and glass wool, in the attenuation of sound frequencies above 2200 Hz. The results motivated further investigation over jute nonwoven fabric to conduct layer analysis (1–7) and to examine the impact of thickness. Thereby, a dataset was compiled consisting of 11 fibrous samples (nine natural and two commercial fibers) and jute nonwoven fabric. The empirical model was developed regardless of the type of fiber or thickness of the nonwoven fabric, and it was successfully validated for three additional fibers (banana, pineapple, and ramie). The predictive model exhibited a high level of accuracy in estimating the NAC, displaying a strong similarity to that of impedance tube measurements. The achieved mean absolute error ranges for the predictions are between 0.02 and 0.03 only. The main discovery of this study revolves around the recognition of frequency of sound as a crucial variable and its application in predicting the NAC for the samples.