Frequency Dependence of the b-Value Used for Acoustic Emission Analysis of Glass Fiber Reinforced Plastics

Abstract

Acoustic Emission Testing (AT) is one of the major non-destructive testing methods used for severity evaluation of structures. Amplitude distributions of AE signals are characterized by b-value and the value is mainly used for the severity evaluation of concrete structures until now. The value is assumed to be independent with propagation distance between acoustic emission sources to AE sensors. We evaluate the influence of the wide frequency band encountered in the fracture behavior of glass fiber reinforced plastic (GFRP) on the b-value analysis. In tensile tests, the b-value was determined from an acoustic emission (AE) source generated near a centered hole in a specimen of GFRP. At 15 mm from the hole, the b-value analysis indicated a decreasing trend with increasing tensile stress. At a propagation length of 45 mm, farthest from the hole, a small number of AE signals were received. The attenuation is more rapid for high-frequency AE signals. Thus, the amplitude distribution bandwidth is wide and the b-value changes. This change in b-value for GFRPs is investigated by analyzing the spectral components of the AE signals. For a single-frequency AE source, the b-value is unchanged with propagation length. In contrast, multiple-frequency AE sources produce changes in b-value proportional to the fraction of each spectral component in the received signal. This is due to the frequency dependence of the attenuation with propagation length. From these results, the b-value analysis cannot be applied to considering frequency dependence of AE attenuation.