PNNL

Abstract

Resonant ultrasound spectroscopy (RUS) is employed as a nondestructive evaluation (NDE) tool in a number of metal/ceramic forming industries [1]. The presence of volumetric defects in an otherwise flaw-free object affects the resonance characteristics of the object. The changes in resonance behavior depend on the number, locations, volume, and material properties of the defects. Since the normal modes of an object depend on its geometry and the position-dependent material properties, namely the density and the complex elastic stiffness tensor, by accurate measurement of a specimen’s resonance frequencies and amplitudes, one can detect and characterize flaws embedded within it. A correct forward mathematical model to predict resonance characteristics is vital to the required analyses. In this paper, we present several computational results depicting the influence of the presence of embedded flaws/defects in a test specimen having simple geometrical shape. The mathematical model is based on the computationally efficient “XYZ Algorithm” of Visscher et al.