This paper discusses a comprehensive study that is developing a quantitativ
e understanding of the acoustic emission (AE) signals that emanate from fat
igue cracks. Two critical components of this study are the development of a
transfer function that quantifies and removes geometric effects from a mea
sured AE waveform and an experimental program that monitors and identifies
AE signals that occur during the fatigue of cylindrical stainless steel spe
cimens under torsion. Typical waveforms are collected during torsional fati
gue and correlated with fracture mechanisms from different stages of testin
g. Three stages of fatigue are identified by AE waveform characterization a
nd confirmed by microscopic replica observation. The other portion of this
study demonstrates the effectiveness of using laser ultrasonic techniques t
o develop transfer functions to quantify and remove geometric effects from
measured acoustic emission waveforms.