This work describes the development of two types of three-dimensional (3D) finite element models to predict stable, Mode I crack growth in thin, ductile aluminum alloys, The two presented models extend the standard 2D form of the Crack Tip Opening Angle (CTOA) methodology, which determines crack extension based on obtaining a critical angle at the crack tip. The more general 3D model evaluates the CTOA at each node along the crack front which enables the development of tunneled profiles. The alternative, constant front approach, enforces uniform growth along the crack front, thereby growing the crack in a self-similar manner. For the constant front approach, evaluation of the CTOA occurs at a specified distance behind the crack front which decouples CTOA evaluation from mesh refinement. Both CTOA-based models include adaptive load control strategies to minimize the effects of discrete Load increments on the growth response, Example analyses demonstrate that the more general 3D approach requires cube-shaped elements on the crack plane to eliminate a bias in growth directions, To evaluate the effectiveness of the constant front approach, this work also describes a validation study using loadcrack extension data from 2.3 mm thick Al 2024-T3 specimens tested at NASA-Langley. The test matrix includes C(T) and M(T) specimens, with varying widths (50-600 mm), a/W ratios, and levels of mechanical restraint to suppress out-of-plane bending, Comparisons of load-crack extension curves from experiments and analyses of a 150 mm C(T) specimen, with out-of-plane bending prevented, provide a calibrated (critical) CTOA value of 5.1 degrees, Analyses using the calibrated CTOA value and the constant front approach provide predictions of peak load for constrained and unconstrained specimens in good agreement with the experimental values, (C) 1999 Elsevier Science Ltd, All rights reserved.
