LIBRARY OF GEOMETRIC INFLUENCES FOR SIF WEIGHT FUNCTIONS F. P. BRENNAN, L. S. TEH and A. J. LOVE NDE Centre, Department of Mechanical Engineering, University College London, UK ABSTRACT Most researchers agree that the SIF is an important if not vital parameter in the assessment of defects prone to linear elastic fracture behaviour. Unfortunately the SIF is difficult to compute or measure, particularly if the crack is situated in a complex three-dimensional geometry or subjected to a non-simple stress state. Enormous developments in computing processing ability in the past few years has led to exciting Finite Element and Boundary Element approaches to the solution of SIFs in complex situations. These however, are likely to always remain the preserve of specialists. The need to provide quick robust and accurate SIF solutions for engineering defect assessment led to the development of a weight function approach to compose complex SIF solutions from simple constituent parts. These simple constituents make up a library that can be used with an appropriate composition algorithm to quickly build solutions for cracks in real engineering geometries. This paper describes the form of the “Library of Geometric Influences” for two dimensional symmetrical notch SIF Weight Functions. Future development of the approach to extend it to asymmetrical and three dimensional geometries is also discussed. 1 INTRODUCTION The SIF is in effect the currency of Linear Elastic Fracture Mechanics and is generally agreed to be the key parameter in the assessment of defects prone to linear elastic fracture behaviour. SIF solutions for many geometries and load conditions exist to varying degrees of validity and accuracy. Some are presented in the form of parametric equations others as figures and tables. Invariably solutions are coded into software combining some quite accurate and reliable solutions with others which are not so useful. The current practice for determination of SIFs is generally to use numerical or experimental techniques, both of which are time consuming and require great care and attention. Exciting developments in Finite Element and Boundary Element approaches have allowed extremely complex aspects of crack behaviour to be studied. The use of these methods are however, likely to always remain the preserve of specialists. It also means that available solutions have generally been developed for specific applications and therefore are generally applicable within often quite restrictive limits of validity. Engineering optimisation and defect assessment of components in service however, often require broad ranging solutions which can be rapidly calculated. Although the approach reported here still requires such reference constituent solutions, their number is limited and can be used to generate an unprecedented number of new solutions. The real value of this approach is that it allows the rapid formulation of SIF solutions in complex geometries under any stress field (including residual and applied stresses).