SYMPOSIA PAPER Published: 01 January 1994
STP13719S

A New Application of Normalization: Developing Curves from Displacement Versus Crack Length and from Displacement Alone

Source

Two new experimental approaches are presented in this study to develop J-R curves: (1) from displacement versus crack length without load measurement and (2) from displacement alone without load and crack length measurement. The approaches are new applications of the method of normalization, which is based on the principle of load separation suggested by Ernst et al. The latest version of the method of normalization introduced by Landes et al. uses the three-parameter LMN functional form to relate the three variables, load, displacement, and crack length. Given any two, the third can be determined. Previously, the method of normalization has been successfully used to develop J-R curves from load versus displacement without measuring crack length and from load versus crack length without measuring displacement. The new approaches can be applied to develop J-R curves for the crack-line wedge-loaded specimen, C(W), for which the load cannot be measured.

This study was conducted in two steps. In the first step, the method of normalization was used for developing J-R curves from displacement versus crack length without load measurement. A power law fit was used to determine some calibration points at the initial part of the deformation curve. They are needed for evaluating the LMN function. In the second step, normalization was used to develop J-R curves from displacement only, without load and crack length measurements. Two new equations that relate load, displacement, and crack length were proposed in this study to use with the method of normalization for predicting the load and crack length from displacement measurements alone.

The approaches developed in both steps have been applied to analyze a number of test results from compact specimens for which all of three variables, load, displacement, and crack length, have already been determined. Five different steels were included. For the first approach the test records were assumed to have only the displacement and crack length measurements, while for applying the second approach, the test records were assumed to have only a displacement measurement. The results show that both new approaches worked very well. Compared with the test data, the predicted error for the load measurement was found within 5% for the first approach and 10% for the second approach. Compared with both the standard compliance method and the regular normalization method, the difference between J-R curves developed was found within 5% for both new approaches.

Author Information

Lee, K
The University of Tennessee, Knoxville, Knoxville, TN
Landes, JD
The University of Tennessee, Knoxville, Knoxville, TN
Price: $25.00
Contact Sales
Related
Reprints and Permissions
Reprints and copyright permissions can be requested through the
Copyright Clearance Center
Details
Developed by Committee: E08
Pages: 422–446
DOI: 10.1520/STP13719S
ISBN-EB: 978-0-8031-5270-0
ISBN-13: 978-0-8031-1990-1