ASTM Committee E08 Task Group E08.08.02 has conducted a round robin to evaluate two new proposed Annexes to Standard E 1820. The round robin involved testing compact and three-point bend specimens at elevated test rates using servohydraulic test machines and then analyzing the resulting data using the normalization method proposed by Landes et al. The test rates were limited so that essentially static (noninertial) analysis could be applied to the results with the objective of obtaining J-integral resistance curves (J-R curves) consistent with those obtained by static tests presently sanctioned by E 1820. The normalization method fits an assumed form of function to the normalized load versus displacement relationship of each specimen and provides a technique to estimate the crack length at each data point. The so-called “normalization function” used is fit to data before crack growth initiates and to one point at the end of the test, for which the final measured crack length is assumed to be available. In this sense it acts as an interpolation scheme to estimate crack lengths and hence crack extensions for use in evaluating the specimen J-R curve.
Five laboratories participated in the round robin. Each laboratory was provided with two to four IT C(T) and/or SE(B) specimens of an A572 bridge steel for testing. Each laboratory was responsible for precracking and testing the specimens, loading the specimens at elevated loading rates between approximately 25 mm/s and 2.5 m/s, collecting the resulting load and displacement data, and analyzing the data according to the proposed normalization annex document prepared by Task Group E08.08.02. Specimen blanks were distributed during July 1998, and tests were run between late 1998 and late 2000 and the data were subsequently analyzed and the results returned. This paper describes the results of these individual analyses and an additional consistent analysis of the data sets completed by the author of this paper.
The objectives of the round robin were to determine if the high rate annex and the normalization method provided accurate and reasonable procedures for obtaining high rate loading J-R curves for structural materials. Since only one material could be tested in the round robin it was hoped that the participants, and possibly others as well, would use the procedures presented to analyze other materials for which they had data and interest. The results of the round robin demonstrated that the method does give good results for the test material, and that all participants were able to apply the method and obtain reasonable results. The author will discuss several weaknesses of the proposed Annexes that were uncovered in the course of the round robin or by other researchers during the past two years, and proposed corrections are presented that hopefully will address these shortcomings.