Glass fibre strength distribution determined by common experimental methods

Abstract

The tensile strength of brittle fibres is routinely described by the Weibull distribution. The parameters of the distribution can be obtained from tests on single fibres and fibre bundles or from model composite tests. However, there is growing evidence that the distribution parameters obtained by different experimental techniques differ systematically. In order to investigate the possible causes of such discrepancies, single-fibre tension, fibre bundle, and single-fibre fragmentation tests are employed in this study to obtain strength distribution of commercial E-glass fibres. The results reveal parameter dependence on the approach used to extract the distribution parameters from experimental data. Particularly, in the case of single-fibre tension tests, the shape parameter obtained from average fibre strength vs. length data is larger than that obtained at a fixed gauge length. It is assumed that the apparent fibre strength scatter is caused by both the inherent flaw structure along a fibre and by the fibre-to-fibre strength variability within a batch, due to slightly differing processing and handling history of the fibres. Fibre fragmentation test results are used to derive the Weibull distribution parameters applicable to the fibre batch. The strength distribution obtained is compared with strength data for the single fibres, and reasonably good agreement is observed.