AN END-EFFECT MODEL FOR THE SINGLE-FILAMENT TENSILE TEST

Citation
Eg. Stoner et al., AN END-EFFECT MODEL FOR THE SINGLE-FILAMENT TENSILE TEST, Journal of Materials Science, 29(24), 1994, pp. 6561-6574
Citations number
14
Categorie Soggetti
Material Science
ISSN journal
00222461
Volume
29
Issue
24
Year of publication
1994
Pages
6561 - 6574
Database
ISI
SICI code
0022-2461(1994)29:24<6561:AEMFTS>2.0.ZU;2-7
Abstract
The effect of cross-sectional shape on tensile strength of pitch-based carbon fibres was investigated by extensive single-filament testing. For this study, round and trilobal pitch-based carbon fibres were prod uced at similar processing conditions. The application of a variety of distributions, including the simple Weibull distribution, to the stre ngth data indicated two sources of failure, one source being the accen tuation of end effects at short gauge lengths. A new mixed distributio n, the end-effect distribution, was proposed to account for these effe cts and applied to the experimental data. The end-effect model provide d an excellent description of the strength distributions of all fibres studied. The end-effect distribution is not complex and is based on s ound physical assumptions. It quantities a recognized inadequacy of th e test method which has not previously been accounted for, and it allo ws separation of end effects from the true fibre strength distribution . The results indicate that end effects can be an important concern fo r gauge lengths as long as 40 mm. Use of this model revealed that, in the absence of end effects, all fibres failed due to macroscopic flaws ; thus, varying the fibre geometry does not results in an unusual fail ure mechanism. However, the tensile strengths of the non-circular fibr e were found to be less dependent on fibre size. Thus, non-circular fi bres can be produced at higher mass flow rates, decreasing filament br eakage and increasing process conversions.