A method based on acoustic emission (AE) generated by fibre fractures
during tensile test has been developed for the determination of the st
rength distribution of E-glass fibre strands containing as many as 400
0 fibres (filaments). This technique is novel in that it allows fast d
ata acquisition and provides accurate data for the load, location and
time of fracture of each filament. A range of statistical data includi
ng the fracture strength and AE signal parameters can be extracted to
characterise the strand of fibres under investigation, Software has be
en developed for the evaluation of the large number of collected data
and for the graphical presentation of the AE signal parameters. The fi
nal test package thus allows automatic testing and provides the analys
t with an overall picture of the real strength distribution of the fib
res in a strand, The test and analysis for 4000 fibres is completed wi
thin 75 min, Complete strand tests of 4000 filaments using this method
have revealed greater detail of the characteristic flaw distribution
in a strand that was not possible with tests conducted on single filam
ents or bundles containing small numbers of filaments (see Ref. [?]).
The new study has provided concrete evidence of the existence of a bim
odal flaw population rather than the unimodal population that has been
previously reported. (C) 1997 Elsevier Science Ltd. All rights reserv
ed.