Page 91 - Handbook of Properties of Textile and Technical Fibres

72 Handbook of Properties of Textile and Technical Fibres

(Yang et al., 1996), even though the bundle test has been designed to straighten all the
fibers prior to clamping. In spite of these shortcomings bundle strength measurements
are nevertheless still very useful for comparative studies, e.g., the effect of dyeing time
and temperature on the strength of wool.
The relationship between average single fiber strength and bundle strength has been
studied by a number of workers, and models have been developed in an attempt to pre-
dict tensile properties of bundles from single fiber properties and vice versa (Platt et al.,
1952; Nachane and Iyer, 1980; Sasser et al., 1991; Frydrych, 1995; Huson and Turner,
2001; Yu et al., 2003). Huson and Maxwell (2004) used experimental single fiber
stressestrain curves to simulate a bundle test and showed that bundle tenacity is highly
dependent on the variations in strain at break (as measured by the coefficient of vari-
ation, CV SB ) of the individual fibers in the bundle as well as the variation in gauge
lengths (CV GL ) within the fiber bundle (Fig. 3.12). The CV GL is a measure of how
well the bundle has been prepared. For the particular wool fibers studied, with a vari-
ation in strain at break of 16.7%, this resulted in a decrease in tenacity relative to
average fiber tenacity of about 25% for perfectly prepared bundles (CV GL ¼ 0). The
effect of a less than perfectly prepared bundle with a broad distribution of fiber lengths
(increased CV GL ) was a further small drop in tenacity. Results from a range of individ-
ual sheep show that CV SB is quite varied, typically varying from 15% to 30% for sheep
on a live weight maintenance diet. If this variability is also present in samples of top
then it is not expected that bundle tenacity would be a good measure of average
intrinsic fiber strength.
As expected, the initial slope of the bundle stressestrain curve was insensitive to
variation in individual fiber breaking strain but highly dependent on gauge length dis-
tribution (Fig. 3.13). The important point to note is that preparing a bundle of fibers
consistently is not easy, thus CV GL will vary making the initial slope a very poor

15

Tenacity (cN/tex) 13

11

9
0 2 4 6 8 10
CV GL (%)
Figure 3.12 Effect of gauge length distribution (CV GL ) on the tenacity of simulated bundle
tensile curves using experimental forceeextension data from single fibers. Results are shown
for a typical distribution of strain at break values, CV SB ¼ 16.7% (,) and for CV SB ¼ 0% (-)
(Huson and Maxwell, 2004).