Page 191 - 3D Fibre Reinforced Polymer Composites

180 30 Fibre Reinforced Polymer Composites
A simple model for estimating the compression fatigue life of stitched composite has
been proposed by Mouritz and Cox (2000). They observed that the compression S-N
curves for stitched and unstitched composites have slopes that are essentially
indistinguishable beneath the experimental scatter, as seen for example in Figure 8.12.
The only significant difference between the S-N curves for stitched and unstitched
composites is the initial knock-down in static compression strength suffered by a
stitched material. As shown in Section 8.3.3, the knock-down in compression strength
due to stitching is usually under 20%. Based on this difference, Mouritz and Cox
(2000) propose that the S-N curve for a stitched composite subject to compression
fatigue can be estimated using Basquin’s law:

s =a, -mlog,, N

where S is the maximum applied compressive fatigue stress, N is the number of load
cycles, a, is the static compressive strength of the stitched composite, and -m is the
slope of the S-N for the unstitched laminate. Figure 8.13 shows fatigue life data for
two stitched composites determined experimentally by Portanvona et al. (1992) and
Vandermey et al. (1991). The solid S-N curves in Figure 8.14 show the theoretical
fatigue life determined using equation (8.1). It is seen that the fatigue curve of stitched
composites can be accurately predicted using the model. An appealing feature of the
model is its simplicity; the compression S-N curve for a stitched composite can be
determined from two simple tests: (1) a static compression test on .the stitched
composite to measure the compressive strength, a,, and (2) a compression-compression
fatigue test on the unstitched composite to determine the slope of the S-N curve, -m.
Stitching can also degrade the tension-tension fatigue resistance of composites
(Aymerich et al., 2001; Herszberg et al., 1997; Shah Khan and Mouritz, 1996, 1997).
For example, Figure 8.14 compares fatigue-life curves for an unstitched and stitched
composite subject to zero-to-tension fatigue loading (Shah Khan and Mouritz, 1997).
An examination of the fatigue damage mechanisms of the composites shown in Figure
8.14 reveals that the stitched laminate started showing evidence of fatigue-induced
damage close to the stitches at a relatively low number of load cycles. It is believed that
the distortion and clusters of broken fibres caused by stitching act as sites for the early
growth of fatigue-induced damage that ultimately leads to complete failure of the
stitched composite. Aymerich et al. (2001) have found, however, that the tensile fatigue
performance is only degraded in fibre dominated composites, such as with a [O], or
[*45/0/90], stacking sequence. The fatigue performance of matrix dominated
composites (eg. [+30/90]s) is improved by stitching because the threads are effective in
arresting or delaying the delamination crack growth under tensile fatigue loading.
Despite the knowledge of the fatigue performance of stitched composites in
compression-compression and tension-tension loading, further research into fatigue is
needed. The conditions under which stitching is beneficial or detrimental to the tensile
fatigue endurance of composites still needs to be resolved. The effects of the various
fatigue conditions (eg. R-ratio, load frequency) and stitching conditions (eg. yarn
thickness, stitch density) on the fatigue endurance and fatigue damage mechanisms of
the main engineering composites, particularly carbodepoxy, is required. Research into
the fatigue performance of stitched composites subject to reversed (compression-
tension) cyclic loading is also needed.