18 TH INTERNATIONAL CONFERENCE ON COMPOSITE MATERIALS DAMAGE CHARACTERISATION AND THE ROLE OF VOIDS IN THE FATIGUE OF WIND TURBINE BLADE MATERIALS J. Lambert 1 *, A.R. Chambers 2, I. Sinclair 3 , S.M. Spearing 4 1-4 School of Engineering Sciences, University of Southampton, Southampton, UK Keywords : GFRP, void, fatigue, computed tomography, fully-reversed, tension-compression 1 Introduction specimens that had sustained different fractions of their estimated lives, an idea of damage sequence The technological development of wind turbines and interaction could also be obtained. and their more widespread use represent an important method towards meeting the growing worldwide energy demand. For increased power 2 Literature generation and a greater efficiency, the general Composites are known to exhibit four main fatigue trend is for larger turbines with an increased blade mechanisms; fibre fracture, fibre/matrix debonding, diameter. Individual blade lengths are currently matrix cracking and delamination. [1] It has been approaching 60m; a significant increase from found that the different mechanisms can dominate around 20m 15 years ago. This emergence of much at different stress levels. At low load levels, matrix longer blades has resulted in an absence of relevant cracking is commonly a predominant mechanism. long-term in-service data. With expected lifetimes At medium loads a combination of matrix cracking of 20 years, the long-term integrity of the blade and interface debonding is observed, while at high material has thus become an important area of loads fibre failures may occur. [1] The low load, high research. As a result, extensive fatigue databases for cycle fatigue nature of the wind turbine is thus wind turbine composites have been compiled in expected to promote matrix cracking. Although recent years. The non-transferability of these matrix cracking has been widely researched, most results between different materials and load work has concentrated on cross-ply laminates, and conditions however implies a large experimental thus transverse (90º) matrix cracking. [2][3] These effort. To reduce this, a greater understanding of the cracks have been noted to grow across the specimen fatigue damage modes and their accumulation is width from the free edge, with their growth rate required. In addition to this, further insight into dependent on load-direction spacing rather than certain microstructural features, such as voids, is their own length. [4][5] required, as their role in fatigue damage accumulation remains uncertain. Accordingly, the Tong et al. [6] have previously investigated crack aim of the present work is to gain insight into the development in [0/90/-45/+45] s glass fibre-epoxy micromechanical damage processes of a wind composite in static tension and tension-tension (T- turbine blade composite material. In order to T) fatigue. Transverse ply cracking appeared first, achieve this, a suitable experimental technique was followed by -45° cracks from the edges of existing needed for imaging and analysis. Computed 90° cracks, and finally +45° matrix cracks. Masters tomography (CT) has been used to this end in the and Reifsnider [7] studied the fatigue crack growth current work, enabling full 3D imaging of a of different quasi-isotropic laminates, confirming sample’s internal volume. The first goal was the the presence of 90° ply cracks that spread to form observation and quantification of microstructural cracks in neighbouring off-axis plies. Two different detail, notably voids, through scanning untested lay-ups were considered, [0/90/±45] s and material. The second main objective was the [0/±45/90] s . Each displayed distinctly different scanning of fatigue-tested coupons to identify the crack saturation patterns, despite initial damage of damage mechanisms occurring. Through scanning the transverse plies in both. This highlights the
dependence on laminate lay-up for damage composite. Both static and flexural fatigue evolution. Gamby et al. [8] have reported on the performance were found to deteriorate with development of cracks from free edges in laminates increasing void content, however the authors argue containing 0° and ±45° plies, noting that the matrix that this global percentage view is too simplistic cracks propagated from the free edges of the off and that factors such as void size, shape, and axis layers, and, as with the transverse cracks distribution must be considered. Mandell and Tsai [18] also concluded that the overall void content previously mentioned, were dependent not on their own length but only on the number of cycles and is an inadequate representation of potential the distance from the free edge. michromechanical influence, and that void size and shape must be considered. A critical void size was Matrix cracking alone does not usually cause proposed, above which fracture mechanics based fatigue failure, however it is often a precursor to crack growth could be used to assess void more critical mechanisms. Matrix cracks are often influence, while at lower sizes voids simply reported to be responsible for delamination onset, contributes to a reduction in the cross-sectional area due to the local stress concentrations at their tips. [9] of the specimen. Overall there is a noticeable Delamination is also known to occur at laminate absence of literature on the effect of voids on edges where interlaminar stresses are highest, due mechanical performance, especially when to the Poisson’s ratio mismatch between differently concerned with fatigue. Moreover, the studies that angled plies. O’Brien [10] experimentally compared do investigate fatigue are usually limited to the difference between local delaminations unidirectional carbon composites, and the effect of occurring within the specimen bulk (associated with voids in multidirectional laminates remains largely a matrix crack), and edge delaminations. The edge unknown. delaminations were found to propagate in a stable manner, whereas local delaminations from matrix cracks created stress concentrations that led to 3 Experimental Detail premature laminate failure strains below that of the 150x25x7mm prepreg glass-epoxy specimens, primary load bearing plies. representative of the material used in wind turbine blade surfaces, with a gauge length of 28mm and a Voids are present in composites either due to the layup of [0/+45/-45] 3s were studied. In addition to manufacturing process, that leaves air trapped in the these directional layers a “fleece” resin rich layer laminate during the curing cycle, or by nucleation containing randomly orientated fibres, used to from volatiles during processing. [11] Critical void improve the surface finish of the composite, was levels below which mechanical performance isn’t present adjacent to each 3-directional fibre ply. affected have been predicted as between 1-4% in 2mm thick aluminium was used for end tabs, various studies. [12]-[14] Whether void content is more adhesively bonded using Araldite 2011 and post- of a concern to static or fatigue properties is an area cured for 30 minutes at 80ºC. Prior to any fatigue of disagreement, with some studies concluding that testing, 9 specimens were scanned using microfocus static performance is largely unaffected by voids, X-ray computed tomography ( µ CT) to enable the [15] while others report the same order in the quantitative void analysis to be performed and reduction of both static and fatigue performance. [16] eventually correlated with their fatigue lives. A Studies explicitly relating to fatigue include that of 25x25x6.5mm volume within the gauge length of Prakash, [17] who performed axial fully-reversed the material was scanned using an XTek™ ( R =-1) tests on unidirectional carbon fibre Benchtop 160i µ CT system. A voxel resolution of composite. It was hypothesised that voids, along 25 µ m was achieved, implying the reasonably direct with other damage, contributed indirectly to detection of voids >40 µ m ( i.e being of the order of reduced fatigue performance through their poor 2 or more voxels across). Using the reconstructed heat dissipation properties. No quantitative analysis µ CT volume as shown in Fig. 1, the voids were [16] was however performed. Chambers et al. segmented using their greyscale histogram peak, performed a controlled analysis; varying the void and the rest of the bulk volume was removed. content through set changes in the cure cycle of the
Recommend
More recommend
