Composite forming mechanisms and materials characterisation 5 'locking angle'of a material.For successful tests,shear force can be calculated from the cross-head force using: F F:=2cos④ 1.1 where is the frame angle and Fpr is the measured axial picture frame force. Test data can be normalised by dividing the shear force by the length of the picture frame,Lpf.Graphs of shear force against shear angle can be produced, where the shear angle is defined as: 0=π/2-2Φ 1.2 The shear angle can be calculated from the cross-head displacement,dpf,by 0=-2cos-1 1.3 The picture frame test procedure is relatively simple to perform and results should be reasonably repeatable if sufficient care is taken in cutting test samples and the correct clamping technique is used.The major benefit of the test is that shear angle and angular shear rate can be easily calculated from the cross-head displacement and displacement rate. Typical results from picture frame experiments conducted on a range of textile reinforcements and prepregs are shown in Figs 1.2 and 1.3.For clarity these graphs represent single experiments,although it should be noted that even well controlled experiments exhibit scatter of up to +20%in shear force for a particular Most materials exhibit some similarities in terms of their shear force curve. Initially the shear resistance is low-for dry fabrics this represents dry friction at A tow crossovers,whilst for prepreg this corresponds to lubricated friction or viscous shear of polymer between fibres or yarns.Towards the end of the test the 菲in.比uet resistance increases significantly-this happens once adjacent yarns come into contact,representing yarn compaction(fabrics)or squeeze flow (prepreg).If the maximum possible deformation.Non-crimp fabrics (Fig.1.2b)exhibit more complex behaviour,as the shear resistance depends on the direction of shear. These materials consist of perpendicular layers of tows held together by a stitching thread.This thread restricts the movement of the tows,so that materials exhibit higher resistance to shear when they are sheared parallel to the stitch. Based on the data given in Figs 1.2 and 1.3,at the simplest level the curve may be approximated using a bi-linear model: Fs/Lp时=Eo tan8 (0<o) 1.4 Fs/Lpr Eo tan 0o Eoc tan (0-00) (8≥o） Clearly the material response(and hence the constants 00,Eo and E in equation 1.4)depends on material type and(for prepreg)experimental conditions such as
`locking angle' of a material. For successful tests, shear force can be calculated from the cross-head force using: Fs Fpf 2 cos 1:1 where is the frame angle and Fpf is the measured axial picture frame force. Test data can be normalised by dividing the shear force by the length of the picture frame, Lpf . Graphs of shear force against shear angle can be produced, where the shear angle is defined as: =2 ÿ 2 1:2 The shear angle can be calculated from the cross-head displacement, dpf , by 2 ÿ 2 cosÿ1 1 2 p dpf 2Lpf 1:3 The picture frame test procedure is relatively simple to perform and results should be reasonably repeatable if sufficient care is taken in cutting test samples and the correct clamping technique is used. The major benefit of the test is that shear angle and angular shear rate can be easily calculated from the cross-head displacement and displacement rate. Typical results from picture frame experiments conducted on a range of textile reinforcements and prepregs are shown in Figs 1.2 and 1.3. For clarity these graphs represent single experiments, although it should be noted that even well controlled experiments exhibit scatter of up to 20% in shear force for a particular angle. Most materials exhibit some similarities in terms of their shear force curve. Initially the shear resistance is low ± for dry fabrics this represents dry friction at tow crossovers, whilst for prepreg this corresponds to lubricated friction or viscous shear of polymer between fibres or yarns. Towards the end of the test the resistance increases significantly ± this happens once adjacent yarns come into contact, representing yarn compaction (fabrics) or squeeze flow (prepreg). If the test were continued, the curve would tend towards an asymptote corresponding to maximum possible deformation. Non-crimp fabrics (Fig. 1.2b) exhibit more complex behaviour, as the shear resistance depends on the direction of shear. These materials consist of perpendicular layers of tows held together by a stitching thread. This thread restricts the movement of the tows, so that materials exhibit higher resistance to shear when they are sheared parallel to the stitch. Based on the data given in Figs 1.2 and 1.3, at the simplest level the curve may be approximated using a bi-linear model: Fs=Lpf E0 tan < 0 1:4 Fs=Lpf E0 tan 0 E1 tan ÿ 0 0 Clearly the material response (and hence the constants 0, E0 and E1 in equation 1.4) depends on material type and (for prepreg) experimental conditions such as Composite forming mechanisms and materials characterisation 5 Copyrighted Material downloaded from Woodhead Publishing Online Delivered by http://woodhead.metapress.com Hong Kong Polytechnic University (714-57-975) Hong Kong Polytechnic University (714-57-975) Saturday, January 22, 2011 6:57:44 PM IP Address: 158.132.122.4
6 Composites forming technologies 1200 Plain weave 1000 (w/N) 4 harness satin 2:2 twill 800 600 400 200 0 0 10 20 30 40 50 60 70 Shear angle(deg) (a) 3500 (E之 3000 -Ebx-936 -Ebx-318 2500 2000 1500 1000 500 0 -50 -30 -10 10 30 50 70 -500 Shear angle(deg) -1000 1.2 Picture frame shear data for dry glass fabric reinforcements.(a)Three woven fabrics with superficial density 800 g/m2.(b)Non-crimp fabrics retained with tricot(Ebx-936)and chain (Ebx-318)stitch oriented at 45 to the tows in each case.Negative shear angle represents deformation perpendicular to the stitch. rate and temperature.However,as a rough guide,0o is likely to be around 40 for dry fabrics and 30 for prepreg,and is indicative of the point at which the material starts to lock as adjacent tows come into contact.The ratio Eo/E is of the order 3-4 for woven fabrics and closer to unity for prepreg.The most
rate and temperature. However, as a rough guide, 0 is likely to be around 40ë for dry fabrics and <30ë for prepreg, and is indicative of the point at which the material starts to lock as adjacent tows come into contact. The ratio E0=E1 is of the order 3±4 for woven fabrics and closer to unity for prepreg. The most 1.2 Picture frame shear data for dry glass fabric reinforcements. (a) Three woven fabrics with superficial density 800 g/m2 . (b) Non-crimp fabrics retained with tricot (Ebx-936) and chain (Ebx-318) stitch oriented at 45ë to the tows in each case. Negative shear angle represents deformation perpendicular to the stitch. 6 Composites forming technologies Copyrighted Material downloaded from Woodhead Publishing Online Delivered by http://woodhead.metapress.com Hong Kong Polytechnic University (714-57-975) Hong Kong Polytechnic University (714-57-975) Saturday, January 22, 2011 6:57:44 PM IP Address: 158.132.122.4
Composite forming mechanisms and materials characterisation 7 500 160°c 400 E之 如 Jeeys 200 100 180°℃ Sunysijqnd 0 0 10 20 30 40 50 60 70 (a) Shear angle(deg) 3000 93g1 2500 2000 4.65°s-1 1500 0.93°s-1 1000 500 0 10 20 30 40 50 60 70 (b) Shear angle(deg) 7.3 Picture frame shear data for pre-impregnated composites.(a)2:2 twill weave glass/polypropylene thermoplastic composite at two temperatures.(b) 5 harness satin weave carbon/epoxy prepreg at various angular shear rates (room temperature). rigorous approach to intra-ply shear characterisation would require every material to be characterised under all possible forming conditions.As this is clearly not a practical proposition,researchers have attempted to develop models to predict materials formability from textile structure3 and matrix rheology.4 This approach is discussed further in Chapter 4
rigorous approach to intra-ply shear characterisation would require every material to be characterised under all possible forming conditions. As this is clearly not a practical proposition, researchers have attempted to develop models to predict materials formability from textile structure13 and matrix rheology.14 This approach is discussed further in Chapter 4. 1.3 Picture frame shear data for pre-impregnated composites. (a) 2:2 twill weave glass/polypropylene thermoplastic composite at two temperatures. (b) 5 harness satin weave carbon/epoxy prepreg at various angular shear rates (room temperature). Composite forming mechanisms and materials characterisation 7 Copyrighted Material downloaded from Woodhead Publishing Online Delivered by http://woodhead.metapress.com Hong Kong Polytechnic University (714-57-975) Hong Kong Polytechnic University (714-57-975) Saturday, January 22, 2011 6:57:44 PM IP Address: 158.132.122.4
8 Composites forming technologies 1.2.2 Bias extension test The bias extension test involves clamping a rectangular piece of bidirectional material such that the tows are orientated initially at +45 to the direction of the applied tensile force.The material sample can be characterised by the aspect ratio,A=Lo/wo,where the sample width wo is usually greater than 100mm. Figure 1.4 shows an idealised bias extension test sample with A=2.The sample is divided into a number of regions which deform at different rates as the test proceeds.Generally it can be shown that the shear angle in region 4 is always twice that in regions denoted B,while region C remains un-deformed.The deformation in region 4 is the same as the deformation produced by the picture frame test,as long as intra-ply shear is the only mechanism;in practice the angle will be somewhat lower than in an equivalent picture frame sample as intra-ply slip will occur (i.e.tow spacing will increase)particularly as the material approaches locking.The sample aspect ratio must be at least two for the three different deformation regions to exist.Increasing the length/width ratio,A,to higher values serves to increase the area of region A. Bias extension tests are simple to perform and can provide reasonably repeatable results.Axial force and cross-head displacement are recorded during a test.The test provides a useful method to estimate the locking angle of a material; once the material in region 4 reaches the locking angle,it usually ceases to shear. As with the picture frame test,different clamping conditions have been suggested by various researchers,but the boundary conditions tend to affect the data much less than for picture frame tests.The method may be preferred for gaining shear data at elevated temperatures for thermoplastic composites,since the influence of relatively cool material adjacent to the metal clamps during high temperature testing is of less importance than in picture frame tests. 名 Clamp area Clamp area w C C B B B B A Lo A B B B Clamp area Clamp area (a) (b) 1.4 Idealised bias extension test sample with A=Lo/wo 2,where Lo and wo are respectively the initial length and width of the specimen
1.2.2 Bias extension test The bias extension test involves clamping a rectangular piece of bidirectional material such that the tows are orientated initially at 45ë to the direction of the applied tensile force. The material sample can be characterised by the aspect ratio, Lo=wo, where the sample width w0 is usually greater than 100mm. Figure 1.4 shows an idealised bias extension test sample with 2. The sample is divided into a number of regions which deform at different rates as the test proceeds. Generally it can be shown that the shear angle in region A is always twice that in regions denoted B, while region C remains un-deformed. The deformation in region A is the same as the deformation produced by the picture frame test, as long as intra-ply shear is the only mechanism; in practice the angle will be somewhat lower than in an equivalent picture frame sample as intra-ply slip will occur (i.e. tow spacing will increase) particularly as the material approaches locking. The sample aspect ratio must be at least two for the three different deformation regions to exist. Increasing the length/width ratio, , to higher values serves to increase the area of region A. Bias extension tests are simple to perform and can provide reasonably repeatable results. Axial force and cross-head displacement are recorded during a test. The test provides a useful method to estimate the locking angle of a material; once the material in region A reaches the locking angle, it usually ceases to shear. As with the picture frame test, different clamping conditions have been suggested by various researchers, but the boundary conditions tend to affect the data much less than for picture frame tests. The method may be preferred for gaining shear data at elevated temperatures for thermoplastic composites, since the influence of relatively cool material adjacent to the metal clamps during high temperature testing is of less importance than in picture frame tests. 1.4 Idealised bias extension test sample with L0=w0 2, where L0 and w0 are respectively the initial length and width of the specimen. 8 Composites forming technologies Copyrighted Material downloaded from Woodhead Publishing Online Delivered by http://woodhead.metapress.com Hong Kong Polytechnic University (714-57-975) Hong Kong Polytechnic University (714-57-975) Saturday, January 22, 2011 6:57:44 PM IP Address: 158.132.122.4
Composite forming mechanisms and materials characterisation 9 Producing graphs of shear force against shear angle can be achieved by following the same data analysis procedure as for picture frame test data, considering region 4 to be equivalent to a picture frame specimen with side length Lbe(see Fig.1.4).However,for increased accuracy,it may be necessary to measure the shear deformation rather than rely on the material following idealised deformation kinematics.This can be achieved by visual analysis, which can prove time consuming without an automated image acquisition and analysis approach. Forces from bias extension tests can be normalised by dividing by a charac- teristic dimension,such as sample width.However,this does not allow data from samples with different aspect ratios to be compared directly.Recent work by Harrison'considered the energy dissipated within regions 4,B and C to develop a more sophisticated normalisation technique for bias extension test data.In addition to allowing results from different aspect ratio samples to be compared, this also allows picture frame and bias extension test data to be correlated directly.Whilst this might allow mechanical data to be extracted in terms of shear force versus shear strain in a form suitable for simulation software,the data analysis procedure is extremely complex and hence at present the picture frame test is preferred for this purpose. 1.3 Axial loading 159 Loading of aligned fibre based materials along the fibre axis or axes typically results in very large forces and very low maximum strains in comparison to intra-ply shear.This might suggest that deformation under axial loading is of secondary importance,and indeed this is reflected in the relatively limited attention received by this topic.Boissels has long argued that this behaviour cannot be neglected,since the high magnitude of the axial stiffness indicates that tensile loading of the fibres accounts for the majority of energy dissipated during forming. Axial loading of textiles and composites can be conducted using standard tensile testing equipment,although as for the bias extension test(Section 1.2.2) wide samples are usually used.Unidirectional fibre materials will typically exhibit a linear force-displacement response when loaded parallel to the fibre axis.This is not the case for textile based materials,which exhibit an initial, non-linear stiffening due to crimp in the tows.As the fibres become aligned with the direction of loading,the response becomes linear and is determined by the fibre modulus and volume fraction.The importance of this 'de-crimping" depends on the properties of the transverse tows,and in particular their resistance to bending and compaction.If the transverse tows are also loaded, then the de-crimping zone will decrease in magnitude.Boissel has analysed a wide range of fabrics using a specially designed biaxial loading frame.Some typical results are given in Fig.1.5 for a plain weave fabric.When loaded
Producing graphs of shear force against shear angle can be achieved by following the same data analysis procedure as for picture frame test data, considering region A to be equivalent to a picture frame specimen with side length Lbe (see Fig. 1.4). However, for increased accuracy, it may be necessary to measure the shear deformation rather than rely on the material following idealised deformation kinematics. This can be achieved by visual analysis, which can prove time consuming without an automated image acquisition and analysis approach. Forces from bias extension tests can be normalised by dividing by a characteristic dimension, such as sample width. However, this does not allow data from samples with different aspect ratios to be compared directly. Recent work by Harrison9 considered the energy dissipated within regions A, B and C to develop a more sophisticated normalisation technique for bias extension test data. In addition to allowing results from different aspect ratio samples to be compared, this also allows picture frame and bias extension test data to be correlated directly. Whilst this might allow mechanical data to be extracted in terms of shear force versus shear strain in a form suitable for simulation software, the data analysis procedure is extremely complex and hence at present the picture frame test is preferred for this purpose. 1.3 Axial loading Loading of aligned fibre based materials along the fibre axis or axes typically results in very large forces and very low maximum strains in comparison to intra-ply shear. This might suggest that deformation under axial loading is of secondary importance, and indeed this is reflected in the relatively limited attention received by this topic. Boisse15 has long argued that this behaviour cannot be neglected, since the high magnitude of the axial stiffness indicates that tensile loading of the fibres accounts for the majority of energy dissipated during forming. Axial loading of textiles and composites can be conducted using standard tensile testing equipment, although as for the bias extension test (Section 1.2.2) wide samples are usually used. Unidirectional fibre materials will typically exhibit a linear force-displacement response when loaded parallel to the fibre axis. This is not the case for textile based materials, which exhibit an initial, non-linear stiffening due to crimp in the tows. As the fibres become aligned with the direction of loading, the response becomes linear and is determined by the fibre modulus and volume fraction. The importance of this `de-crimping' depends on the properties of the transverse tows, and in particular their resistance to bending and compaction. If the transverse tows are also loaded, then the de-crimping zone will decrease in magnitude. Boisse16 has analysed a wide range of fabrics using a specially designed biaxial loading frame. Some typical results are given in Fig. 1.5 for a plain weave fabric. When loaded Composite forming mechanisms and materials characterisation 9 Copyrighted Material downloaded from Woodhead Publishing Online Delivered by http://woodhead.metapress.com Hong Kong Polytechnic University (714-57-975) Hong Kong Polytechnic University (714-57-975) Saturday, January 22, 2011 6:57:44 PM IP Address: 158.132.122.4