The importance of measuring residual stresses 9 waviness arises through the manufacturing of a composite structure when the fibers undergo compressive axial loads resulting in thermal residual stresses. Since the matrix cannot provide any crosswise support,the fibers become deformed(micro-buckling),leading to waviness.The strength of the structure and the overall quality of the composite material will be reduced by fiber waviness (Parlevliet et al.,2007). Thermal residual stresses exceeding the yield strength of the resin can lead to the formation of cracks in composite materials.Cracks multiply along the interface when the fiber-matrix interface bond is weak.When the interface is strong,cracks can grow into the matrix.The thermal residual stresses can reach values near to those of the transverse ply strength,which can lead to the premature cracking of the ply during processing.Such cracks,which are commonly referred to as microcracks,are sometimes visible in transparent composites and create failure initiation sites in ensuing service life.Microcracks might propagate into transverse ply cracks,forming initiation points for delamination and the ultimate failure of the laminate,in the same manner as longitudinal splitting.More significantly,they can entail premature failure in cyclic loading conditions(Parlevliet et al.,2007). Besides the formation of transverse cracks,the difference in the magnitude of residual stresses between 0-degree and 90-degree plies in cross-ply laminates can also cause delamination.Inter-laminar failure of composite materials is indicated by gradual propagation of the delamination,eventually leading to loss of stiffness and structure strength.The free edge effect is another mechanism causing delamination,which also entails matrix cracking.Delamination at free edges is related to high inter-laminar stresses,developing because of the non-uniform features at the free edges.Delamination may also take place around any geometric concentration of stress,for instance around holes,cut-outs and changes in section. This significantly reduces the composite structure's capability to bear loads (Parlevliet et al.,2007). An important factor for affordable composite manufacturing is the ability to produce composite structures within tight dimensional tolerances.Residual stresses develop through the processing of composite structures and often entail dimensional changes,warpage of structures framed on flat tooling,and spring-in of flanges on angled sections.These sorts of defects can be observed more clearly in thin laminates,in which the magnitude of the deformation is often greater, particularly in the case of warpage.Warpage,or dimensional instability,may be the outcome of unbalanced cooling and tool-part interaction.Tool-part interaction has been proven to significantly contribute to warpage,for thin pieces especially. Furthermore,a differential temperature distribution in the mold can result in warpage (Fig.1.2)(Parlevliet et al.,2007). The main outcomes of residual stresses are strength reduction and shape distortion.Stresses at the fiber-matrix,lamina-laminate and structural levels all influence the strength of the component,while only lamina-laminate and structural level stresses influence dimensional stability to any great degree. Woodhead Publishing Limited,2014
The importance of measuring residual stresses 9 © Woodhead Publishing Limited, 2014 waviness arises through the manufacturing of a composite structure when the fi bers undergo compressive axial loads resulting in thermal residual stresses. Since the matrix cannot provide any crosswise support, the fi bers become deformed (micro- buckling), leading to waviness. The strength of the structure and the overall quality of the composite material will be reduced by fi ber waviness (Parlevliet et al. , 2007). Thermal residual stresses exceeding the yield strength of the resin can lead to the formation of cracks in composite materials. Cracks multiply along the interface when the fi ber- matrix interface bond is weak. When the interface is strong, cracks can grow into the matrix. The thermal residual stresses can reach values near to those of the transverse ply strength, which can lead to the premature cracking of the ply during processing. Such cracks, which are commonly referred to as microcracks, are sometimes visible in transparent composites and create failure initiation sites in ensuing service life. Microcracks might propagate into transverse ply cracks, forming initiation points for delamination and the ultimate failure of the laminate, in the same manner as longitudinal splitting. More signifi cantly, they can entail premature failure in cyclic loading conditions (Parlevliet et al. , 2007). Besides the formation of transverse cracks, the difference in the magnitude of residual stresses between 0-degree and 90-degree plies in cross- ply laminates can also cause delamination. Inter- laminar failure of composite materials is indicated by gradual propagation of the delamination, eventually leading to loss of stiffness and structure strength. The free edge effect is another mechanism causing delamination, which also entails matrix cracking. Delamination at free edges is related to high inter- laminar stresses, developing because of the non- uniform features at the free edges. Delamination may also take place around any geometric concentration of stress, for instance around holes, cut- outs and changes in section. This signifi cantly reduces the composite structure’s capability to bear loads (Parlevliet et al. , 2007). An important factor for affordable composite manufacturing is the ability to produce composite structures within tight dimensional tolerances. Residual stresses develop through the processing of composite structures and often entail dimensional changes, warpage of structures framed on fl at tooling, and spring- in of fl anges on angled sections. These sorts of defects can be observed more clearly in thin laminates, in which the magnitude of the deformation is often greater, particularly in the case of warpage. Warpage, or dimensional instability, may be the outcome of unbalanced cooling and tool- part interaction. Tool- part interaction has been proven to signifi cantly contribute to warpage, for thin pieces especially. Furthermore, a differential temperature distribution in the mold can result in warpage ( Fig. 1.2 ) (Parlevliet et al. , 2007). The main outcomes of residual stresses are strength reduction and shape distortion. Stresses at the fi ber- matrix, lamina- laminate and structural levels all infl uence the strength of the component, while only lamina- laminate and structural level stresses infl uence dimensional stability to any great degree
10 Residual stresses in composite materials 1.2 Distorted glass fiber reinforced polyetherimide(Cetex)due to non-uniform cooling of the hot plate press (Parlevliet et al.,2007). 1.4 The importance of residual stress measurement In recent years,the growing employment of advanced laminated composites has drawn a lot of attention to process-induced residual stresses.Explaining such residual stresses has therefore been a subject of great interest(Liu,1999).While the reliable measurement and prediction of residual stresses is a challenge,their distribution and size are the critical factors in determining how a composite will behave (Colpo,2006). The residual stresses developing through processing and operating conditions cannot be neglected.They can significantly compromise a laminate's strength.If these stresses are not precisely understood,they may result in material failures where the tensile stresses developed in the matrix surpass its critical tensile strength.Once this has occurred,microcracks could develop,permitting hydrogen to seep into the core.The formation of microcracks in other structures exposes the fibers to possibly degrading environmental conditions and possible chemical attack in storage facilities (Myers,2004).These stresses are usually small; however,they may be comparable with those generated by mechanical loads due to the low stresses imposed by the design codes.Thus if the residual stresses are not taken into account,the overall stress state is misrepresented,potentially increasing the risk of failures related to environmentally assisted cracking(Reid and Paskaramoorthy,2009). The residual stress state must be superimposed on any stress state resulting from external loading,in order to estimate the actual stress state existing when a structure undergoes external loading.When the overall stress surpasses the design stress limit of the material,this combined stress can entail premature structural failure.Thus it is important to evaluate residual stresses in order to predict the failure mode of a composite (Seif and Short,2002;Seif et al.,2006).If residual stresses are not taken into account throughout the structural design phase,a higher safety factor should be considered for the structure,usually resulting in overweight and over-designed structures(Stamatopoulos,2011). Woodhead Publishing Limited,2014
10 Residual stresses in composite materials © Woodhead Publishing Limited, 2014 1.4 The importance of residual stress measurement In recent years, the growing employment of advanced laminated composites has drawn a lot of attention to process- induced residual stresses. Explaining such residual stresses has therefore been a subject of great interest (Liu, 1999). While the reliable measurement and prediction of residual stresses is a challenge, their distribution and size are the critical factors in determining how a composite will behave (Colpo, 2006). The residual stresses developing through processing and operating conditions cannot be neglected. They can signifi cantly compromise a laminate’s strength. If these stresses are not precisely understood, they may result in material failures where the tensile stresses developed in the matrix surpass its critical tensile strength. Once this has occurred, microcracks could develop, permitting hydrogen to seep into the core. The formation of microcracks in other structures exposes the fi bers to possibly degrading environmental conditions and possible chemical attack in storage facilities (Myers, 2004). These stresses are usually small; however, they may be comparable with those generated by mechanical loads due to the low stresses imposed by the design codes. Thus if the residual stresses are not taken into account, the overall stress state is misrepresented, potentially increasing the risk of failures related to environmentally assisted cracking (Reid and Paskaramoorthy, 2009). The residual stress state must be superimposed on any stress state resulting from external loading, in order to estimate the actual stress state existing when a structure undergoes external loading. When the overall stress surpasses the design stress limit of the material, this combined stress can entail premature structural failure. Thus it is important to evaluate residual stresses in order to predict the failure mode of a composite (Seif and Short, 2002; Seif et al. , 2006). If residual stresses are not taken into account throughout the structural design phase, a higher safety factor should be considered for the structure, usually resulting in overweight and over- designed structures (Stamatopoulos, 2011). 1.2 Distorted glass fi ber reinforced polyetherimide (Cetex ® ) due to non-uniform cooling of the hot plate press (Parlevliet et al. , 2007)
The importance of measuring residual stresses 11 1.5 Issues in the measurement of residual stresses It is difficult to measure the specific residual stress contributing to matrix-fiber failures with sufficient spatial resolution to predict their effects.The development of residual stress implies nonlinear material behavior and often entails material discharge,phase transformations,and mechanical and thermal problems.For most problems,predictive capabilities are inadequate.Therefore,being able to measure residual stress is essential to satisfy two goals: 1.minimizing failures pertaining to residual stresses;and 2.to promote the development of predictive capabilities through the verification of models. Residual stresses can be investigated through empirical methods,or by modeling of production mechanisms(Prime,1999a,b).Empirical methods may or may not be destructive.The methods which are considered destructive normally entail cutting or drilling processes,in order to relax the residual stresses.Then residual stresses are calculated considering the changes of dimension that have taken place.For the non-destructive methods,diffraction techniques are employed (Prime,1999a,b). Since advanced structural composites were developed,the need to understand their behavior has led to a large body of research in residual stress determination. Many existing methods have been developed for the characterization of residual stresses and their effects.Some of these methods have been developed from existing tests performed on other materials,and some are completely new.Existing methods are often categorized into two broad groups:destructive and non- destructive.Destructive testing implies damaging or removing a section of material so that the specimen may no longer be usefully employed.Non-destructive tests are often preferred to destructive tests for these reasons.Furthermore,with non-destructive testing,tests can be repeated on the same specimen in order to improve accuracy.Generally,the latter methods also offer the means to test the specimen over time (Myers,2004). 1.6 Techniques for measuring residual stress in composites There are a wide range of experimental methods to measure the residual stresses. These methods can be categorized into four main groups(Reid,2009): 1.Methods considering the response to released residual stresses:Methods that monitor the elastic response of a laminate to the release ofresidual stresses are probably the most widely used measurement techniques.A variety of methods for releasing residual stresses within a laminate are available.These include layer removal,Sachs method,hole drilling,ring-core method,deep-hole drilling,incremental slitting method,contour method,sectioning method, Woodhead Publishing Limited,2014
The importance of measuring residual stresses 11 © Woodhead Publishing Limited, 2014 1.5 Issues in the measurement of residual stresses It is diffi cult to measure the specifi c residual stress contributing to matrix- fi ber failures with suffi cient spatial resolution to predict their effects. The development of residual stress implies nonlinear material behavior and often entails material discharge, phase transformations, and mechanical and thermal problems. For most problems, predictive capabilities are inadequate. Therefore, being able to measure residual stress is essential to satisfy two goals: 1. minimizing failures pertaining to residual stresses; and 2. to promote the development of predictive capabilities through the verifi cation of models. Residual stresses can be investigated through empirical methods, or by modeling of production mechanisms (Prime, 1999a,b). Empirical methods may or may not be destructive. The methods which are considered destructive normally entail cutting or drilling processes, in order to relax the residual stresses. Then residual stresses are calculated considering the changes of dimension that have taken place. For the non- destructive methods, diffraction techniques are employed (Prime, 1999a,b). Since advanced structural composites were developed, the need to understand their behavior has led to a large body of research in residual stress determination. Many existing methods have been developed for the characterization of residual stresses and their effects. Some of these methods have been developed from existing tests performed on other materials, and some are completely new. Existing methods are often categorized into two broad groups: destructive and nondestructive. Destructive testing implies damaging or removing a section of material so that the specimen may no longer be usefully employed. Non- destructive tests are often preferred to destructive tests for these reasons. Furthermore, with non- destructive testing, tests can be repeated on the same specimen in order to improve accuracy. Generally, the latter methods also offer the means to test the specimen over time (Myers, 2004). 1.6 Techniques for measuring residual stress in composites There are a wide range of experimental methods to measure the residual stresses. These methods can be categorized into four main groups (Reid, 2009): 1. Methods considering the response to released residual stresses : Methods that monitor the elastic response of a laminate to the release of residual stresses are probably the most widely used measurement techniques. A variety of methods for releasing residual stresses within a laminate are available. These include layer removal, Sachs method, hole drilling, ring- core method, deep- hole drilling, incremental slitting method, contour method, sectioning method
12 Residual stresses in composite materials radial cutting method,matrix removal methods and micro-indentation techniques. 2.Methods considering changes due to failure:The estimation of residual stresses through the measurement of the change in apparent failure strength can,in principle,be applied to any material with a well-defined failure (or yield)strength.However,the method appears only to have been applied to the measurement of transverse residual stresses in composite materials and in the form of the 'first ply failure method'. 3.Methods considering changes in the material structure:Methods that rely on changes in the material structure include X-ray diffraction,neutron diffraction, Raman spectroscopy,photoelasticity,and the use of acoustic waves.Only the use of acoustic waves requires contact with the specimen and all methods are potentially non-destructive. 4.Methods considering the response to changes in temperature:These methods include measurement of curvature,Cure referencing method,and Local heating methods. Each of these methods has its own advantages and shortcomings (Reid,2009). From another perspective,experimental methods for the estimation of residual stresses can be divided into two categories,destructive and non-destructive. Furthermore,the non-destructive methods are divided into those that use the inherent material properties,those that use sensors,and finally those that use in- plane and out-of-plane deformation.The non-destructive methods in general can provide results for large areas a laminate,whereas the measurements acquired using destructive methods concern only a small area of the structure.Some of the existing non-destructive methods make use of the inherent material properties of the composites,since some material properties change when the material is exposed to stresses or stains.Photoelasticity,Micro-Raman spectroscopy and measurement of electrical conductivity of fibers are three methods of this kind (Stamatopoulos,2011).Several methods are based on in-plane and out-of-plane deformations,such as methods based on interferometry,warpage of asymmetrical composite materials,Neutron diffraction and X-ray diffraction. There are also destructive methods that attempt to estimate the residual stresses. The main disadvantage of the non-destructive methods described above is that they do not provide information on the distribution of global residual stresses in the composite or along the plies.The methods which can measure the distribution of residual stresses are based mainly on destructive techniques.The general principle shared by all destructive methods is that some stressed material is removed and the resulting deformations (usually displacements or strains)are measured(Schajer and Prime,2006).The destructive methods include first-ply failure,layer removal method,the incremental hole-drilling method,the deep- hole method and the crack compliance method (Stamatopoulos,2011).The following chapters review the range of destructive and non-destructive techniques. Woodhead Publishing Limited,2014
12 Residual stresses in composite materials © Woodhead Publishing Limited, 2014 radial cutting method, matrix removal methods and micro- indentation techniques. 2. Methods considering changes due to failure : The estimation of residual stresses through the measurement of the change in apparent failure strength can, in principle, be applied to any material with a well- defi ned failure (or yield) strength. However, the method appears only to have been applied to the measurement of transverse residual stresses in composite materials and in the form of the ‘fi rst ply failure method’. 3. Methods considering changes in the material structure : Methods that rely on changes in the material structure include X-ray diffraction, neutron diffraction, Raman spectroscopy, photoelasticity, and the use of acoustic waves. Only the use of acoustic waves requires contact with the specimen and all methods are potentially non- destructive. 4. Methods considering the response to changes in temperature : These methods include measurement of curvature, Cure referencing method, and Local heating methods. Each of these methods has its own advantages and shortcomings (Reid, 2009). From another perspective, experimental methods for the estimation of residual stresses can be divided into two categories, destructive and non- destructive. Furthermore, the non- destructive methods are divided into those that use the inherent material properties, those that use sensors, and fi nally those that use inplane and out-of- plane deformation. The non- destructive methods in general can provide results for large areas a laminate, whereas the measurements acquired using destructive methods concern only a small area of the structure. Some of the existing non- destructive methods make use of the inherent material properties of the composites, since some material properties change when the material is exposed to stresses or stains. Photoelasticity, Micro-Raman spectroscopy and measurement of electrical conductivity of fi bers are three methods of this kind (Stamatopoulos, 2011). Several methods are based on in- plane and out- of-plane deformations, such as methods based on interferometry, warpage of asymmetrical composite materials, Neutron diffraction and X-ray diffraction. There are also destructive methods that attempt to estimate the residual stresses. The main disadvantage of the non- destructive methods described above is that they do not provide information on the distribution of global residual stresses in the composite or along the plies. The methods which can measure the distribution of residual stresses are based mainly on destructive techniques. The general principle shared by all destructive methods is that some stressed material is removed and the resulting deformations (usually displacements or strains) are measured (Schajer and Prime, 2006). The destructive methods include fi rst- ply failure, layer removal method, the incremental hole- drilling method, the deephole method and the crack compliance method (Stamatopoulos, 2011). The following chapters review the range of destructive and non- destructive techniques
The importance of measuring residual stresses 13 1.7 References Barnes,J.A.and Byerly,G.E.(1994)Formation of residual stresses in laminated thermoplastic composites,Composites Science and Technologies,51(4),479-94. Brinksmeier,E.,Cummett,J.T.,Leskovar,P,Knig,W.,Peters,J.and Tnshoff,H.K. (1982)Residual stresses-measurement and causes in machining processes,Annals of CRP,31,491-510. Cheng,W.and Finnie,I.(2007)Residual Stress Measurement and the Slitting Method, New York,Springer. Colpo,F.(2006)Residual Stress Characterization in a Single Fibre Composite Specimen by Using FBG Sensor and the OLCR Technique,PhD Thesis,Lausanne,EPFL. Deve,H.E.and Maloney,M.J.(1991)On the toughening of intermetallics with ductile fibers-role of interfaces,Acta Mettallurgica et Materialia,39(10), 2275-84. Finnie,I.,Cheng,W.and McCorkindale,K.J.(1990)Delayed crack propagation in a steel pressure vessel due to thermal stresses,International Journal of Pressure Vessels and Piping,42,15-31. Fitzpatrick,M.E.and Lodini,A.(eds)(2003)Analysis of Residual Stress by Diffraction using Neutron and Synchrotron Radiation,London,Taylor Francis,Inc. Gibson,R.F.(1994)Principles of Composite Material Mechanics,New York, McGraw-Hill. Jain,L.K.and Mai,Y.W.(1996)On residual stress induced distortions during fabrication of composite shells,Journal of Reinforced Plastics and Composites,15(8), 793-805 Liu,H.Y.,Zhang,X.,Mai,Y.W.and Diao,X.X.(1999)On steady-state fibre pull-out.Part II:Computer simulation,Composite Science and Technology,59(15),2191-9. Liu,S.C.(1999)Residual Stress Characterization for Laminated Composites,Ph.D. Thesis,University of Florida.Available at:etd.fclaedu/UF/amp7373/liu.pdf Masubushi,K.(1980)Analysis of Welded Structures:Residual Stresses,Distortion and their Conseguences,Oxford,Pergamon Press. Myers,D.G.(2004)Method for Measurement of Residual Stress and Coefficient of Thermal Expansion of Laminated Composites,MSc Thesis,University of Florida. Nakamura,T.and Suresh,S.(1993)Effects of thermal residual stresses and fiber packing on deformation of metal-matrix composites,Acta Metallurgica et Materialia,41(6), 1665-81 Nath,R.B.,Fenner,D.N.and Galiotis,C.(2000)The progressional approach to interfacial failure in carbon reinforced composites:Elasto-plastic finite element modeling of interface cracks,Composites:Part A,31(9),929-43. Parlevliet,P.P.,Bersee,E.N.and Beukers,A.(2006)Residual stresses in thermoplastic composites-a study ofthe literature.Part I:Formation ofresidual stresses,Composites: Par1A,37(11),1847-57. Parlevliet,P.P.,Bersee,E.N.and Beukers,A.(2007)Residual stresses in thermoplastic composites-a study of the literature.Part III:Effects of thermal residual stresses, Composites:Part A,38,1581-96. Prime,M.B.(1999a)Measuring residual stress and the resulting stress intensity factor in compact tension specimens,Fatigue and Fracture of Engineering Materials and Structures,22(3),195-204. Prime,M.B.(1999b)Residual stress measurement by successive extension of a slot:the crack compliance method,Applied Mechanics Reviews,52(2),75-96. Woodhead Publishing Limited,2014
The importance of measuring residual stresses 13 © Woodhead Publishing Limited, 2014 1.7References Barnes , J. A. and Byerly , G. E. ( 1994 ) Formation of residual stresses in laminated thermoplastic composites , Composites Science and Technologies , 51 (4) , 479 – 94 . Brinksmeier , E. , Cummett , J. T. , Leskovar , P. , Knig , W. , Peters , J. and Tnshoff , H. K. ( 1982 ) Residual stresses – measurement and causes in machining processes , Annals of CIRP , 31 , 491 – 510 . Cheng , W. and Finnie , I. ( 2007 ) Residual Stress Measurement and the Slitting Method , New York , Springer . Colpo , F. ( 2006 ) Residual Stress Characterization in a Single Fibre Composite Specimen by Using FBG Sensor and the OLCR Technique , PhD Thesis, Lausanne , EPFL . Deve , H. E. and Maloney , M. J. ( 1991 ) On the toughening of intermetallics with ductile fi bers – role of interfaces , Acta Mettallurgica et Materialia , 39 (10) , 2275 – 84 . Finnie , I. , Cheng , W. and McCorkindale , K. J. ( 1990 ) Delayed crack propagation in a steel pressure vessel due to thermal stresses , International Journal of Pressure Vessels and Piping , 42 , 15 – 31 . Fitzpatrick , M. E. and Lodini , A. (eds) ( 2003 ) Analysis of Residual Stress by Diffraction using Neutron and Synchrotron Radiation , London , Taylor & Francis , Inc. Gibson , R. F. ( 1994 ) Principles of Composite Material Mechanics , New York , McGraw-Hill . Jain , L. K. and Mai , Y. W. ( 1996 ) On residual stress induced distortions during fabrication of composite shells , Journal of Reinforced Plastics and Composites , 15 (8) , 793 – 805 . Liu , H. Y. , Zhang , X. , Mai , Y. W. and Diao , X. X. ( 1999 ) On steady- state fi bre pull- out. Part II: Computer simulation , Composite Science and Technology , 59 (15) , 2191 – 9 . Liu , S. C. ( 1999 ) Residual Stress Characterization for Laminated Composites , Ph.D. Thesis, University of Florida . Available at: etd.fcla.edu/UF/amp7373/liu.pdf Masubushi , K. ( 1980 ) Analysis of Welded Structures: Residual Stresses, Distortion and their Consequences , Oxford , Pergamon Press . Myers , D. G. ( 2004 ) Method for Measurement of Residual Stress and Coeffi cient of Thermal Expansion of Laminated Composites , MSc Thesis, University of Florida . Nakamura , T. and Suresh , S. ( 1993 ) Effects of thermal residual stresses and fi ber packing on deformation of metal- matrix composites , Acta Metallurgica et Materialia , 41 (6) , 1665 – 81 . Nath , R. B. , Fenner , D. N. and Galiotis , C. ( 2000 ) The progressional approach to interfacial failure in carbon reinforced composites: Elasto- plastic fi nite element modeling of interface cracks , Composites: Part A , 31 (9) , 929 – 43 . Parlevliet , P. P. , Bersee , E. N. and Beukers , A. ( 2006 ) Residual stresses in thermoplastic composites – a study of the literature. Part I: Formation of residual stresses , Composites: Part A , 37 (11) , 1847 – 57 . Parlevliet , P. P. , Bersee , E.N. and Beukers , A. ( 2007 ) Residual stresses in thermoplastic composites – a study of the literature. Part III: Effects of thermal residual stresses , Composites: Part A , 38 , 1581 – 96 . Prime , M. B. ( 1999 a) Measuring residual stress and the resulting stress intensity factor in compact tension specimens , Fatigue and Fracture of Engineering Materials and Structures , 22 (3) , 195 – 204 . Prime , M. B. ( 1999 b) Residual stress measurement by successive extension of a slot: the crack compliance method , Applied Mechanics Reviews , 52 (2) , 75 – 96