Digital Image Correlation measurements as a tool of composites deformation description
Tomasz Brynk ⇑,Rafal M.Molak,Miroslawa Janiszewska,Zbigniew Pakiela
Warsaw University of Technology,Faculty of Materials Science and Engineering,Warsaw,Poland
椿木a r t i c l e i n f o Article history:
Received 1November 2011
Received in revid form 29March 2012Accepted 4April 2012
Available online 12May 2012
Keywords:
Carbon–fiber-reinforced polymers (CFRP)Digital Image Correlation (DIC)Mechanical properties
a b s t r a c t
The aim of prented investigations was to perform uniaxial tensile tests of flat specimens made of car-bon fiber reinforced composites with strain measurements performed by means of Digital Image Corre-lation (DIC)method.Samples with three reinforcement types were examined.DIC method was ud in two point tracing mode (optical extensometer)for longitudinal and transversal strain measurements.The results of this measurement were ud for Young’s modulus and Poisson’s ratio calculations neces-sary in the designing process of the mobile robot manipulator.Additionally strain fields maps were eval-uated for the lected moments of tensile tests.Obtained strain maps showed inhomogeneity of strain distributions and its dependence on the reinforcement fibers types.Strain maps showed also the areas where delamination occurs.Prented methodology may be applied for validation of the results of com-posite materials deformation modeling.
Ó2012Elvier B.V.All rights rerved.怎么安装软件
1.Introduction
In mechanical testing of structural materials usually only two points bad strain measurement devices or strain gages are uti-lized.Such measurements do not deliver much uful information concerning strain concentration and decohesion of materials.This information may be delivered when using optical methods of strain measurements.One of the methods,which became widely ud in recent years due to development of PC computers and im-age acquisition methods and devices [1,2],is Digital Image Correla-tion.The method has been invented in 80s of XX century [3–5].DIC method has been successfully applied in wide range of applications related with mechanical testing of materials.One of many exam-ples are minisamples tensile tests [6],measurements of PLC bands strain development [7,8]and stress intensity factor determination by means of DIC and inver methods [9,10].There are many examples of composites samples testing there are many literature examples of DIC measurements in test with monotonic [11,12]as well as dynamic loading conditions [13].DIC measurement were also successfully utilized in crack resistance tests [14].There are reported investigations concerning strain measurement on the sur-faces of the whole components made of composites [15,16].
The aim of prented investigation was to perform DIC mea-surements during uniaxial tensile tests f
or determination of Young’s modulus and Poisson’s ratio (during elastic range of load-ing)in carbon reinforced composite polymers (CFRP)samples and to obtain full-field strain maps at the moments of tests just after the delamination commenced.Test were performed on the three types of samples made of carbon reinforced epoxy composites with different orientation of fibers and different type of reinforcement.Tested composites are candidates for the application in the mobile robot extension arm.Result of measurements will be utilized in the designing process of such a part.2.Materials
Variety alloys have successfully been replaced by composites in aerospace,automotive and marine applications [17]due to the best combination of specific strength and elastic modulus.In all of the areas there is a main trend to reduce the weight.
In composites materials the fibers are protected from abrasion and environmental effects by the surrounding resin.The resin gives the composite durability,shape and appearance.The fibers are responsible for the high stiffness and strength.The resin has influ-ence on interlaminar shear strength which is esntial for applica-tions underbending loads.
The matrix material ud in this study was EPIKOTE resin MGS LR 285produced by the Hexion Specialty Chemicals company.For the purpo of the study three types of the samples with various fi
bers orientations were produced.Tape reinforcement composites with unidirectional carbon fibers,as well as composites reinforced with 2D fabrics,plain and twill were manufactured.三大合成材料
Fig.1a shows plain weave,where each weft roving pass over one and under one warp roving,and each warp roving pass over one and under one weft roving.Fig.1b prents twill weave,where
0927-0256/$-e front matter Ó2012Elvier B.V.All rights rerved./10.atsci.2012.04.034
Corresponding author.
E-mail address:tbrynk@inmat.pw.edu.pl (T.Brynk).
applying speckle pattern each of rectangular subts of a few pixels size have unique distribution of grey level inside.Position of the subt on the image after deformation is determined by minimiz-ing correlation function in the form:
Cðx;y;u;vÞ¼X n2
i;j¼Àn
ðIðxþi;yþjÞÀIÃðxþuþi;yþvþjÞÞ2
where x,y–pixel coordinates;u,v–displacements;n–subt size; (I(x+i,y+j)–pixel intensity on reference image;I⁄(x+u+i,y+v+j)–pixel intensity on the image after deformation.
The most important parameters related to calculations using DIC software are subt size(n),calculation step(parameter which defines the spacing between pixels which are taken into account during calculations)and interpolation type.Prented results were obtained with the subt size of19pix,step equal to3and cubic b-spline interpolation.Strain maps were obtained using strain windows size parameter equal to15.Strain window size defines the size of the local neighbourhood in which the deriva-tives of displacementfield are calculated,and is responsible for smoothing of the maps.
Table1
Samples characterization.
狗狗图片大全图片Type of reinforcement Gauge ction dimensions(mm)Number of layers Direction offibers Fibers content Resin/
Hardener
Mass Volumetric
KDU1007,unidirectional tape149.9Â14.5Â1.24II(parallel to the direction of loading)0.510.41MGS285/286 Fabric98130,plain weave150Â24.6Â2.410X(45°to the direction of loading)0.500.40MGS285/286 Fabric98130,twill weave150Â24.6Â2.210X(45°to the direction of loading)0.500.40MGS285/286
When interpolation between pixel intensities is applied DIC dis-placement measurements may achieve,according to software producer,up to0.01pix accuracy.Typical systematic error of dis-placement measurements in real experimental conditions is esti-mated to be at the level of0.05pix(due to minor out of plain movement and lens distortions).The applied optical system allowed for displacement measurement with scaling factor equal to0.094mm/pix,therefore the expected,systematic error was esti-mated to be at0.047l m level.
做好自己的图片Results of longitudinal displacement obtained by means of DIC measurements,namely‘optical extensometer’,were corre-lated with the data registered by load cell and were ud to pre-
Fig.4.Strain maps of tape reinforced sample for images registered just after delamination occurred.
pare stress–strain plots for investigated samples.The slope linear part of such a plots were taken into account when calculating Young’s modulus(40subquent data points with the highest R2coefficient of linearfit were evaluated).When using‘optical extensometer’DIC software traced only two subts positions and distance between them therefore it allows for relatively fast DIC calculations for all registered images of samples surfaces.‘Optical extensometer’function was also applied when measur-ing the transversal strain.Measurement results for digital images registered in t陕西一本线
he same time period of each test like that ud for Young’s modulus calculations were utilized for Poisson’s ratio determination.
Digital images acquired just after the delamination of compos-ite materials has commenced were ud for calculations of the strain maps.The areas on the samples surfaces were the delamina-tion occurred were automatically removed from the strain maps
samples surfaces.Due to relatively high total deformation of com-posite samples reinforced with2D fabrics some part of measure-ment area for this samples were moved out of the area of obrvation of optical system.Therefore there was not possible to obtain strain maps in areas from the left sides for this samples (e Figs.5and6).
Composites with plain and twill weave have similar values of ultimate strength and stiffness however the sample with plain weave reinforcement showed higher elongation to break value.It may be explained by later strain localization and smoother strain distribution for this sample which is clearly visible when compar-ing e xx and e yy strain maps(e Figs.5and6).Delamination in ca of plain weave type composite commenced near the edge of the sample and is located in a few distant places at the same time (e uncolored regions in Fig.5).This is the evidence of more uni-formfiber straining occurred in weave type when comparing to
160T.Brynk et al./Computational Materials Science64(2012)157–161
show that application of DIC measurements widespread the exper-imental data amount from test comparing to the usage of typical electromechanical extensometer which allow only two point dis-placement tracing.浑浑噩噩的意思
统计学习方法
5Conclusions
Application of DIC method for strain measurements during ten-sile tests of composite materials allowed for measurement of more parameters then derived directly from stress–strain curves ob-tained with strain measurement made by means of electrome-chanical extensometer.Transver displacement measurement obtained by means of‘optical extensometer’were ud for deter-mination of the Poisson’s ratio values.
Strain maps obtained for the moments just after delamination commenced were obtained for all tested samples.The usage of the global threshold function allowed for removing the areas where speckle pattern was damaged from the strain maps.Thus gave the opportunity to clearly illustrate strain distribution depen-dence on the reinforcement type and indicate the areas where delamination occurred.Analyzes of such a strain maps for2D fabric reinforced samples made a contribution in understanding the high-er elongation(obrved in pair with almost the same ultimate strength)of plain weave reinforced composite in comparison with twill weave counterpart.Strain maps with excluded delamination areas were also obtained in ca of tape reinforced composite, which had significantly higher strength and stiffness but also much lower elongation to failure.
The obtained results show applicability of DIC in accurate strain measurements during tensile test of different CFRP samples.Be-sides determination of elastic constants they helped with the bet-ter understanding of the damage mechanisms in carbon epoxy composites and may be utilized for validation of the models of composite materials.Acknowledgements
This work was co-financed by the European Regional Development Fund within the framework of the1.priority axis of the Innovative Economy Operational Programme,2007–2013, submeasure1.1.2‘‘Strategic R&D Rearch’’under Contract No. POIG.01.02.01-00-014/08.
References
[1]M.A.Sutton,J.J.Orteu,H.Schreier,Image Correlation for Shape,Motion and
Deformation Measurements,Basic Concepts,Theory and Applications, Springer,New York,2009.
[2]B.Pan,K.Qian,H.Xie,A.Asundi,Meas.Sci.Technol.20(2009)1–17.
[3]M.A.Sutton,W.J.Wolters,W.H.Peters,W.F.Ranson,S.R.McNeill,Image Vision
Comput.3(1983)133–139.
[4]W.H.Peters,W.F.Ranson,Opt.Eng.21(1982)427–431.
[5]W.H.Peters,W.F.Ranson,M.A.Sutton,T.C.Chu,J.Anderson,Opt.Eng.22
(1983)738–742.
[6]R.M.Molak,K.Paradowski,T.Brynk,L.Ciupinski,Z.Pakiela,K.J.Kurzydlowski,
Int.J.Pres.Ves.Pip.86(2009)43–47.
[7]W.Tong,H.Tao,N.Zhang,L.G.Hector Jr.,Scripta Mater.53(2005)87–92.
[8]J.Zdunek,T.Brynk,J.Mizera,Z.Pakieła,K.J.Kurzydłowski,Mater.Charact.59
(2008)1429–1433.
[9]J.Abanto-Bueno,J.Lambros,Eng.Fract.Mech.69(2002)1695–1711.
[10]T.Brynk,M.Rasinski,Z.Pakiela,L.Olejnik,K.J.Kurzydlowski,Phys.Status
Solidi A207(2010)1132–1135.
[11]P.Lecomte-Grosbras,B.Paluch,M.Brieu,G.Saxcé,L.Sabatier,Compos:Part A
40(2009)1911–1920.
[12]Z.Y.Wang,H.Q.Li,J.W.Tong,M.Shen,F.Aymerich,P.Priolo,Compos Sci
Technol68(2008)1975–1980.
[13]S.Giancane,F.W.Panella,R.Nobile,V.Dattoma,Proced.Eng.2(2010)1307–
1315.
[14]G.Catalanotti,P.P.Camanho,J.Xavier,C.G.Dávila,A.T.Marques,Compos.Sci.
Technol.70(2010)1986–1993.
[15]L.B.Meng,G.C.Jin,X.F.Yao,H.Y.Yeh,Polym.Test.25(2006)42–48.
[16]X.F.Yao,L.B.Meng,J.C.Jin,H.Y.Yeh,Polym.Test.24(2005)245–251.
[17]P.K.Mallick,Fiber reinforced compos.,Mater,Manuf.,Des.,third ed.,CRC Press,
Boca Raton,2008.
[18]ISO527:2009:Plastics–Determination of tensile properties–Part5:Test
conditions for unidirectionalfibre-reinforced plastic composites.
T.Brynk et al./Computational Materials Science64(2012)157–161161
