The McGraw: ed 54 _ Elements of Fracture Mechanics _ 4s E with the maximum crack opening equal to 40.a/E at x, = 0. In fracture mechanics problems, if Z(z,) is expressed with the crack tip as origin, Jz, appears in the denominator of all the three modes. Another formal definition of K (3.7], which is an alternative yet equivalent to the two definitions already presented through Eqs (3.3) and (3.4), is, K= nz, Z@,) (3.36) 20 z x If this definition is applied to Eq, (3.29), we have the familiar result, K,= 0 ra We realize that the simple expression of Eq, (3.29) was obtained by neglecting some terms from Eq. (3.28). Some of us may, at this stage, like to split hairs and be interested in knowing how approximate the results are for the relation of Eq. (3.29). To have some idea, we would explore how 62, differs from the correct solution on x, = 0 plane. For the biaxial case under consideration, the rigorous solution of Eq. (3.24) is simplified for x, = 0 to Ze) = —* for Ixjl>a (3-#) Substituting in Eq. (3.19b) leads to on A (gay The origin is moved to the crack tip by the transformation x; = @ + x, and, with some mani- pulation, we obtain on (142, /a) ie yw 2a) o (2x,/a)? The approximate value of G2, is obtained from Eq, (3.32b) for 0= 0 as on 1 (2x. /a)" The percentage difference between the two values, rigorous and approximate (labeled as ‘no approximation’ and ‘one term’ respectively in Fig. 3.7), increases with the increase in distance from the origin on the crack plane. At the distance where x,/a = 0.15, the difference is 9.8%. The approximate relation of Eq. (3.29) can be improved by retaining higher terms in the expansion of (1 + z,/2a)""” of Eq. (3.28). In fact, the form of Z;(z,) changes to K Zp ae t Anes! + Agee? + Agee? + (3.37) (2n2,) where A,, Ay, Ay, ete. are real numbers. If we retain the first two terms, dis modified to K, oe 6. 30. 3r 28)) Gm= B— Naz cos g| 1 tsin5 sin + [1 -sin* >) | (en 2 29" ot aa 2))The McGraw-Hill companies Stress Intensity Factor 55 35 30 & S25 6 No approximation 20 Two terms ‘One term 15, 0.05 a 05 Fig. 3.7 oy, on the crack plane for biaxial loading In Fig. 3.7 the solution of ‘two terms’ is also plotted for 6, from the above results for = 0. It shows that the inclusion of one more term considerably reduces the difference between the rigorous and approximate values. For the biaxial loading problem, Fig. 3.8(a) shows the photo-elastic fringes (simulated on a computer) in the approximate case of using only the first term in Eq. (3.37). The lobes of the fringes are almost normal to the crack plane. Figure 3.8(b) shows the fringes when two terms are used and itis evident that the lobes of the fringes are inclined. This result is close to the fringes observed in real life cases. @ ©) Fig. 3.8 Photoelastic fringes for the biaxial loading case with analysis of (a) one term, and (b) two terms (Courtesy: Dr K. Ramesh, Dept. of Applied Mechanics, IIT Madras)Me Ciel aa ‘Companies 56 _ Elements of Fracture Mechanics The stress field [Eq. (3.32)] developed in this section for a biaxially loaded plate is usually claimed to be the stress field of an uniaxially loaded plate (Secs 3.2 and 3.3). What is the contribution of the far field stress o,, = 6? It does not cause any substantial change in the stress field in the vicinity of the crack tip because a, does not open up the crack. Invoking the principle of superposition, we can separate the biaxial stress (Fig. 3.9) to two configurations, (a) and (b). Configuration (a) is of importance to us; however, a simple solution to it is still not available. Configuration (b) does not try to open the crack but it does modify the stress field to a certain extent near the crack tip. However, the solution to configuration (b) is not simple and we usually neglect its effect in engineering solutions to fracture mechanics. the Lid E: I Peto Fig. 3.9 Separating biaxial loading case to configs (a) and (b) We close this subsection by realizing that the exact Westergaard solution is not available for a centre crack in a plate loaded uniaxially in the direction normal to the crack plane. However, a simple Westergaard function is available for biaxially loaded plates whose stress field in the vicinity of the crack tip is not substantially different from the case of an uniaxially loaded plate. Therefore, for most of the practical purposes, the solution of a biaxially loaded plate is employed for both uniaxially and biaxially loaded problems. 3.5.2, Mode II (Sliding Mode) A centre-cracked problem in an infinite plate for Mode II loading is considered, as shown in Fig. 3.3(a). Ithas been found that the following expression of the Airy Stress Function gis convenient: ®=-x,ReZy (3.38) where, Zy is a complex function and its form would be so chosen such that all the boundary conditions are satisfied. One can easily show that the above expression for ® satisfies the biharmonic equation and the stress field is given by Oy = 21m Zy +x Re Zip (3.39) Oy =~ x, Re Zi (3.39b) oy = Re Zy.-x, Im Zi, (3.390)The McGraw-Hill Companies Stress Intensity Factor 57 Further, stress components are converted to strain components and then integrated to yield for plane strain cases as w= yul Z(t) Z +x,ReZy] (3.40a) y= lle 20) ReZy ~ x2 1mZy:] (3.40b) For the case of Mode II, as shown in Fig. 3.3, the Westergaard function is given as: 2 Zn= a (G41) (@-#) which satisfies all the boundary conditions. They are similar to boundary conditions of the Mode I problem [Fig. 3.2(a)]. Like in the case of Mode I, we transform the origin to the crack tip by the transformation relation z = a + z, and obtain the simplified but approximate relation as +(na)!? Zu= 7 (3.42) (202,) Invoking the definition of the stress intensity factor [Eq. (3.36)], we obtain: Ky= 0 Vma Substitution of Z, and its derivative in Eqs (3.39a-c), leads to: -K, 3 (2nr) 2 22 o oe 30 = os © sin © cos 3.43 cos > sin > cos > (3.43) O2- cos 2 Losing sin%?) 2 2 22 anl™ Substitution of Zy, and Zin Eqs (3.40a and b) gives the displacement field for a plane strain problem as Ku fr of 20) uy = —| — sin —| 2 - 2v + cos* — (3.44% alae) 82 2) Gas) Kafr \ ey. 20) uy = —*| cos — || 1-2v—sin? — (3.44b) 2h le) ak 2) \e displacement field for a plane stress field can be obtained as (3.45a) (3.45b)Be a Red 58 Elements of Fracture Mechanics 3.5.3, Mode III (Tearing Mode) A large plate with a centre crack, subjected to a far field shear stress 633 = 1 (Fig.3.4), is considered. For this case of Mode III, m=0 =0 Us = (2, X2). For this displacement field, components of strain tensor turn out to be fi, = G2 = Oy = = 0 2 12; og Le 3° 2ax! > 2x, leading to, Gy = O33 = G3 = G2 =0 aw O13 = 2 Big = HS (3.46a) ox, aw yy = tt Ey =H 7 (3.46b) It should be kept in mind that the problem of Mode III is not a case of plane stress or plane strain. It is considerably simpler because many components of displacement, stress and strain are zero. We would, therefore, not be using the biharmonic equation. Instead, the problem will be solved with displacement component w as the dependent variable and then there is no need to worry for compatible conditions any more. Out of the three equilibrium equations, only the last one provides the non-trivial equation, aos , Aon _ 4 ax,” Ax, Converting it into displacement components using Eqs (3.46a and b) Pw | ew oe + SU 0 ay” Og which is the well known Laplace equation and is also written as Vw =0 The Westergaard approach is also applicable to this differential equation by choosing w in the form of 1 we Fle 2 (3.47) where Zz is a complex function of variable z, Again one can show that the above expression satisfies the governing differential equation. Also, substituting 7 in Eqs (3.46a and b), we obtain stress components as 63 = Im Zin (3.48a) G3 = ReZiy (3.48b)aCe mee ed Stress Intensity Factor _59 The form satisfies all the boundary conditions. Z’y: is transformed to the origin at the crack with relation 2 =a+ z, and after neglecting the small terms, we obtain: va Zee (3.49) me Oy? Using tra = Ki and expressing z, in polar coordinates, the equation is simplified to Kin ( 8 5) 5 7z| cos; isin (3.50) (ear 2 Substituting Zj, in Eq. (3.48), we obtain the stress field as (51a) (3.51b) Integrating Eq. (3.49) and substituting t Vira = Kyy, we have tum (2) "Kul! = ku(2)” (os8 risin’) which on substitution in Eq. (3.47), gives the displacement field as with 3.6 ConcLuDING REMARKS We determined the stress and displacement fields around a crack tip in infinite plates for all the three modes. These results are quite useful to a designer, because in many practical applications a crack in a plate is quite small in comparison to the lateral extent of the plate. The Westergaard’s approach adopted in this chapter is not the general way of solving problems. In fact, problems of infinite plates are easier to solve because stress fields far away from the crack are simple; we mainly focus our attention to the cracked faces and with some luck we may be able to guess a solution, which satisfies all the boundary conditions. In the case of work-components,The McGraw-Hill Companies 60 Elements of Fracture Mechanics which have their edge /edges close to the crack, the job of guessing a Westergaard function, which not only satisfies conditions at cracked faces but also meets the requirements at the edges, becomes, a difficult task. Problems dealing with bodies of finite dimensions will be taken up in Chapter 4. QUESTIONS 1. What is a singularity? What kind of singularity describes a stress field near the vicinity of a crack tip in LEFM? Is it expected to be different for elastic-plastic fracture mechanics? 2. Stress field is the same for plane stress and plane strain problems. Why is it not so for displacement fields? In problems of plates, stress components are expressed in the Cartesian coordinate system whereas the location at which stress is considered is defined in polar coordinates. Why is such a mixed approach adopted? 4, For many problems of practical applications, solutions of infinite plates are applicable. Justify the statement. 5. Displacement near the crack tip is determined by integrating strain components. Why do we equate integration constants to zero? 6. Mode I case has been solved for a biaxial case and its stress and displacement fields are taken to be approximately the same as of an uniaxial case. Justify. 7. Why do we not use biharmonic equation to solve Mode II problems for a centre crack in an infinite plate? PROBLEMS 1. Show that @ =x, ReZ, chosen for the Mode Il problem, satisfies the biharmonic equation. Determine stress components and displacement components (plane stress) in terms of Zy. 2. Fora centre crack in an infinite plate loaded in Mode II, determine stress components and displacement components (plane stress) near the vicinity of a crack tip in terms of Ky. 3. Show that w 1 A Im Zy , chosen for the Mode III problem for a centre-cracked infinite H plate, satisfies the Laplace equation Vw = 0, Determine stress components and all displacement components in terms of Zin. Also, determine stress and displacement fields in the vicinity of the crack tip in terms of Kyy 4, Ina large plate, a crack of length 2a is inclined with an angle a-with x-axis (Fig. 3.10). The plate is loaded in x, direction with 6, =. (i) Find the stress intensity factors (ii) For 6 = 80 MPa, 2a = 20 mm and c= 30*, determine K, and Ky Determine stress at point H of Problem 4, ifr = 1 mm and @ = 45°. 6. Determine the critical crack length in a centered-cracked plate, loaded in Mode I, if critical stress intensity factor K,, = 60 MPa Vm and far field stress is 120 MPa. 7. Determine stress components (6,,, Og, G,¢) and displacement components (1) in polar coordinates for plane stress of Mode I.The McGraw-Hill companies Stress Intensity Factor 61 ILL d ttt Fig. 3.10 The figure of Problem 4 REFERENCES 3.1 Irwin, G.R. (1958). Fracture, Handbuch der Physik, S. Flugge (ed.), Springer-Verlag, Berlin, Vol. VI, pp. 551-590. 3.2. Kanninen, M.F. and Popelar, C.H, (1985). Advanced Fracture Mechanics, Oxford University Press, New York. 3.3 Sokolnikoff, IS. (1956). Mathematical Theory of Elasticity, McGraw-Hill Book Company, New York. 3.4 Fung, ¥.C. (1965). Foundation of Solid Mechanics, Prentice-Hall Inc., Englewood, New Jersey. 5 Westergaard, HM. (1939). “Bearing Pressures and Cracks,” Journal of Applied Mechanics, 61, pp. A49-53. 3.6 Carrier, G.E., Krook, M. and Pearson, C.E. (1966). Functions of a Complex Variable, McGraw-Hill Book Company, New York. 3.7 Gdoutos, E-E. (2005). Fracture Mechanics—An Introduction, Springer, The Netherland. 3.8 Anderson, T.L. (2004), Fracture Mechanics: Fundamentals and Applications, CRC, Press-Book. 3.9 Sanford, RJ. (2003). Principles of Fracture Mechanics, Prentice Hall, Upper Saddle River. 3.10 Janssen, M., Zuidema, J. & Wanhill, RJ.H. (2004). Fracture Mechanics, Spon Press, Abingdon. 3.11 Ramesh, K. (2007). e-Book on Engineering Fracture Mechanics, IIT Madras, URL: http//apm.iitm.ac.in/smlab/kramesh /book_4.htm.The McGraw: feed Chapter 4 SIF of More Complex Cases Models are useful, but without exception, they have blind spots built into them. Deepak Chopra 4.1 OTHER APPLICATIONS OF WESTERGAARD APPROACH We have so far solved the basic problems of Modes 1 II and III in the previous chapter through the approach made available by Westergaard. The chapter defined stress intensity factor and determined stress and displacement fields near a tip of a crack in an infinite plate loaded with a far field stress. In this section, we would take up some more problems which are also important for the practical cases. 4.1.1 Wedge Loads on Cracked Surfaces Consider the problem of two wedge loads P (per unit length), acting symmetrically on each cracked Fig. 4.1 Symmetrically applied wedge loads on the surfaces of a crack