Page 228 - T. Anderson-Fracture Mechanics - Fundamentals and Applns.-CRC (2005)
P. 228

1656_C004.fm  Page 208  Thursday, April 21, 2005  5:38 PM





                       208                                   Fracture Mechanics: Fundamentals and Applications


                          Let us now introduce a moving coordinate system (x, y) attached to the crack tip, where
                       x = X − a(t) and y = Y. The rate of change of each wave potential can be written as
                                                  d ψ  i  =  ∂ ψ  i  − V  ∂ ψ  i
                                                  dt     t ∂    x ∂  (i = 1, 2)                  (A4.6)
                       where V(= −dx/dt) is the crack speed. Differentiating Equation (A4.6) with respect to time gives

                                                     ∂ 2 ψ    2  ψ  2  ψ ∂  ∂ ψ ∂
                                              ˙˙ ψ = V  2  i  − 2 V  i  +  i  − V ˙  i           (A4.7)
                                               i      x ∂  2  ∂∂     t ∂  2  x ∂
                                                              xt
                       According to Equation (A4.5) the first term on the right-hand side of Equation (A4.7) is proportional
                       to the stress tensor. This term should dominate close to the crack tip, assuming there is a stress
                       singularity. Substituting the first term of Equation (A4.7) into Equation (A4.4) leads to

                                                         ∂  2 ψ  ∂  2 ψ
                                                      β 2   1  +  1  =  0                       (A4.8a)
                                                       1  ∂x  2  ∂y  2


                       and
                                                         ∂  2 ψ  ∂ 2 ψ
                                                      β 2   2  +  2  =  0                       (A4.8b)
                                                       2  ∂x  2  ∂y  2

                       where

                                                        2               V   2
                                                       V
                                               β =−        and  β =−     
                                                 2
                                                                   2
                                                   1
                                                                      1
                                                       c
                                                1                2      c 2 
                                                        1
                       Note that the governing equations depend only on instantaneous crack speed; the term that contains
                       crack acceleration in Equation (A4.7) is negligible near the crack tip.
                          If we scale y by defining new coordinates, y  = β y and y  = β y, Equation (A4.8) becomes
                                                                   1
                                                              1
                                                                              2
                                                                         2
                       the Laplace equation. Freund and Clifton [32] applied a complex variable method to solve Equation
                       (A4.8). The general solutions to the wave potentials are as follows:
                                                               Fz
                                                        ψ = Re[ ( )]                             (A4.9)
                                                          1       1
                       and
                                                               Gz
                                                        ψ = Im[ ( )]
                                                          2       2
                       where F and G are as yet unspecified complex functions, z  = x + iy  and z  = x + iy .
                                                                       1
                                                                              1
                                                                                           2
                                                                                    2
                          The boundary conditions are the same as for a stationary crack: σ  = τ  = 0 on the crack surfaces.
                                                                                 xy
                                                                             yy
                       Freund and Clifton showed that these boundary conditions can be expressed in terms of second
                       derivatives for F and G at y = 0 and x < 0:
                                         1+ (  β  2 )  2  ′′( ) +  + ′′( )]Fx  −  + 2 [G ′′( )x  +  + ′′( )]G x  −  =  0  (A4.10a)
                                                                β [Fx
                                                                 2
                                                                   G
                                                    Fx
                                         2 [ ′′ ( ) − ′′ ( )]+  1 β Fx  +(  1 )  2  [ ′′ x  G  ( )] =  0  (A4.10b)
                                                                     ( ) − ′′ x
                                          β
                                           1     +       −             +       −
   223   224   225   226   227   228   229   230   231   232   233