Fatigue Crack Paths 2003

0,345

μθ=04.01°

-0,210

point load

0,0148260

θμ=40.01°

point load

a/b=100

a/b=100

a/b=11.75

a/b=11.75

K I I/ K 0

K I / K 0

0,12

a/b=3.00

a/b=3.00

L/a

a/b=0.75

a/b=0.75

-1 -0,5

0 0,5 1 1,5

2

-3 -1,5

0 1,5

3 4,5

6

a)

L/a

0,456

point load

-00,210123

θμ=04.05°

0,024680

point load

θμ=04.05°

a/b=100

a/b=100

a/b=11,75

a/b=11,75

K I / K o

K I I / K 0

a/b=3.00

a/b=3.00

a/b=0.75

a/b=0.75

L/a

-1,5

1,5

-1

-0,5

0 0,5

1

1,5

-3

-1,5

0

3 4,5

6

L/a

b)

Figure 3. KI and KII vs L/a for different a/b ratios, μ=0.1 (a) , μ=0.5 (b).

For a Hertzian load, positive values of KI (even though with low values) can be

produced also for L/a<0 (only zero values are expected in these conditions under point

like load [13]). The plateau of KII vs L/a is a consequence of the sticking in the closed

portions. This phenomenon can be observed at the lower a/b ratio for μ=0.1, whereas it

appears also at larger a/b ratios for μ=0.5.

b)

a)

0,6

0,6

μ=0.1 θ=40°

μ=0.5

0,5

0,5

θ= 40°

0,4

0,4

0,3

0,3

K I I / K 0

K I I / K 0

-0,21012

0,2

0,1

point load

point load

0

a/b=100 a/b=01.17.575

a/b=100

a/b=11.75

-0,1

a/b=0.75

-0,2

0,02

0,04 0,06

0,08

0

0,05

0,1

0,15

0

KI/K0

KI/K0

Figure 4. KII vs KI at different a/b for μ=0,1 (a) and μ=0,5 (b).