Crack Paths 2009

x 1 0

p

,

p

0

#1

#2

#1

0

8

#2

y/W 0

9

6

da 1.0

8

dN

4

7

2

6

0

0.5

5

42

Tip 1

Tip2

1000

0

2000

a)

b)

4

cycles,N

0.01

0.02

1.0

0

x/W

∞y th ε ε / = 1.0 and

x∞ y∞ ε ε / = 0.6: a) The

Figure 5. Crack growth rate for the case with

branched crack, and b) the branches’ growth rates per cycle, da/dN, normalised with

max(da/dN), during 2000 cycles.

2

0

#1

#3

#3

#1

0.004 y/W- 0

dadN10.02

8

#2

#2

6

24

Tip 1

Tip2

Tip3

1000

2000

0

a)

b)

0

cycles,N

0.01

0.02

-0.008 0

x/W

∞y th ε ε / = 0.5 and

x∞ y∞ ε ε / = 0.6: a) The

Figure 6. Crack growth rate for the case with

branched crack, and b) the branches’ growth rates per cycle, da/dN, normalised with

max(da/dN), during 2100 cycles

C O N C L U S I O N S

Branching during dissolution driven crack growth is studied using an adaptive FE

method. A strain-assisted evolution law is used for the inherently blunted crack. No

criterion for crack growth is needed as for a sharp crack, neither for the growth

direction.

The inherent sensitivity for perturbations of strain along the crack front is typical for

some types of real stress corrosion cracks, for which variations of strains and could be

reasoned to arise due to microstructure and local electro-chemical conditions.

1079