Table of Contents Table of Contents
Previous Page  86 / 352 Next Page
Information
Show Menu
Previous Page 86 / 352 Next Page
Page Background

82

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

0

10

20

30

40

50

60

70

80

90

100

initial crack length

a

0

= 9.4 mm

Experiment

CGM

von Mises

WM 18

F [kN]

CTOD [mm]

Figure 14:

F

-CTOD curves obtained experimentally, using von Mises criterion (without crack

growth) and the CGM (with crack growth) for joint width 18 mm

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

0

10

20

30

40

50

60

70

80

90

100

initial crack length

a

0

= 8 mm

Experiment

CGM without nucleation of secondary voids

CGM with nucleation of secondary voids

F [kN]

CTOD [mm]

WM 12

Figure 15:

F

-CTOD curves obtained experimentally and using the CGM, with and without

nucleation of secondary voids, for joint width 12 mm

The influence of an important dimension of the specimen, initial crack length

a

0

, is

analysed for two SENB specimens of joint width 6 mm, for

a

0

= 8 mm and

a

0

= 10.4 mm,

corresponding to the experimentally measured lengths.

F

-CTOD curves obtained

experimentally and using the CGM are given in Fig. 16. It is clear that the difference in

behaviour of these two specimens can be assessed using the CGM, and that damage

modelling in the ligament is in good agreement with actual material behaviour.