Page 290 - Handbook of Materials Failure Analysis

286 CHAPTER 12 A nonlocal damage-mechanics-based approach

over the local damage model and these will be discussed in this section in the context
of simulation of fracture mechanics specimen with different geometry, size, and
loading conditions.
Initially, the standard 1T SEB specimen was simulated to obtain the load-
displacement and the fracture resistance behavior at room temperature, where the
fracture mode is purely of ductile nature. The effect of the FE mesh size (in the
crack-tip region) on the load-displacement behavior and J-R curve of the 1T SEB
specimen was studied using both local and nonlocal damage models. The objective
is to demonstrate that the results of nonlocal formulation are independent of mesh
size at the crack-tip and hence, this can be used further in simulation of fracture
toughness variation and its scatter in the DBTT region, where both ductile and cleav-
age modes of fracture compete with each other before the specimen fails by unstable
cleavage mode.
The effect of crack-depth of the specimen on the load-displacement response of
the 1T SEB specimens was later studied by comparing the results of a shallow-
cracked (a/W¼0.13) specimen with that of a deeply cracked (a/W¼0.53) specimen.
Different sizes (i.e., 1T, 2T, and 4T) of SEB specimens were also analyzed. The
effect of specimen geometry and loading condition on the load-deformation behavior
was studied by comparing the results of simulation of the 1T SEB specimen with that
of the 1T CT specimen. The effect of the use of symmetric boundary conditions on
the results of local and nonlocal models was also investigated. All these studies were
conducted initially at room temperature (upper-shelf region).
Subsequently, the scatter in fracture toughness and its variation with temperature
was also simulated for various types of specimens in the DBTT region and the effects
of the previously mentioned parameters (i.e., specimen geometry, size, crack-depth,
loading conditions, etc.) were also studied. Experiments were conducted on all these
different types of fracture mechanics specimens at room temperature as well as the
DBTT regime and the results of FE analysis were compared with those of experiment.
In order to demonstrate the effectiveness of the damage model in predicting the crack
growth in the actual dissimilar metal welded specimen, simulation of a 1T SEB
specimen with initial crack at the ferrite-buttering interface has been carried out.

5 MATERIAL AND EXPERIMENT
Experiments have been conducted on different types of fracture mechanics speci-
mens for study of ductile crack growth in upper-shelf, lower-shelf as well as DBTT
regime. The specimens have been machined from the pressure vessel steel DIN
22NiMoCr3-7. For FE analysis of the specimens, the material properties of the above
material have been used. This material is homogeneous from the point of chemical
composition and possesses good ductility. The true stress-strain curves of the mate-
rial at different temperatures are shown in Figure 12.1 and these are used in the FE
analyses. Other material properties including the Rousselier’s constants, character-
istic length parameter, and the different void volume fractions (i.e., initial, final, and