Computer Simulation of Fast Crack Propagation and Arrest in Steel Plate with Temperature Gradient Based on Local Fracture Stress Criterion

2009 ◽  
pp. 617-617-24 ◽  
Author(s):  
S Machida ◽  
H Yoshinari ◽  
S Aihara
1991 ◽  
Vol 22 (10) ◽  
pp. 2287-2296 ◽  
Author(s):  
J. H. Chen ◽  
G. Z. Wang ◽  
Z. Wang ◽  
L. Zhu ◽  
Y. Y. Gao

1990 ◽  
Vol 38 (12) ◽  
pp. 2527-2535 ◽  
Author(s):  
J.H. Chen ◽  
L. Zhu ◽  
H. Ma

Author(s):  
C. Jacquemoud ◽  
S. Marie ◽  
M. Nédélec

In order to define a design criterion for graphite components, it is important to identify the physical phenomena responsible for the graphite fracture, to include them in a more effective modelling. In a first step, a large panel of experiments have been realised in order to build up an important database; results of tensile tests, 3 and 4 point bending tests on smooth and notched specimens have been analysed and have demonstrated an important geometry related effects on the behavior up to fracture. Then, first simulations with an elastic or an elastoplastic bilinear constitutive law have not made it possible to simulate the experimental fracture stress variations with the specimen geometry, the fracture mechanisms of the graphite being at the microstructural scale. That is the reason why a specific F.E. model of the graphite structure has been developed in which every graphite grain has been meshed independently, the crack initiation along the basal plane of the particles as well as the crack propagation and coalescence have been modelled too. This specific model has been used to test two different approaches for fracture initiation: a critical stress criterion and two criteria of fracture mechanic type. They are all based on crystallographic considerations as a global critical stress criterion gave unsatisfactory results. The criteria of fracture mechanic type being extremely unstable and unable to represent the graphite global behaviour up to the final collapse, the critical stress criterion has been preferred to predict the results of the large range of available experiments, on both smooth and notched specimens. In so doing, the experimental observations have been correctly simulated: the geometry related effects on the experimental fracture stress dispersion, the specimen volume effects on the macroscopic fracture stress and the crack propagation at a constant stress intensity factor. In addition, the parameters of the criterion have been related to experimental observations: the local crack initiation stress of 8MPa corresponds to the non-linearity apparition on the global behavior observed experimentally and the the maximal critical stress defined for the particle of 30MPa is equivalent to the fracture stress of notched specimens. This innovative combination of crack modelling and a local crystallographic critical stress criterion made it possible to understand that cleavage initiation and propagation in the graphite microstructure was driven by a mean critical stress criterion.


Author(s):  
John Knott

This paper describes characteristics of transgranular cleavage fracture in structural steel, viewed at different size-scales. Initially, consideration is given to structures and the service duty to which they are exposed at the macroscale , highlighting failure by plastic collapse and failure by brittle fracture. This is followed by sections describing the use of fracture mechanics and materials testing in carrying-out assessments of structural integrity. Attention then focuses on the microscale , explaining how values of the local fracture stress in notched bars or of fracture toughness in pre-cracked test-pieces are related to features of the microstructure: carbide thicknesses in wrought material; the sizes of oxide/silicate inclusions in weld metals. Effects of a microstructure that is ‘heterogeneous’ at the mesoscale are treated briefly, with respect to the extraction of test-pieces from thick sections and to extrapolations of data to low failure probabilities. The values of local fracture stress may be used to infer a local ‘work-of-fracture’ that is found experimentally to be a few times greater than that of two free surfaces. Reasons for this are discussed in the conclusion section on nano -scale events. It is suggested that, ahead of a sharp crack, it is necessary to increase the compliance by a cooperative movement of atoms (involving extra work) to allow the crack-tip bond to displace sufficiently for the energy of attraction between the atoms to reduce to zero.


Author(s):  
Amaury Bousquet ◽  
Stéphane Marie ◽  
Philippe Bompard

The integrity assessment of Reactor Pressure Vessels, mainly based on crack initiation, can be completed by studying crack propagation and arrest. Whereas engineering approaches do not take into account dynamic effects, these effects are important in unstable cleavage crack propagation, arrest and possible propagation re-initiation events. This study deals with physical mechanisms of cleavage crack propagation and numerical computations related to brittle fracture in the framework of local approach to fracture. Experiments were carried out on thin CT 25 specimens made of 16MND5 PWR vessel steel at five temperatures (−150°C, −125°C, −100°C, −75°C, −50°C). Two kinds of crack path, straight or branching path, were observed. Branching cracks appear for the highest critical loadings at initiation, that increase the elastic stored energy and the effect of plasticity. The elastic-viscoplastic behavior of the ferritic steel was studied up to a strain rate of 104 s−1 and taken into account in the numerical simulations. The eXtended Finite Element Method (X-FEM) was used in CAST3M FE software to model crack propagation. Numerical computations combine a local non linear dynamic approach with a RKR type fracture stress criterion. The different physical micro-mechanisms, involved in cleavage fracture, were examined by the means of SEM fracture surface analyses at different temperatures and strain rates for the two kinds of crack path. The links of the critical fracture stress with both temperature and strain rate for straight crack path as well as analyses of branching crack phenomena were considered by the means of Scanning Electron Microscopy (SEM) fracture surface analyses, 3D quantitative optical microscopy and FE computations in order to aim at a robust physical justification of the propagation model which has already been developed at CEA in the frame of the B. Prabel PhD.


Sign in / Sign up

Export Citation Format

Share Document