Development of Fracture Toughness Testing echniques for Non-sharp Defects

The novelty in this proposal is the development of novel inspection methods for flaw tip acuity characterisation and guidance for fracture toughness measurement in samples with non-sharp defects. The outcomes of this research programme will produce a step-improvement in fracture mechanics-based defect assessment methods, of genuine overall impact across a range of industries including nuclear; aerospace; oil and gas; wind power generation; and the defence sector.
SI assessment procedures conventionally categorise flaws either as infinitely sharp cracks or local thinning areas (LTAs) Defects failing the LTA geometry check - the flaw tip radius (r) must exceed its depth (a), r/a 1 - are assumed as sharp cracks and assessed by fracture mechanics principles. The LTA rule may be appropriate for general corrosion or mechanical gouges where there is a change in wall thickness and plastic collapse is assumed to be the failure mechanism. For non-sharp defects (e.g. welding/additive manufactured porosity, pitting corrosion defects), the failure mechanism is not, in principle, clearly defined.
For non-sharp defects, higher fracture toughness and load bearing capacity are expected compared to the crack case. Simplifying blunt defects into LTAs can be non-conservative, on one hand, and into sharp cracks can be excessively so, on the other. As a result, there is the need to develop procedures that account for the non-sharp defects more realistically.
A critical factor for achieving a complete characterisation of bluntness effect on failure is the development of representative and reproducible fracture testing guidance for notched specimens. It has been observed that cleavage fracture and ductile tearing from blunt tips is more complex than from fatigue pre-cracked standard specimens. The measurement and definition of initiation and crack growth is not as straightforward for specimens with non-sharp flaws, i.e. short cracks have been observed to be in competition on different planes. The introduction of a notch into a specimen, instead of a fatigue pre-crack, can be considered an in-plane constraint loss. It has also recently been observed that there is an out-of-plane constraint effect. Thus, specimens that would satisfy minimum thickness requirements for a fatigue pre-crack in typical fracture toughness tests (e.g. E-399) may result in non-conservative toughness measurements in notched-components.
Experimental notched fracture toughness tests for cleavage will be designed and performed on notched single edge notch bend (SEN(B)) specimens. Changes in sample thickness and notch tip acuity will be performed to achieve different levels of notch tip constraint and achieve plane strain conditions. The validation of the fracture toughness results requires a sufficiently large number of test results due to the inherent scatter associated with fracture toughness values. An S355 structural steel will be used for the entire initial testing program.
A parametric study of the effect of sample thickness and notch tip acuity using cleavage models for specimens tested will be developed. Modified boundary layer (MBL) models and SEN(B) models will be studied. 3D MBL models will be used to assess both the in-plane effects of notch tip acuity and out-of-plane effects of model thickness on stress distributions ahead of the crack tip. Two-parameter fracture mechanics (J-Q) will be used to characterise the in-plane stress fields under different levels of constraint and for various notch tip radius, thus using a unique framework to quantify the level of deviation of the stress/strain fields from high crack tip constraint conditions (Q=0) for both types of effects. Experimental outputs will be used to validate results from the SEN(B) models used in this task. The resulting toughness values from the experimental work will be used to capture the behaviour in the numerical models.