The effect of specimen and flaw dimensions on fracture toughness: Dissertation

Fracture toughness values obtained from standardized tests, with strict specimen size requirements to produce specimen size independent results, are used in the process of fracture toughness assessment of structural components. Often results from the standardized tests can not be applied to real flaws due to different flaw geometries in test specimen and structure or the properties of the examined material are such that strict specimen size requirements of the standards are in practise impossible to fulfill. In both cases the main cause to problems arises from combined constraint characteristics of the flaw and specimen type. In this study a wide range of specimen and flaw dimensions were examined both by experimental tests and computational finite element analysis for constraint assessment. The methods for taking into account specimen and flaw dimensions are introduced and applied to investigated configurations. These methods include constraint correction or indexing methods, namely T-stress, Q-parameter and small-scale yielding correction (SSYC) and statistical treatment for specimen size effect in case of cleavage fracture. Experimentally studied configurations were Charpy-size impact specimens with a/W ranging from 0.05 to 0.5, static 3-point-bend (3PB) specimens with thickness ranging from 5 to 300 mm and plates in 4-point-bending (4PB) with elliptical surface cracks on the tension side. All flaws were fatigue precracked. Computationally analysed configurations included 3-dimensional surface-edge notched 3PB specimens (SE(B)) with deep (a/W = 0.5) and shallow (a/W = 0.1) crack configurations and compact tension (CT) specimens with a/W = 0.6. The W/B ratios for SE(B) and CT specimens were 1, 2 and 4. The W/B = 2 configuration was also analysed as side grooved. The strain hardening exponent (n) of the deformation plasticity material model had values 5, 10 and 20. Computationally analysed were also two sizes of elliptical surface cracks in a 4PB plate with material strain hardening exponent n = 10. For the 4PB plates a formula for fracture toughness (Jc) calculation was proposed. The conducted 3-dimensional finite element analyses of the cracked configurations were more detailed than any analysis in previously published literature. Thus analyses revealed a great deal of new information of the in-plane constraint behaviour inside the specimen. Toughness scaling models (= SSYC) and J-Q trajectories were created and their evolution as functions of specimen dimensions, material hardening properties and applied loading were investigated. The SSYC and J-Q approach were found to describe constraint properties adequately. This finding received partial verification from the experimental tests, where SSYC was found to be superior in comparison to Q-parameter and T-stress. The statistical treatment for specimen thickness description was very promising and a development for that method was proposed. Based on the results, the differences in apparent fracture toughness values obtained from various different specimen configurations can be better understood and taken into account.