Abstract
Application of loads in two perpendicular directions, also known as biaxiality, has been recognized to have significant effects on the ability of a material to resist fracture, known as fracture toughness, normally defined for high-constraint conditions. There are contradictory results in literature; most propose that biaxiality cannot drive fracture toughness below a minimum value found in specific geometries used for material characterization. This can allow for more lenient assessment of components and longer life expectancies. However, some research proposes that biaxiality could further decrease fracture toughness potentially leading to high financial costs and safety risks. Initially historical data is used to investigate how biaxial loading could be analysed using standards that have been developed to assess the fitness for service of a component. These mostly use uniaxial fracture toughness data to assess the integrity of components, which is shown to be a safe and conservative practice for all biaxiality levels. Conservatism varies with flaw geometry and biaxiality, reaching zero for equibiaxially loaded through thickness cracks. Use of marginally higher fracture toughness values from different geometries or loading conditions result in potentially unsafe predictions and raise concern for this configuration. Biaxiality in literature has mostly been studied experimentally in combination with surface flaws. To decouple the two effects an innovative experimental program is conducted here on through thickness flawed rectangular and cruciform specimens loaded in uniaxial and biaxial bending respectively. Crack propagation shows that biaxiality is captured, while specimens show varying plasticity levels with some including considerable ductile tearing. Finite element analyses are conducted to analyse the tests. Biaxiality is shown to constrain the plastic flow and reduce fracture toughness in comparison to uniaxial loading. Thefracture toughness values estimated with the application of experimental displacements to the models, for both geometries, are much higher than those for geometries used for material characterization denoting lower constraint levels than them.
Finally, it is concluded that the combination of thickness and crack geometry of the steel
specimens tested here do not raise concern for the safety of a component in biaxial
bending containing through thickness cracks. There is a potentially significant margin of
conservatism in assessing such a component with high constraint data and considerably
higher fracture toughness values could be used instead.
| Date of Award | 27 Sep 2022 |
|---|---|
| Original language | English |
| Awarding Institution |
|
| Supervisor | Christopher E Truman (Supervisor) & Mahmoud Mostafavi (Supervisor) |