Many service components in power generation and aerospace industries operate at high temperatures and stresses that make them susceptible to creep deformation and damage. Their complex geometries and load multi-axiality are often treated only approximately in assessing their structural integrity via assessment codes that are based on standard creep tests. For example, the forward creep (defined here as constant load creep) test of round bars is not a true representation of the stress state that service components generally experience. The experiments conducted in this work used notched bar specimens to simulate the effect of stress triaxiality. The results from these experiments were then used to validate a well-established creep ductility exhaustion damage model. Although the damage model is largely based on uniaxial creep rupture tests, it has been previously adapted so that it can be applied to more complex states of stress. Rupture calculations were conducted prior to experimental testing to obtain an estimation of the duration of the experiments. The finite element simulation results, which utilised previously developed creep deformation and damage models, were then compared to the experimental data. It was shown that the model predicted the correct trend for the creep deformation and failure of the specimens and primary, secondary and tertiary creep behaviour of notched bars could be captured. The tests imply that the effective creep ductility was smaller at lower stresses, i.e., at slower strain rates creep strain was more damaging.

Motivated by the need to more accurately account for real, in-service, operating conditions, this paper aims to investigate whether creep strain accumulated at different strain rates is equally damaging. Previous research has suggested that creep strain is more damaging when accumulated more slowly in creep of notched bars. The research presented here seeks to address this question by considering the accumulation of creep strain during stress relaxation of notched bars. Repeat stress relaxation tests with varying dwell lengths were conducted so that the relative damaging effects of the early, rapid accumulation and later, slow accumulation of creep strains could be compared. Another aim was to determine how a lower test temperature affects this creep strain accumulation. In repeat relaxation tests the load is reestablished repeatedly after relaxation dwells of equal duration, until rupture of the specimen occurs. The material used was an ex-service powerplant stainless steel Type 316H. Notched bar specimens were used to introduce stress triaxiality at the notch tip to imitate the multiaxial loads plant components are subjected to during in-service operation. The stresses and strains in the specimens were then assessed using finite element analysis; a user subroutine was implemented so the onset and propagation of creep damage could be simulated throughout the specimens’ creep life. The research found that the material in question had a lower creep ductility at 515°C than at 550°C. The research also showed that creep strain accumulated rapidly at the start of a dwell is significantly less damaging than creep strain accumulated more slowly towards the end of the dwell.

No abstract