The effect of notches and pits on corrosion fatigue strength.

An investigation has been undertaken to examine the fatigue behaviour of two martensitic steels in air and aggressive environments. The steels studied are, 18% Ni marageing steel and FV520B, the later being a stainless steel turbine blade material and the former being a marageing steel that suffers general corrosion in mild environments. Both steels were heat treated to give similar tensile strength.The design and manufacture of an autoclave allowed push-pull fatigue tests to be conducted in aggressive environments at elevated temperatures.Corrosion potential was monitored using a three electrode cell and was controlled during testing. Base-line fatigue tests were conducted with a range of constant corrosion potentials, using both notched and plain FV520B specimens. In addition fatigue tests with pulsed corrosion potential were performed to asses the effect of transient corrosion conditions on the corrosion fatigue strength. The pulsed tests were designed to simulate service transients in the oxygen content and general chemical hostility in the condensing steam environment during start-up and shut down of the steam turbine.Post test examination of fractured samples was performed using Scanning Electron Microscopy (SEM) and optical microscope techniques. The fractography results were used to quantify microstructural and fracture features of the steels.A model based on the size and geometry of the initial corrosion pitting has been proposed to asses the fatigue life of FV520B in an aggressive environment.The effect of pitting on the corrosion fatigue strength of FV520B has been modelled using linear elastic fracture mechanics (LEFM) type approach. The model has shown a good correlation between predicted fatigue lives with experimental results.The results suggest that the fatigue life is governed by the mechanical stress concentrating effect of the pits rather than the electrochemical damage caused by the environment.Finite Element Analysis (FEA) of the notch allowed calculation of the elastic stress intensity factor (K[t]) for the specimen geometry used. The experimental results together with numerical results of FEA were used to calculate of the notch strength reduction factor (K[f]) for the material. This has been used to derive the notch sensitivity factors (q) for both materials.The results of fatigue tests in air showed that although both materials have similar tensile strength their plain fatigue strengths are different. The sensitivity of the fatigue strength to notches was also found to be significantly different. The marageing steel showed a higher sensitivity to a notch than the FV520B.An empirical model has been proposed to quantify the notch sensitivity and the effects of various microstructural features on the fatigue strength. A model has been developed to predict the serviceable life of a peak hardened FV520B turbine blade subjected to aggressive low load conditions during start-up and non-aggressive high load conditions during continual running. The model is based on the conclusions suggested in the work of a threshold stress intensity factor being reached where a fatigue crack will grow from a corrosion pit at the root of a notch. The model is then used to highlight the life reduction caused to steam turbine blades due to increased numbers of start-up cycles.