HIGH TEMPERATURE LOW CYCLE FATIGUE AND CORRELATION TO CRACK GROWTH RATE IN TWO NICKEL BASE SUPERALLOYS (POWDER, PHASE ANALYSIS, HOLD TIME EFFECT, PREDICTION, OXIDATION)

Low cycle fatigue of two advanced nickel base superalloys, PM/HIP Astroloy and PM/HIP Rene'95, was studied at 650(DEGREES)C and 725(DEGREES)C. These alloys are potential candidate materials for aircraft turbine disk applications. The main purpose of this study was to determine the relative importance of creep and oxidation that might occur at high temperature.; Fully reversed strain controlled tests were performed at a frequency of 0.33Hz in high purity argon. Creep was simulated by superposition of 2 or 5 min hold times at the maximum tensile strain. Microstructural features, fracture characteristics and dislocation structures were examined by phase separation, SEM, EDAX and TEM techniques. The nature of the crack tip deformation also was examined by AES.; Employing a tensile hold was more damaging than raising temperature. Raising temperature from 650(DEGREES)C to 725(DEGREES)C did not change the transgranular crack propagation mode, whereas employing hold times caused mixed transgranular (TG) + intergranular (IG) propagation. Crack initiation was also changed to IG by employing a tensile hole which was otherwise TG. In Rene'95 pores were the major crack initiation site, while in Astroloy second phase particles, such as borides and grain boundary (gamma)', were observed at the origin.; Dislocation structures were similar in both types of tests. However, occasional stacking faults were found in hold time tests. Examination of secondary cracks showed no apparent creep damage. Oxidation in high purity argon appeared to be the major factor in LCF life degradation due to hold times.; Based on Kaisand-Mowbray model, an interactive computer program was developed to predict fatigue crack growth rate from low cycle fatigue data. With that program, the model was tested for several nickel base superalloys. Excellent agreement between the measured and predicted crack growth rate was observed. A sensitivity analysis showed that initial crack length was the most significant parameter in the prediction of crack growth rate.