PSI - Issue 38

Alok Gupta et al. / Procedia Structural Integrity 38 (2022) 40–49 Author name / Structural Integrity Procedia 00 (2021) 000 – 000

46

7

the fracture lines (at loc. 2 in Fig. 10(b)) which is due to formation sub-grains with low angle boundaries. The misorientation parameter depicted by the Kernel Average Misorientation (KAM) maps, typically used to represent degree of damage due to cyclic plastic strain [Kamaya (2012)], were also compared for these two locations. Misorientations above 7  were excluded in the KAM map to avoid grain boundaries contributing to the map. A higher KAM density is seen for loc. 2 (with a larger spread of green and yellow color in Fig. 10(d)) as compared loc.1 (Fig. 10(c)). The dislocations were preferentially distributed near the grain boundaries and sub-grain boundaries. From Fig. 10, formation of finer primary α ’ sub-grains and higher dislocation density seem to be contributing towards cyclic softening of SLM Ti-6Al-4V material. Additionally, some local micro-cracks also develop within the material in the areas of dislocations pile-up. Propagation of these microcracks under cyclic loading and their coalescence during the crack growth phase contribute to reduce the LCF life of SLM Ti-6Al-4V [Kamaya (2009)]. The higher mechanical strength (Table 1) was due to the needle shape martensitic α ’ grains of the SLM Ti-6Al-4V material as seen in Fig. 10(a) [Simonelli et al. (2014)]. With the expected small plastic strain levels in the BT1 test, any growth/coalescence of micro-cracks and dislocation density was less likely. The bracket successfully passed the initial design target of 3000 cycles in BT1 test which was done at the upper bound of the loading levels expected during normal operation of the engine. Under normal operation of the engine, the bracket experiences stresses much lower than the elastic limit. The BT2 test was carried out with a displacement range of ±4mm where the plastic strains levels were higher, which is evident in the shape of the BT2 test loops which were wider, see Fig. 7, than those for the BT1 test. Noting, that the features with geometric discontinuities, i.e. of stress concentration, are expected to be the locations of high stresses in the bracket and if a defect or micro-notch due to surface roughness is present at one of these locations, then it becomes a prime site for crack growth, eventually leading to fracture. The bracket under the BT2 test saw the failure of a straight strut at the 211 th cycle, when a 15% drop in force was observed which indicated that the bracket retained 85% of its original strength even after the failure of a strut. This suggested that the bracket has the much-needed redundancy in its load path and will continue to bear the lower level of loading expected once the extreme event has passed. 4.3. Relationship of property-performance The tensile (TT1 and TT2) and LCF coupon tests (LT1 and LT2) were useful to set the loading level and other test parameters for the BT1 and BT2 tests, which helped to prove fatigue performance of the bracket under cyclic loadings expected under normal operating environment and an extreme event of an aero-engine. The LCF coupon tests (LT1 and LT2) resulted in failures at a very low number of cycles and have shown cyclic softening behavior. The number of cycles to 1 st strut failure of the bracket were in similar order as noticed in the LT2 LCF coupon test (211 v/s 294 in LT2 test) which indicated that the total strains at certain locations in the bracket may be around 1.2%. A crack initiation and propagation at such a high strain is likely if the location is found with a defect or microcracks due to rough surface or a surface defect (LOF void or pore). The tensile and fatigue data at different loading levels (for both specimen and bracket tests) can also be useful in finite element simulation, model co-relation and development of life predictive method for the bracket, which should gradually remove the need for an expensive bracket tests in the future. 4.4. Appraisal on design of bracket The SLM process was used to build a weight optimized bracket in a ‘ struts & connector s’ shape. The performance of the bracket was successfully tested against the design target of 3000 cycles (minimum) without failure for LCF load condition in this study. In a parallel study by the authors, the HCF performance of the bracket has also been proven [Gupta et al. (2021b)]. Additionally, it has been demonstrated that in case of a strut failure during engine operation, the bracket would not suddenly loose its full load carrying capability due to its design involving multiple 4.2. Bracket performance