6. Conclusions
Using the microstructural analysis, were not detected differences between the intrinsic microstructures shown by specimens taken from fractured rods fail by different mechanisms, or between samples obtained from the same rods near fracture zones and on distant zones. Were not detected evidences of the contributions from original or microstructural defects due to material itself, or to processing such as possible responsible failure of cable, which should then be given to mechanical and environmental conditions during the service. For both conditions analysed in the fatigue tests (with and without initial corrosion) a great dependence was found between the fatigue life and the maximum stress in the fatigue cycle, i.e. a small variation in the applied stress causes a great variation in the fatigue cycle. The fractographic analysis showed that the failure of the cable occurred in two phases: at first the rods were being attacked by stress corrosion seriously affecting more than 50% of the rods, which progressively took them to fracture; when the number of remaining rods no longer met the efforts applied, they broke by ductile fracture.