PSI - Issue 2_A

Mikhail V. Bannikov et al. / Procedia Structural Integrity 2 (2016) 1071–1076 Aut or name / Structur lIntegrity Procedia 00 (2016) 000–0 0

1074 4

Fig. 2. Geometry of specimen. Values of sizes (in mm) 2R1= 5, R2= 15, L2=15, L1=17.49 were calculated to satisfy for resonant conditions by Bathias (2005). The ultrasonic generator excites the converter, which initiates periodic stress amplified by the horn. The peak displacements are located at the ends of the sample, where they can be measured by a dynamic sensor. The maximum stress is concentrated in the center of the sample. The distribution of stresses and displacements in the converter-amplifier-sample system with the coefficient of cycle asymmetry R = -1 is shown in Fig. 3, a. Fig 3, b. shows the sample with a thin layer of liquid gallium, which was applied to the previously polished and degreased surface.

a)

b) Fig. 3.The resonant fatigue machine. a) The principal scheme with stress and displacement distribution by Bathias (2005) b) The image of specimen with thin layer of liquid gallium. The results of tests of pure iron samples in air and in contact with the liquid metal are shown in fig. 4. The data analysis allows us to make a conclusion that for the examined material tested in air, there exists the presumed “fatigue limit” of 3•10 9 cycle loading to fracture with the stress amplitude of 160 MPa. However, the application of a thin layer of liquid gallium alloy on the surface of the sample reduces its fatigue life by more than 2 orders of magnitude: 6•10 6 – 2.3•10 7 cycles to failure. The process of fracture of samples of pure iron in the air was not of an ordinary character. Usually, the loss of resonant frequency of the sample in the ultrasound machine is an irreversible process and corresponds to the fatigue crack initiation. In the case of pure iron the loss of the resonant frequency and, consequently, shutdown of the machine (and thus the interruption of loading) does not always lead to the macroscopic fracture of the sample and nucleation of a crack. Such behavior of the sample is the result of irreversible changes in the structure of the material, which can be considered critical from the view point of the sample deformation endurance under ultrasonic loading. However, these changes are not associated with the formation of fatigue crack and mechanically correspond to a sharp increase of the dissipation ability of the sample and impossibility of its testing in the air-cooled regime provided in the fatigue-load machines. The measurements of the sample temperature during tests showed that in the final stage of the process the temperature abruptly increases up to 220 °C, which is accompanied by a decrease of the resonant frequency.