PSI - Issue 17

M. Carrera et al. / Procedia Structural Integrity 17 (2019) 872–877 M. Carrera, P. Lopez-Crespo, P.J. Withers/ Structural Integrity Procedia 00 (2019) 000 – 000

874

3

The use of XRD and other high energy techniques can provide meaningful information of the inner working mechanisms inside a growing fatigue crack (Withers, (2013), Kelleher et al. , (2010)). In this paper, XRD is used in conjunction with FEM analysis to study the plastic zone size and shape through the thickness. Plain strain and plain stress solutions are also used for validation of results obtained, as well as application of elastic-plastic fracture mechanics.

2. Experimental method

2.1. Q1N material and fatigue experiments

A tempered and annealed bainitic steel, similar to Q1N (HY80) described in Robertson (1994) is used in the experiments. Its chemical composition is as exposed in Table 1.

Table 1. Bainitic steel composition, expressed in weight %. The balance is Fe.

C Cu 0.16% 0.25% 0.31% 0.01% 0.008% 1.42% 2.71% 0.41% 0.10% Si Mr P S Cr Ni Mo

The basic material properties obtained in tensile test are as follow : yield stress σ y = 699MPa and ultimate strength σ u = 858MPa. The experiments were based on a CT sample, with dimensions W = 62.5mm and B = 12mm (ASTM E647-00). X-ray diffraction tests were done following a procedure described previously (Steuwer et al. (2010)). Energy dispersive mode configuration was employed, with 60x60µm beam aperture, so that the side resolution (x, y) was 60µm. Due to the small size of grains in the sample, this high resolution was possible. The length of the gauge volume in the through thickness direction was about 1.4mm. A mix of plain strain and stress is obtained from the test, since the measured results are volumetric. Nevertheless, plain strain conditions are assumed, as the main part of the analysed volume is from the mid plane of the specimen. During the fatigue cycles of the XRD test, the slight movement of the sample was carefully corrected and adjusted, especially during maximum loads at K max = 46.6 MPa√m. This allowed taking all strain maps with the sample as reference, instead of the laboratory device. Sample movement was estimated to be adjusted up to ± 50µm. Pre-crack ing of the sample took place along 51.000 cycles at 10Hz, with ΔK = 28 MPa√m and K min /K max = 0.03. Strain measurements were taken at two different crack lengths.

Figure 1: Flux diagram of the process to obtain plastic zone from strain fields maps, around a fatigue crack tip