Sarvesh Pal

Rail squats: progress in understanding the Australian experience

Rail squats or studs, being a sub-surface crack in a track below a depression of the rail surface, have been studied in the Australian CRC for Rail Innovation project R3.105, from a number of perspectives. Examination of squats on track by the Rail Corporation of NSW has revealed statistics about their distribution. Examination with optical and scanning electron microscopy at the University of Queensland has shown the frequent presence of a brittle white etching layer (WEL) on the rail surface. This has led to successful adaptation of eddy currents to detect WEL rather than cracks. Tests at the University of Queensland have investigated whether the WEL can form below the normal 720℃ needed to form austenite. Examination of crack surfaces shows beach marks indicative of growth in modes II and III. Neutron diffraction testing of Australian rail has shown residual stresses in a railhead with a WEL, similar to those reported elsewhere, but with a broad surface layer in compression, not a narrow running band. Elasto-plastic finite element simulation has shown residual compression transforms into residual shear at a crack tip, which increases that due to plastic deformation from successive wheel passages, tending to encourage the crack to continue in a direction of sub-surface growth. The rate of growth of squats measured on-track in Sydney shows crack growth that corresponds to a power law with a low exponent, implying that the stress intensity at the crack tip is not a strong function of crack length. This is partly due to the localised nature of contact stresses, which allows the crack to grow beyond the region loaded. However, accounting for this effect leads to the prediction of a smaller reduction in growth rate than that observed. It is likely the low exponent reflects a smaller effect of water on crack growth, as the crack enlarges. There is also a redistribution of contact loading that occurs as a crack grows.

Neutron residual stress measurements in rails

Neutron News Volume 24 • Number 3 • 2013 9 Introduction Rails were among the fi rst objects of study by neutron diffraction strain measurement and the fi rst experiments were done as early as the late 1980s [1, 2]. This interest is easy to explain: the problem of rail fracturing is critical from the public safety point of view and the penetrating ability of neutrons suggested the possibility of breakthrough experiments and fast progress in this fi eld. It was well-established that residual stresses, both near-surface and interior, played a signifi cant role in the development of defects which led to rail failure. This suggested three distinct approaches of neutron diffraction strain measurement that could contribute to various problems of the rail industry. The fi rst method was to map the complete triaxial stress distribution non-destructively in the interior of an intact rail, ideally before and after signifi cant service. Another approach was to use slices, for example to characterize how different processing methods produce favourable or detrimental stress distributions in rails. A third technique was to make non-destructive measurements, but in critical and not very deep portions of rails, for example, to examine defects and their relation to rail failure in the top running surface of rails, e.g. “white layer” formation. In the case of the most direct and straightforward approach to measure stresses in the interior of an intact rail neutron beams would penetrate the whole thickness of the railhead (70–75 mm across) while the half-attenuation length in steel is only 6 mm. Therefore, neutron beam attenuation is critical so that experiment optimisation and planning become highly important issues even for the most intense neutron sources. Such optimisation would involve (i) adjustment of the neutron beam wavelength, scattering geometry and sample orientation to minimize the neutron beam fl ight path inside of the rail and (ii) balancing the choice of spatial resolution and number of mesh points and the available measurement time. Even with optimisation, measurement time is very signifi cant (many hours) and the number of measurement points might not be great (a hundred) with low spatial resolution (5 × 5 × 5 mm3). An approximate solution would be to use slices cut out of the continuous rail sections and this approach was used in the milestone experiments by P. Webster that were published in Neutron News in 1991 [3]. Slices appropriate for neutron penetration, 5–20 mm thick, either transverse, longitudinal or sometimes oblique, can be cut out of long section and measured much faster (minutes), with high spatial resolution (2 × 2 × 2 mm3) and covering larger rail cross-sections with large number of points (up to a thousand) than for an intact rail. This approach is suitable when there is a necessity to measure a long series of samples, i.e. providing important information on how different processes produce different initial stress distributions. However, the price for the convenience of working with thin (<10 mm) slices is that the stress component normal to the cut plane is completely eliminated, while all other stress tensor components are partially relaxed. Therefore, results of such measurements are not necessarily what railway engineers want to know: “what is the stress distribution in the original rail and how does it change with service?” Neutron residual stress measurements in rails

Revisiting Stress Corrosion Cracking of Steel in Caustic Solutions for Developing Cracking Susceptibility Diagrams for Improved Applicability

Stress corrosion cracking tests were conducted using Bayer solutions of different chemistry at different temperatures for extraction of alumina from bauxite ores. The validity of the commonly used caustic cracking susceptibility (CS) diagram for steels exposed to plain caustic solutions was assessed by testing the notched and precracked specimens. This study presents first results toward the development of a model susceptibility diagram for actual Bayer solutions, and for improved applicability of the traditional plain caustic diagram. For mechanistic understanding of caustic cracking, tests were also carried out under imposed electrochemical conditions.

Circumventing Practical Difficulties in Determination of Threshold Stress Intensity for Stress Corrosion Cracking of Narrow Regions of Welded Structures

Determination of the threshold stress intensity for stress corrosion cracking (KIscc) of narrow areas such as weld and heat-affected zone (HAZ) of a weldment is a nontrivial task because of the requirements of large specimens in testing by the traditional techniques and the difficulty of restricting crack propagation to narrow regions in such specimens. This article describes a successful application of the circumferential notch tensile (CNT) technique to determine the KIscc of narrow regions of the weld and HAZ. Also, the microstructure of the HAZ of the manual metal arc-welded steel was simulated over a relatively small length of specimens and its KIscc in a hot caustic solution was determined successfully. Intergranular stress corrosion cracking was confirmed with a scanning electron microscope.