D. I. Fletcher

Development of a machine for closely controlled rolling contact fatigue and wear testing

In order to gain detailed insight into the processes taking place during rolling contact fatigue (RCF) and rolling contact wear, the test machine used must be closely controllable and offer comprehensive data collection facilities. This paper describes a machine that has been developed to offer these facilities over a wide range of test conditions, and that has the facility for the early detection of cracks by an eddy current method. An existing twin-disk contact simulation machine has been comprehensively updated to enhance the range of tests possible and, most importantly, to offer close control over very short numbers of machine cycles, allowing the very early stages of the development of fatigue cracks and wear damage to be examined. Results of tests to examine the capabilities of the rebuilt machine and a study of the repeatability of results under two particular test conditions are presented.

Predicting the Life of Steel Rails

Abstract A model of plastic strain accumulation, wear, and rolling contact fatigue (RCF) crack initiation in rail steel has been developed. Local to the contact zone, material is subject to severe cyclic stresses taking it beyond yield and leading to incremental accumulation of plastic deformation (ratcheting). This model is based on a ratcheting law derived from twin-disc, rolling-sliding contact experiments and can simulate thousands of ratcheting cycles with corresponding strain hardening. The model is being further refined to account for detailed microstructural changes. Sections of worn and fatigued rail, removed from service, have been metallurgically analysed. To obtain further data on rail-steel deformation and RCF crack initiation, twin-disc tests have been performed using discs cut from across a railhead and wheel rim. Two heat treatments were applied to some rail discs to investigate the effect of pro-eutectoid ferrite phase distributions and volume fractions. Tests were run to failure (defined by an eddy current crack-detection system) and to percentages of fatigue lives. Micro- and nano-hardness tests, and microstructural observations, have been used to suggest a micromechanism of fatigue crack initiation for the model. Application of this model will contribute to reduced maintenance costs and an improved understanding of RCF development.

A simple method of stress intensity factor calculation for inclined fluid-filled surface-breaking cracks under contact loading:

Abstract A simple method has been developed for the calculation of stress intensity factors for fluid-filled cracks under contact loading (typical of rolling contacts) and results compare well with existing data from more complex models.

Equilibrium of crack growth and wear rates during unlubricated rolling-sliding contact of pearlitic rail steel

Abstract It is generally accepted that large rolling contact fatigue cracks in rails do not develop during unlubricated rolling-sliding contact, and damage under these conditions is restricted to wear of the rail steel. However, close examination of a worn rail steel surface reveals the presence of a multitude of wear flakes, the roots of which closely resemble shallow rolling contact fatigue cracks. Experiments have been conducted under unlubricated rolling-sliding conditions to examine the early development of flakes, or cracks, using a laboratory-based, twin-disc test machine to simulate the contact pressure and slip characteristic of the contact between a rail and a locomotive driving wheel. Small defects were found after as few as 125 unlubricated contact cycles. It was found that an equilibrium between crack growth rate and surface wear rate was established after approximately 10 000 cycles, leading to a shallow steady state crack depth. Initial crack growth by ratchetting (accumulation of unidirectional plastic strain until the critical failure strain of the material is reached), followed by shear stress-driven crack growth described by fracture mechanics, was found to be a sequence of mechanisms in qualitative agreement with the observed crack growth and steady state crack depth.

The role of wear in enhancing rail life

For many engineering components the primary problems to be prevented during operation are wear and fatigue crack growth. However, in many cases one of these mechanisms dominates to the extent that the other may be neglected. In cases where both processes of almost equal importance in determining component life their interaction or coupling becomes critical. This paper presents a generic introduction to the concept of interaction between wear and fatigue in the failure of an engineering component. To illustrate the topics discussed a detailed example is presented of wear-fatigue interaction in the failure of a railway rail.

A simple method of stress intensity factor calculation for inclined surface-breaking cracks with crack face friction under contact loading

Abstract A simple method has been developed for the calculation of stress intensity factors for cracks growing by a shear mechanism under contact loading (typical of rolling contacts) which includes the consideration of friction between the crack faces. The results compare well with the existing data from more complex models.

The effect of intermittent lubrication on the fatigue life of pearlitic rail steel in rolling-sliding contact:

Abstract Railway flange lubrication can be unreliable, resulting in intermittent rather than continuous lubrication. Twin-disc contact simulation tests were carried out to investigate the influence of intermittent lubrication on rail steel fatigue life, using both a colloidal suspension of molybdenum disulphide in an oil carrier fluid (similar to a commercial flange lubrication product) and water. A brief interruption of the lubricant supply during molybdenum disulphide lubrication produced rapid surface failure of the rail steel, which was followed by rapid contact fatigue crack growth upon reapplication of the lubricant. During water lubrication the interruption of the lubricant (water) supply did not produce such rapid crack growth. The results were found to be in qualitative agreement with the predictions of the three-mechanism model of rolling contact fatigue, which combines the mechanisms of ratchetting (accumulation of unidirectional plastic strain until the critical failure strain of the material is reached) and the fracture mechanics-based mechanisms of shear stress driven and tensile fluid assisted crack growth. Analysis of the test results using this model revealed crack face friction to be an important factor controlling crack growth, which may be responsible for the difference in behaviour between contacts lubricated with the molybdenum disulphide-based lubricant and those lubricated by water.

Rapid method of stress intensity factor calculation for semi-elliptical surface breaking cracks under three-dimensional contact loading

Abstract Fast methods of stress intensity factor calculation for inclined surface breaking cracks under contact loading are presented. Cracks are loaded in normal and tangential traction by a three-dimensional Hertzian elliptical contact patch, and friction between the crack faces is considered. Stress intensities are calculated from Greens functions originally developed for two-dimensional cracks through application of stresses on a plane below the three-dimensional contact patch, in place of those previously considered for a two-dimensional contact. This approach gives the method great speed advantages over fully three-dimensional methods. Both semi-circular and semi-elliptical cracks are examined. The validity of the approximations and the results are judged by validation with results from alternative fully three-dimensional cases. Very good agreement is found between trends in stress intensity factor with changes in crack size and applied tractions. Absolute values of stress intensity factor agree well for semicircular and shallow semi-elliptical cracks, but values were below those of the reference case for deep, narrow semi-elliptical cracks. Calibration of the model to overcome this under-prediction is discussed. A case of special value in railway rail-wheel contact modelling is that of a contact offset to the side of a crack, representing the wheel running alongside rather than directly across an existing crack. This configuration results from the common procedure of grinding the rail to change or maintain its cross-sectional profile. The three-dimensional contact patch methods presented here enable this case to be modelled while retaining very fast running times for the calculations.

Visualization and Modelling to Understand Rail Rolling Contact Fatigue Cracks in Three Dimensions

The article presents an extensive survey of experimental data on rolling contact fatigue (RCF) crack shape and propagation characteristics in rails removed from service, where such cracks are angled to the rail axis. The data include re-analysis of previously published experimental data to extract crack shape information and new experimental work on crack shapes at different stages in the early RCF life. Periods from initiation (ratcheted ‘flake cracks’) have been considered through very early growth to the limit of one prior austenite (PA) grain and on to rail-head visual cracks. Techniques included multi-sectioning through single cracks and crack zones, on used rail and test discs, to build up real three-dimensional (3D) data on crack shapes and propagation characteristics. This data have been compared with the UK rail system guidance charts relating to visual crack length and respective vertical depth; all data fell within the indicated guidance zones. The configuration of such angled cracks, typically found in curves, so aligned due to the vector of both lateral and longitudinal traction, rather than just axially, was identified as an important case for modelling. A fracture mechanics-based model has been developed to predict modes I and II stress intensity factors for such cracks covering multiple PA grains. An important geometry effect is revealed by which a contact approaching a crack angled to the rail axis is effectively ‘offset’ from the approach direction considered in 2D models, thereby resulting in lower predicted peak stress intensity factor values, compared with 2D, for the prediction of crack growth rates.

Rail surface fatigue and wear

Abstract: rail fatigue and wear are degradation processes which drive rail maintenance and replacement, and hence have economic importance in railway operation. The underlying mechanisms of these processes are described, and the importance of their interaction in determining the life of the rail is discussed. Methods for detailed prediction of crack growth rate and rail wear rate are introduced, providing a link between vehicle loads and the rail damage produced.