I.R. McColl

Lubricated fretting wear of a high-strength eutectoid steel rope wire

Abstract The fretting wear behaviour of heavily work-hardened eutectoid steel wire is an important issue in the construction and usage of locked coil steel ropes. These ropes, which are used for high duty applications in aggressive environments, such as those encountered in the mining industries, can exhibit a variable service life. Earlier papers examined the design features of this class of rope and aspects of the operating regime, which may contribute to this variable life. The as-drawn surface of the wire was determined to play a significant role in the fretting behaviour, initially suppressing wear and friction. In this paper, the influence of low viscosity oils, with and without graphite additions, on the fretting behaviour of the as-drawn wire are examined. Oil bath lubrication suppresses effectively wear and friction throughout the tests. Of greater significance, however, a smear of oil, more typical of service conditions, is also effective, at least during the early stages of testing. The addition of graphite to this smear of oil aids its retention, and its ability to lubricate and form a shield around the fretting interface. The presence of oil inhibits the ingress of oxygen and consequently, at higher normal forces, when significant breakdown of the oil film occurs, micro-welding and tearing of the clean metal surfaces follows. Even so, wear and friction are still suppressed effectively.

An investigation of the fatigue and fretting performance of a representative aero-engine spline coupling

The fatigue behaviour of a representative high-performance aero-engine spline coupling is investigated under test conditions designed to simulate in-service conditions. The test load cycles consist of major cycle torque and axial load, simulating maximum thrust, combined with minor cycle rotating bending moment and fluctuating torque, simulating life-limiting conditions at take-off. The objective of the study is to develop understanding of the fatigue behaviour of the coupling over a range of loading conditions, including interaction between low-cycle fatigue, fretting fatigue and fretting wear. This information is necessary for the development of fatigue and fretting-fatigue life prediction techniques. The test results are interpreted with the help of three-dimensional finite element models, which include the frictional contact between the spline teeth.

Macroscopic fretting variables in a splined coupling under combined torque and axial load

A three-dimensional study of frictional contact in a helical splined coupling for the assessment of macroscopic fretting variables is presented. The study is based on an experimentally validated finite element model of the coupling under combined torque and axial loads. The effect of axial profile modification for reduced contact stresses in spline teeth and the effect of friction coefficient are considered. The motivation for the work is the need for representative information about fretting variable distributions in splined couplings for the development and application of simplified fretting test configurations.

Fretting damage in locked coil steel ropes

Abstract Locked coil steel ropes are used for high duty applications in aggressive environments, such as those encountered in the mining industry. The service life of this class of rope can be variable. This paper examines the design features of the rope, and the operating regime, which may contribute to this variability in service life. These ropes are constructed from cold drawn eutectoid and hypo-eutectoid steel wires. A rope is examined after a period in service close to its design life. Of the degradation mechanisms considered, fretting is found to produce significant wear at points of contact between the layers of the rope, near its outer surface. Initiation of fretting wear is encouraged by the breakdown of the added lubricant and the magnitude of the contact stresses. Two further papers will examine the fretting behaviour of a typical eutectoid steel wire used in this class of rope. In the first of these papers the residual lubricant coating from the wire manufacturing process is identified as playing an important role in the fretting behaviour. The second paper covers the influence of low viscosity oils, and solid additives, on the fretting behaviour.

The role of debris in the fretting wear of a SiC reinforced aluminium alloy matrix composite

The addition of 20 vol.% of 3mm SiC particles to a 2124 Al‐Cu‐Mg alloy matrix can achieve appreciable improvements in fatigue strength, by 50%, and fretting wear behaviour against a 0.4% carbon steel, over sliding distances up to 80 m. Over greater distances, the fretting performance is inferior to that of the matrix alloy. However, anodising of the composite, to a thickness of 10 mm, gives more than a factor of three reduction in wear, at least for sliding distances up to 800 m. In this paper, the role of debris in the fretting wear of the composite, for strokes of 40, 80 and 120 mm, is examined. Tests were carried out using a crossed flat-against-flat specimen arrangement. Localised adhesive transfer, from the composite to the steel counterface, occurred during the early stages of fretting, particularly at the lowest value of stroke. It was sufficient to separate the contacting surfaces, so that wear was initially patchy. As fretting continued, layers of compacted debris were formed. Cross-sectional scanning electron microscopy/X-ray analysis revealed the presence of Al, Fe, Si, C and O, but an absence of atomic contrast in the majority of layers, suggesting that, at a minimum, intimate mixing of composite and counterface debris had occurred. Debris particles ejected from the contact area increased in size with increasing fretting stroke. Transmission electron microscopy of these particles, after tests at all values of stroke, showed them to be agglomerations of sub-micron particles containing iron and possibly a spinel, FeAl2O4 or Fe3O4, suggesting that a degree of mechanical alloying had occurred. The matrix alloy, Al2O3 and Fe2O3 were not discernable, although SiC particle remnants were occasionally found.

Fretting wear of a high-strength heavily work-hardened eutectoid steel

Abstract This is the second paper in an informal series of three on the fretting behaviour of a heavily work-hardened eutectoid steel wire, used in the construction of locked coil steel ropes. The fretting behaviour of the wire, both in the service surface condition and after mechanical abrasion, is examined. Residual solid die lubricant, from the wire drawing manufacturing process, present both on the surface of the wire and as a reservoir in the drawing marks is determined to play an important role in the fretting behaviour. In particular, it is capable of maintaining the coefficient of friction at a low value, at least during the early stages of fretting.

Fatigue life prediction for a barrelled spline coupling under torque overload

Abstract Aeroengine spline couplings experience a wide range of loading conditions leading to contrasting service life limiting phenomena, including fatigue, fretting fatigue and fretting wear. Highly loaded couplings may employ incomplete contact axial profiles, while the contact geometry transverse to the spline axis is nominally complete with theoretical stress singularities at the contact edges. Life assessment of such components is consequently complex. The effect of torque overload conditions on the fatigue life of a barrelled, aeroengine type spline coupling is investigated experimentally. The experimental results are interpreted using three-dimensional finite element analyses, incorporating frictional contact and elastic-plastic material behaviour and the results of simple tension-tension fatigue tests. Torque-life and finite element predicted stress-life relationships are generated for spline life prediction purposes. Good correlation is obtained between the spline coupling and simple tension-tension fatigue test results, interpreted via the finite element predicted stress ranges.

Fretting wear of a fine particulate reinforced aluminium alloy matrix composite against a medium carbon steel

Abstract Particulate reinforced aluminium alloy matrix composites, produced by a powder metallurgy route, can show improved specific properties, including elastic modulus and tensile and fatigue strength, over monolithic alloys. However, these materials are difficult to join using metallurgical processes without impinging on their mechanical performance. Mechanical fasteners or adhesives, which are normally used to join these materials, can result in fretting wear problems under vibration and fatigue regimes. The fretting wear performance of a 2124 AlCuMg alloy reinforced with 17 vol.% of 3 μm silicon carbide particulate is compared with an equivalent monolithic 2024 AlCuMg alloy, both in the T4 condition, when tested against a cold-worked medium carbon steel. During the early stages of fretting the composite shows significant advantages, reducing the degree of debris transfer to the counterface. However, as fretting progresses the performance of the composite becomes inferior to that of the equivalent monolithic alloy. At high numbers of fretting cycles the degree of direct contact between the fretting faces is minimal and wear progresses via layers of debris. This behaviour is interpreted in terms of the transfer and loss of debris as wear progresses, involving deformation-adhesion, delamination, abrasion, debris oxidation and attrition of the reinforcing particles.

Influence of surface and heat treatment on the fretting wear of an aluminium alloy reinforced with SiC particles

Abstract Because particulate reinforced metal matrix composites cannot be welded without unacceptable disruption of local particle distribution and matrix microstructure, they are normally joined using mechanical fasteners or adhesives, and consequently they are susceptible to fretting wear and fretting fatigue. The fretting wear behaviour of a 2124 AlCuMg alloy reinforced with a uniform distribution of 17 vol% of 3 μm silicon carbide particles has been compared with that of the equivalent monolithic alloy. With the matrix alloy in the naturally aged (T4) condition, it was shown previously that for test durations up to ∼1 × 106 fretting cycles (40 μm stroke) the composite possessed significant advantages over the monolithic alloy, with a much reduced level of debris loss and transfer to the steel counterface. For longer test durations a reversal in behaviour took place with the wear rate of the composite increasing above that of the monolithic alloy. In an attempt to improve the longer duration fretting performance of the composite, two approaches have been adopted: (1) artificial ageing (14 h at 170°C) of the matrix alloy to peak strength (T6 condition), and (2) application of an anodizing surface treatment to the composite. The artificial ageing treatment has little influence on the wear rate of the composite. In contrast, the effect of the anodising surface treatment is most marked. Anodizing of the composite (T4 condition) was carried out in 1.5M sulphuric acid at a current density of 5 mA cm−2 at 20°C, after a light alkali etch. A period of 45 min was sufficient to produce a 10 μm thick layer into which the reinforcing particles were assimilated. Testing was undertaken with the anodized layer not sealed. Shorter duration tests (≤ 1 × 106 fretting cycles) showed increased levels of material transfer from the anodized composite to the steel counterface. However, over longer duration tests (≤ 1 × 107 fretting cycles) anodizing of the composite achieved an appreciable reduction (factor of ≥ 3) in its wear rate, and in the wear rate of the steel counterface. The behaviour of the anodized layer is interpreted in terms of its early fragmentation introducing oxide debris into the fretting interface which restricts metal-to-metal contact long after the anodized layer has been consumed.

A methodology for modelling the effects of nitriding on fatigue life

A methodology is presented for modelling the effects of nitriding for fatigue life prediction. The method is illustrated by application to tooth bending fatigue failure in notch conditions representative of the fillet regions in spline teeth. The methodology first involves the prediction of nitrided layer stress distributions which take account of the graduation in material stress—strain curve through the nitrided layer and the effects of residual stresses. Second, the life prediction aspect requires that the effects of the graded material properties and residual stresses on mean stress levels and thus on fatigue constants be accounted for. Comparisons between predicted and experimentally-determined lives show that the methodology significantly improves the predictions for nitrided specimens.