A B S T R A C TThe present paper studies the occurrence of micropitting damage in gear teeth contacts. An existing general micropitting model, which accounts for mixed lubrication conditions, stress history, and fatigue damage accumulation, is adapted here to deal with transient contact conditions that exist during meshing of gear teeth. The model considers the concurrent effects of surface fatigue and mild wear on the evolution of tooth surface roughness and therefore captures the complexities of damage accumulation on tooth flanks in a more realistic manner than hitherto possible. Applicability of the model to gear contact conditions is first confirmed by comparing its predictions to relevant experiments carried out on a triple-disc contact fatigue rig. Application of the model to a pair of meshing spur gears shows that under low specific oil film thickness conditions, the continuous competition between surface fatigue and mild wear determines the overall level as well as the distribution of micropitting damage along the tooth flanks. The outcome of this competition in terms of the final damage level is dependent on contact sliding speed, pressure and specific film thickness. In general, with no surface wear, micropitting damage increases with decreasing film thickness as may be expected, but when some wear is present micropitting damage may reduce as film thickness is lowered to the point where wear takes over and removes the asperity peaks and hence reduces asperity interactions. Similarly, when wear is negligible, increased sliding can increase the level of micropitting by increasing the number of asperity stress cycles, but when wear is present, an increase in sliding may lead to a reduction in micropitting due to faster removal of asperity peaks. The results suggest that an ideal situation in terms of surface damage prevention is that in which some mild wear at the start of gear pair operation adequately wears-in the tooth surfaces, thus reducing subsequent micropitting, followed by zero or negligible wear for the rest of the gear pair life. The complexities of the interaction between the contact conditions, wear and surface fatigue, as evident in the present results, mean that a full treatment of gear micropitting requires a numerical model along the lines of that applied here, and that use of overly simplified criteria may lead to misleading predictions.

Surface initiated rolling contact fatigue, leading to a surface failure known as pitting, is a life limiting failure mode in many modern machine elements, particularly rolling element bearings. Most research on rolling contact fatigue considers total life to pitting. Instead, this work studies the growth of rolling contact fatigue cracks before they develop into surface pits in an attempt to better understand crack propagation mechanisms. A triple-contact disc machine was used to perform pitting experiments on bearing steel samples under closely controlled contact conditions in mixed lubrication regime. Crack growth across the specimen surface is monitored and crack propagation rates extracted. The morphology of the generated cracks is observed by preparing sections of cracked specimens at the end of the test. It was found that crack initiation occurred very early in total life, which was attributed to high asperity stresses due to mixed lubrication regime. Total life to pitting was dominated by crack propagation. Results provide direct evidence of two distinct stages of crack growth in rolling contact fatigue: stage 1, within which cracks grow at a slow and relatively steady rate, consumed most of the total life; and stage 2, reached at a critical crack length, within which the propagation rate rapidly increases. Contact pressure and crack size were shown to be the main parameters controlling the propagation rate. Results show that crack propagation under rolling contact fatigue follows similar trends to those known to occur in classical fatigue. A log-log plot of measured crack growth rates against the product of maximum contact pressure and the square root of crack length, a parameter describing the applied stress intensity, produces a straight line for stage 2propagation. This provides the first evidence that growth of hereby-identified stage 2 rolling contact fatigue cracks can be described by a Paris-type power law, where the rate of crack growth across the surface is proportional to the contact pressure raised to a power of approximately 7.5.

Micropitting is a type of surface damage that occurs in rolling-sliding contacts operating under thin oil film, mixed lubrication conditions, such as those formed between meshing gear teeth. Like the more widely studied pitting damage, micropitting is caused by the general mechanism of rolling contact fatigue but, in contrast to pitting, it manifests itself through the formation of micropits on the local, roughness asperity level. Despite the fact that micropitting is increasingly becoming a major mode of gear failure, the relevant mechanisms are poorly understood and there are currently no established design criteria to assess the risk of micropitting occurrence in gears or other applications. This paper provides new understanding of the tribological mechanisms that drive the occurrence of micropitting damage and serves to inform the ongoing discussions on suitable design criteria in relation to the influence of contact slide-roll ratio (SRR) on micropitting. A triple-disc rolling contact fatigue rig is used to experimentally study the influence of the magnitude and direction of SRR on the progression of micropitting damage in samples made of case-carburised gear steel. The test conditions are closely controlled to isolate the influence of the variable of interest. In particular, any variation in bulk heating at different SRRs is eliminated so that tests are conducted at the same film thickness for all SRRs. The results show that increasing the magnitude of SRR increases the level of micropitting damage and that negative SRRs (i.e. the component where damage is being accumulated is slower) produce more micropitting than the equivalent positive SRRs. Measurements of elastohydrodynamic film thickness show that in the absence of bulk heating, increasing SRR does not cause a reduction in EHL film thickness and therefore this cannot be the reason for the increased micropitting at higher SRRs. Instead, we show that the main mechanism by which increase in SRR promotes micropitting is by increasing the number of micro-contact stress cycles experienced by roughness asperities during their passage through the rolling-sliding contact. Therefore, the asperity stress history should form the basis of any potential design criterion against micropitting.

Condition monitoring of machine health via analysis of vibration, acoustic and other signals offers an important tool for reducing the machine downtime and maintenance costs. The key aspect in this process is the ability to relate features derived from the recorded sensor signals with the physical condition of the monitored asset in real time. This paper uses simple machine learning techniques to examine the ability of specific time-domain features obtained from vibration signals to predict the progression of surface distress in lubricated, rolling-sliding contacts, such as those found in rolling bearings and gears. Controlled experiments were performed on a triple-disc rolling contact fatigue rig using seeded-fault roller specimens where micropitting damage was generated and its progression directly observed over millions of contact cycles. Vibration signals were recorded throughout the experiments. Features known as condition indicators were then extracted from the recorded time-domain signals and their evolution related to the observed physical state of the associated specimens using simple machine learning techniques. Five time-domain condition indicators were examined, peak-to-peak, root-mean-square, kurtosis, crest factor and skewness, three of which were found not to be redundant. First, a classification model using KNN nearest neighbor was built with the three informative condition indicators as training data. The cross-validation results indicated that this classifier was able to predict the presence of micropitting damage with a relatively high precision and a low rate of false positives. Secondly, a k-means clustering analysis was performed to measure the significance of each condition indicator by leveraging patterns. The peak-to-peak condition indicator was found to be a good predictor for progression of micropitting damage. In addition, this indicator was able to distinguish between micropitting and pitting failure modes with a high success rate. Finally, the condition indicator response was correlated with the predicted damage state of the test specimen obtained through an existing physics-based surface distress model in order to illustrate the potential of hybrid models for improved prognostics of damage progression in rolling-sliding tribological contacts.

No abstract