Study and modelling of surface topography changes during running-in in rolling-sliding contacts under mixed lubrication

Abstract

The initial interaction between surfaces in mechanical systems that leads to wear is known as running-in. During this period, asperities undergo elastic and plastic deformation, generating wear debris and reducing surface roughness. This process continues until a steady state is achieved, facilitated by the transition of the lubrication regime from boundary/mixed to elasto-hydrodynamic / hydrodynamic lubrication regime, wherein the surfaces are fully separated by a lubricant film. This thesis aims to provide a more comprehensive understanding of the running-in phenomenon in bearing contacts, where it significantly influences the overall performance and service life of bearings. The project simulated a pure-rolling and rolling-sliding contact of a roller-element bearing using a mini traction machine (MTM) and a twin disc tribometer. The mini traction machine tests were employed to closely monitor the surface topography changes occurring during running-in, both under pure-rolling conditions and with induced slip, on AISI 52100 Bearing Steel test samples, while maintaining constant maximum pressure of 1 GPa, entrainment velocity of 3.7 m/s, lubricant (PAO Base Oil) temperature of 100 °C and lambda ratio (λ) of 1.15. The results reveal that test surfaces subjected to slip experienced a decrease in asperity valley depth due to plastic deformation and material flow from adjacent peaks, in addition to a reduction in asperity peak height, thus providing new evidence of the accurate mechanism behind the reduction in surface roughness during the running-in period. 5 The twin disc tribometer, due to its higher load capacity, was utilized to further enhance the understanding of the running-in phenomenon. A set of parametric tests were conducted to allow for a comprehensive understanding of the true impact of slip, entrainment velocity, and pressure on surface topography changes during running-in. The key findings from these tests reveal that running-in can be classified into three categories: Category 1, no wear / elastic contact; Category 2, mild wear; and Category 3, wear, severe plastic deformation, and plastic material flow. Furthermore, it was noted that average roughness, Ra, and root mean square roughness, Rq, decrease in a linear manner during this process, while skewness, Rsk, behaves non-linearly, decreasing with mild wear due to asperity peak removal, and subsequently remaining unchanged or slightly increasing at the onset of plastic deformation and material flow. The surface roughness changes observed during running-in were modelled using various machine learning algorithms. However, linear regression performed the best, successfully capturing the linear behaviour of Ra and Rq with 6% error. The coefficients obtained from the linear regression model indicate that changes in Ra and Rq during running-in are influenced by an increase in slide-roll ratio and a decrease in minimum film thickness.

More information

Identifiers

ORCID for Maruti Sai Dhiraj Sakhamuri: orcid.org/0000-0001-7607-3087

ORCID for Robert Wood: orcid.org/0000-0003-0681-9239

Catalogue record

Export record