Paper: Strain-rate effects on the plastic indentation and abrasion of elastohydrodynamic contacts by debris particles.
Author: George K. Nikas
Where: Proceedings of the Institution of Mechanical Engineers (IMechE), Part J: Journal of Engineering Tribology, 2014, 228(1), 22-45. Available from the publisher.
A numerical model of plastic indentation and abrasion of elastohydrodynamic contacts by debris particles previously developed by the author is extended to include the dependence of material flow stress on strain rate. Using the Johnson-Cook viscoplasticity model, the flow stress of all materials involved in the indentation process is expressed as a function of plastic strain, strain rate, and temperature. This complements other elements of the model, including strain-gradient plasticity, work hardening, frictional heating from particle extrusion, thermal softening, melting, and material loss due to adhesion. Following a laborious programme of experimental validation and numerical comparisons, the predictions of the model are shown to be in excellent agreement with the experimental results on soft and hard particles in rolling and rolling-sliding elastohydrodynamic contacts. The incorporation of strain-rate effects further improved the agreement between theoretical and experimental results previously established with simpler versions of the model that ignored the strain-rate factor. Strain rate is also shown to affecting several parameters in the process of surface damage, including the magnitude of contact stresses and flash temperatures, as well as the behaviour of a particle in a concentrated contact. It is also shown that for an optimum contact velocity linked to strain-rate effects and fluid film thickness in lubricated contacts, surface damage is minimised, particularly for large and hard particles.
Some figures from this work
Figure 1 shows an example of the dents created by a 36 micron, spherical particle of M50 steel in a rolling elastohydrodynamic contact. The particle and the contact counterfaces have approximately the same cold hardness of 750 Vickers. The top row of figures concerns the results when strain-rate effects are not accounted for. In the bottom row of figures, strain-rate has been accounted for. The visual difference is clear to see and the same is true for dent dimensions, as depicted in figure 1.
Fig. 1. Dent by a 36 micron, M50 steel particle (750 Vickers) in a rolling contact with counterfaces of the same hardness (748 HV), excluding and including strain-rate effects. Views are in the rolling direction (left) and transverse direction (right).
2 shows indentation results as function of the rolling velocity for a
30 micron, spherical M50
steel particle (750 Vickers) in a rolling elastohydrodynamic
contact with counterfaces
of equal hardness (748 HV). Dent depth, volume and average slope in
the rolling direction are all depicted. The maximum flash temperature
from the frictional heating during particle extrusion in the contact
is also shown. The left graph in figure 2 shows the results when
strain rate is unaccounted for. In the right graph, strain rate has
been accounted for. In lubrication literature, the particle analysed
in this case is considered relatively large and very hard. Regardless
of strain rate effects, it is obvious in both diagrams of figure 2
that at a critical velocity of 2 m/s, dent depth, volume and slope are
globally minimised. At the same velocity, maximum flash temperature
also shows a distinctive local minimum. Strain rate notably affects
the gradient of the curves below the critical velocity in conjunction
with film thinning, as well as above the critical velocity in
conjunction with film thickening and thermal softening. It is also
noticeable that the flash temperatures are significantly lower when
strain rate effects are included by up to about 410 °C for higher velocities. This is critical
in terms of surface damage or metallurgical changes to the contact
surfaces such as the formation of white layers and embrittlement,
including the development of surface thermo-cracks when the
frictionally heated surface is abruptly exposed to fresh lubricant
upon removal of the squashed particle.
Fig. 2. Strain-rate effects on indentation data as function of the rolling velocity for a 30 micron, M50 steel particle (750 HV) in a rolling contact with counterfaces of equal hardness (748 HV).
Several examples concerning various particles are analysed in the paper, including copper particles (90 Vickers) and high-carbon steel particles (750 Vickers) in rolling and rolling-sliding contacts. Input data for all cases are adopted from the experimental literature and the numerical results are thoroughly compared with those derived experimentally. A large part of the paper is devoted to an exhaustive validation with experimental data and the model is shown to being in excellent agreement with the most well documented data available in the literature.
Homepage of Dr Nikas