Observations of dry wear and the development of an improved wear law

Unlubricated wear of metal components is a significant factor in the deterioration of various metal structures and assemblies and has been the focus of intensive research for several decades. Despite this, accurately predicting the quantity and development of wear from information about the geometry, material and loading of the wearing contact appears to be beyond the reach of engineering science at present. The development of more generally applicable, reliable wear prediction method (“wear law”) would be of great benefit in terms of industrial practice and would also close a significant gap in the current state of tribological knowledge. The research presented in this thesis establishes both the limitations of traditional wear prediction approaches based on two-body contact mechanics and/or Archard’s wear law as well as a possible alternative in the tribological circuit approach, which was initially proposed by Berthier and then refined into a wear model for a closed contact geometry by Fillot, Iordanoff & Berthier. Through experimental research, a number of assumptions and questions put forward by these authors were examined; in particular the interaction between the debris ejection path length, surface profile and debris particle size was found to govern the ejection of wear debris and thus wear. The model was also found to predict a strongly non-linear dependence on normal force for an annular contact geometry, the same geometry in which exactly such dependence has been observed experimentally, but not yet explained. Additional experimental work revealed a tendency for the contact to eject debris along the shortest available path. With this knowledge, a mathematical expression for debris ejection in multiple directions was formulated. With it, the approach proposed by Fillot, Iordanoff & Berthier was expanded to an open contact geometry, successfully achieving a transition of a wear model between geometries. This would also indicate, that the approach studied could be a generally applicable wear model. To aid the experimental work, a method for observing wear without disturbing it has been developed, using X-ray computed tomography. This is the first such application of X-ray CT to wearing contacts, the first instance of wear being extensively quantified entirely non-destructively (i.e. without dismantling the wearing contact for studies) and has been proven to be very successful as a tool for observing and quantifying wear. The success obtained can also serve as a powerful justification for further development of this approach and progress towards the observations of wear in-situ using penetrating radiation.