Oil-in-water (O/W) emulsion has been widely used as a coolant and a flame-retardant lubricant in the rolling mill. Understanding of the physical properties of the oil film between the roll and strip is necessary because it affects the rolling productivity and surface quality of the sheets. The oil film generally consists of ‘plate-out oil’ and ‘supplied oil’ directly introduced from an O/W emulsion nozzle as oil droplets. However, the detailed mechanism of the oil film formation is still unclear. In this study, we investigated the thickness of plate-out oil on a roller in rolling process and the thickness of oil film formed between ball and disk using a disk-on-ball apparatus with an optical interferometer with changes to the rolling speed, particle size and concentration of the O/W emulsion. As a results, when a volume of plate-out oil was sufficient, the oil film thickness decreased by supplying O/W emulsion because of the re-emulsifying. When the plate-out volume was insufficient, the oil film thickness became thicker just after a high-concentrated O/W emulsion with a large particle size was supplied. In addition, the oil film thickness became thinner when the sliding speed became higher as a result of a growth of vortex which occurred around the inlet zone.

The adhesive wear is classified into two parts, initial and steady-state. The transition between initial wear and steady-state wear is influenced by a lot of factors. The aim of this paper is to pile up the basic data for initial wear simulation. The wear volume of sliding similar metals in air was determined immediately after the transition from initial wear to steady-state wear occurred. The effects of the transition sliding length Lt, mean contact pressure p, hardness H of specimens and relative humidity U on the initial wear volume Vt were investigated. Friction/wear tests under repeated sliding were carried out with a rotating disk on a plate configuration for austenitic stainless steels. Reasonably good agreement was observed between the straight line following Vt/a = Lt/b = p/c and the experimental data for Vt, Lt, and p in a 3-dimensional Vt-Lt-p diagram, where a, b, and c are constants. As a disk and/or a plate changed the H value, plotting Vt against Lt yielded a straight line, expressed as Vt/a = Lt/b. The experimental data for Vt, Lt, and U in a 3-dimensional Vt-Lt-U diagram are approximately on the same plane, described by EVt + FLt = 0, where E and F are constants. The line of intersection of the EVt + FLt = 0 and the Vt-Lt planes, corresponds to Vt/a = Lt/b. The Vt-U curve takes a maximum value at a U of 65%.