System for Simulating Mould OscillationIn the present study, the system for the belt rotation was totally changed in order to investigate the effect of the oscillation on the oil infiltration. In the previous experiment that neglects the oscillating motion, a normal AC motor was used just for the belt movement at constant velocities. 26) In contrast, in this study, an AC servomotor was installed with a controller that is connected to a function generator (see Fig. 2). Using this system, the belt velocity was set as a function that consists of both time independent and time dependent velocities. The time independent velocity, V C , corresponds to casting velocity, and the time dependent velocity, V M , represents oscillating velocity of mould, which is written as:.................... (1) Where, t is the time, S and f indicate the stroke and frequency of oscillation, respectively.During the experiment, both the rotating velocity of the servomotor and displacement of the acrylic plate measured by the linear gauge sensor are transmitted simultaneously to a personal computer. This makes it possible to observe dynamic change in the film thickness of the oil in relation to the cyclic variation in the belt velocity. In a continuous casting mould, the in-situ observation of the film thickness of mould flux cannot be realized during operation. The present cold model, which overcomes this problem, will provide a better understanding of the infiltration behavior of mould flux.Precisely, there remains a difference between the mould flux infiltration in continuous casting and the oil infiltration in this experiment as indicated in Fig. 1. In continuous casting, the strand is moving downward at casting velocity, V C , and the mould is reciprocating at V M . On the other hand in the present experiment, for simplicity of the mechanism in the apparatus, the acrylic plate is fixed in the vertical direction and only the belt is moving at V C ϩV M . This discrepancy will be discussed on the basis of our theoretical model. 23,24,26) In our previous report, 26) the infiltration of silicone oil in this experiment was formulated by the Navier-Stokes equation for the simple case neglecting the oscillation. For the purpose of including the oscillation for the present study, only its boundary condition at the belt surface must be changed. Thus, replacing the boundary condition uϭV C with uϭV C ϩV M at zϭ0 in Eq. (3) in Ref. 26) and solving it, one obtains a force with which the acrylic plate is pushed away from the belt by pressure generated in the oil channel, F (N/m)......... (2) Where L represents length of the oil channel, m the dynamic viscosity of oil, r the density of oil, g the gravity constant, P 0 , P L the pressure at the entrance and exit of the channel, respectively. P 0 is set to pgd in which d is depth of oil pool, P L is zero. Where x is the distance from the entrance of the channel along the belt surface, and h(x) is the film thickness of oil as a function of x. Each of the three terms of Eq. (2) represents the contributio...