Investigation of droplet collision characteristics with moving film and its comparison with stationary film: unsteady and 3D CLSVOF method

This research deals with a numerical study of a drop impacting a moving film as a transient three-dimensional and its comparison with static film. For the modeling, the CLSVOF (Coupled Level Set and Volume of Fluid) method has been used for drop impact analysis. The parametric effects of film Reynolds number (830 < Re_f < 4478), non-dimensional film thickness (0.25 < h^* < 0.75), drop Weber number (249 < We < 1762), non-dimensional time (0.5 < τ < 4.0), and impact angle (0°, 30°, and 60°), on the interface evolution, was investigated. Further, the results have been compared to drop-impacting static film obtained by previous researchers, whereby in most cases consists of the results with drop-impacting static film have been observed. However, the drop impacting the moving film indicated a different behavior in some cases because of the unique effect of the moving film movement. Doubling the film velocity as well as a 57% reduction in the drop velocity, caused an 8.3% and 6.66% increase in the crown asymmetry and suppression of the downstream crown. The crown slip toward the downstream flow was observed in all cases. Doubling the fluid film velocity, on average, reduced the crown height by 18.5% and increased its diameter by 10.48%. However, its effect on the crater diameter was negligible (1%). An increase in the film thickness from 0.5 mm to 1 mm (h^* from 0.25 to 0.5), fed the upstream crown, and on average, increased its height by 10.57%. At a low impact velocity of 3 m/s (We = 249), the crown diameter was, on average, 26.7% larger than its diameter at the velocity of 7 m/s (We = 1358). By increasing the drop impact angle to 60 degrees and overcoming the effect of the fluid film velocity, the crown behavior was the same as its behavior with the drop impact on the static film. Finally, to predict the upstream crown height, downstream crown height, crown diameter, and crater diameter, four novel five-variable correlations (depending on τ, θ, Re_f, We, and h^*) are developed. It is concluded that a good agreement exists between the numerical data, and correlation results.