A new phenomenological model to predict drop size distribution in Large-Eddy Simulations of airblast atomizers

Abstract

A new atomization model for prefilming airblast atomizers is presented and applied in the Large-Eddy Simulation of an academic experiment. The model, named PAMELA, expresses the drop size Probability Density Function of the spray in the form of a Rosin–Rammler distribution whose parameters depend on flow conditions. A mechanism of liquid fragmentation is proposed where a Rayleigh–Taylor instability develops in the transverse direction. The wavelength of this instability (i) is assumed to be proportional to the Sauter Mean Diameter of the spray, and (ii) scales with a Weber number based on the atomizing edge thickness, providing a first link between flow conditions and the Rosin–Rammler parameters. The second link is found by introducing a second Weber number based on the thickness of the boundary layer developing on the prefilmer. A first comparison with academic experiments shows that the model assumptions are valid and allows to calibrate the model constants. PAMELA is then implemented in a LES solver to perform the numerical simulation of an academic airblast atomizer. The obtained drop size distribution and spatial structure of the spray are in good agreement with measurements, demonstrating the validity of the proposed approach in the context of LES, and that the proposed PAMELA model may now be used to describe the liquid spray in LES of industrial nozzles.