ILASS Americas, 17th Annual Conference on Liquid Atomization and Spray Systems, Arlington, VA, May 2004. — 11 p.Abstract
Two classical secondary atomization models commonly used in multidimensional computational fluid dynamics (CFD) codes were evaluated against single droplet experimental measurements. The Taylor Analogy Breakup (TAB) and Kelvin-Helmholtz (KH) instability models were compared to measurements of breakup time ranging from the bag to the catastrophic breakup regimes (Pilch & Erdman 1987, Dai & Faeth 2001), drop size in the bag (Chou & Faeth 1998) and shear (Hsiang & Faeth 1993, Faeth, et al. 1995, Chou, et al. 1997) breakup regimes, and drag coefficient evolution in the bag breakup regime (Chou & Faeth 1998). Droplets having a liquid-to-gas density ratio greater than 751 and Oh 0.1 were studied. The applicability of these submodels over a range of We was in-vestigated and the most sensitive constants within these models identified; appropriate ranges were also determined for physical accuracy. Furthermore, an evaluation of two different numerical approaches, namely abrupt (TAB) ver-sus continuous stripping (KH), was undertaken over the range of breakup regimes considered. The comparison was conducted within a zero-dimensional (0D) single droplet framework capable of replicating unsteady momentum boundary conditions. Included in the 0D code was a simplified drag model, which updates the relative velocity and drag coefficient of the drop at each timestep, assuming a constant ambient flow field. The results revealed that suit-able bounds on key model constants could be identified to estimate breakup time, drop size, or drag coefficient, in-dividually, for a specific regime; however, the simultaneous prediction of all three led to inherent tradeoffs between different regimes and between drop size and breakup time predictions.
