Internal flow dynamics of spill-return pressure-swirl atomizers

Abstract

The sprays produced by spill-return pressure-swirl atomizers are strongly dependent on the nature of the internal fluid dynamics. Several spill-return atomizers were compared in terms of the spatial and temporal behaviour of the internal air-core, liquid sheet thickness and its perturbations. The only difference amongst the test configurations was the geometrical arrangement of the spill-line (SL) orifice through which the liquid was spilled away. The flow field inside the swirl chamber was examined using high-speed imaging with image post processing using an in-house Matlab code and three orthogonal velocity components acquired using Laser Doppler Anemometry. The dimensions of the production atomizers did not allow direct visualization of their internal flow, so a scaled, modular, transparent plexiglass model was used. Its flow characteristics were equivalent to the original atomizer. The refractive index of the atomizer body was matched to the test liquid using a solution of 1-Bromonaphthalene and kerosene fuel type JET A-1. The test conditions were derived from the original atomizer and were limited to inlet port Reynolds numbers, from 700 to 2000 and spill-to-feed ratios, SFR, from 0 to 0.75. An inviscid analysis, originally derived for Simplex atomizers, was modified and applied to the spill-return version. This approach allows a theoretical prediction of the discharge coefficient and air-core diameter dependent solely on SFR. An axially located SL orifice inhibits any internal air-core forming in the swirl chamber. Off-axial SL orifices generate and stabilize the air-core, which leads to the regular formation of a liquid sheet and a high-quality spray. Nevertheless, some configurations changed the breakup nature of the liquid sheet and consequently the spray quality. Moreover, the turn-down ratio of the liquid supply rate and spray stability depend on the distance of the SL orifices from the swirl chamber centreline. The flow energy losses increase with SFR. The outcomes from this analysis allow the optimization of the SL configuration for specific application and extend the classical inviscid analysis.