Many fuel injector geometries proposed for lean-premixed combustion systems involve the use of radial swirlers. At the high swirl numbers needed for flame stabilization, several complex unsteady fluid mechanical phenomena such as vortex breakdown and recirculation zone precession are possible. If these unsteady aerodynamic features are strongly periodic, unwanted combustion induced oscillation may result. The present paper reports on an isothermal experimental study of a radial swirler fed fuel injector originally designed by Turbomeca, and examines the dynamical behavior of the unsteady aerodynamic flow structures observed. Particle Image Velocimetry (PIV) is used to capture the instantaneous appearance of vortex structures both internal to the fuel injector, and externally in the main flame-stabilizing recirculation zone. Multiple vortex structures are observed. Vector field analysis is used to identify specific flow structures and perform both standard and conditional time averaging to reveal the modal characteristics of the structures. This allows analysis of the origin of high turbulence regions in the flow and links between internal fuel injector vortex breakdown and external unsteady flow behavior. The data provide a challenging test case for Large Eddy Simulation methods being developed for combustion system simulation.

The accurate discrimination of the aerodynamic parameters affecting the flight of sports balls is essential in the product development process. Aerodynamic studies reported to date have been limited, primarily because of the inherent difficulty of making accurate measurements on a moving or spinning ball. Manufacturers therefore generally rely on field trials to determine ball performance, but the approach is time-consuming and subject to considerable variability. The current paper describes the development of a method for mounting stationary and spinning footballs in a wind tunnel to enable accurate force data to be obtained. The technique is applied to a number of footballs with differing constructions and the results reported. Significant differences in performance are noted for both stationary and spinning balls and the importance of the ball orientation to the flow is highlighted. To put the aerodynamic data into context the results are applied in a flight model to predict the potential differences in the behaviour of each ball in the air. The aerodynamic differences are shown to have a considerable effect on the flight path and the effect of orientation is shown to be particularly significant when a ball is rotating slowly. Though the techniques reported here are applied to a football they are equally applicable to other ball types.