Mixing of a Plain Jet into a Swirling Crossflow

Many gas turbine combustion systems employ swirl to help mix fuel and air and subsequently stabilize the reaction. In the case of liquid fuels, strategic introduction of the liquid into the swirling flow can provide improved combustion performance. In the present work, radial injection of a plain jet into a swirling flow is considered. A large body of work is available in the literature for liquid injection into a uniform crossflow which neglects the possible influence of swirl on the subsequent fuel plume distribution and performance. The goal of the present work is to (1) establish the role of swirl on the behavior of a liquid jet injected into a crossflow and (2) explore how well correlations based primarily on atmospheric data scale to high temperature and pressure conditions. To accomplish this, pressure, temperature, pressure drop, and fuel flow were varied for an axial swirl premixing module with plain jet fuel injection. Instantaneous planar images of the spray were collected using planar laser induced fluorescence and an intensified CCD camera. The images obtained are processed to generate a number of performance characteristics associated with the fuel plume including area, centerline distance, angle of rotation and plume unmixedness. Two approaches are taken regarding analysis. First, a standard least squares fit maximizing R is carried out for a presumed model form. Due to the presence of mulitcollinearity in the parameters studied, a reduced model with maximum preditcablity was completed based on physical understanding and general dependencies of the characteristics on individual nondimensional parameters. Maximizing the predictablity and dropping one or more regression parameters did not greatly effect the fit. In addition to general behavior of a jet injected into a swirling crossflow, the results also revealed that centerline distance and plume unmixedness have the greatest potential to be scaled from atmospheric to high temperatures and pressures found in gas turbines. * corresponding author Introduction It is well established that atomization of liquid fuels can play a significant role in the performance and emissions associated with combustion devices. As a result, understanding how to optimize this process and the subsequent transport, evaporation, mixing, heat transfer and combustion phenomena that occur can give rise to reduced emissions, improved efficiency and better overall operability of combustion systems. Despite the complications, innovative design strategies have been developed to do so. One such approach is lean premixed prevaporized (LPP) combustion. Operating a combustion device, such as a gas turbine, at a lean premixed prevaporized condition curbs the formation of nitrogen oxides (NOx). LPP utilizes excess amounts of air to avoid locations of fuel rich combustion that raise flame temperatures and promotes the creation of nitric oxide (NO). Achieving LLP combustion depends on the fuel injection system’s ability to create a homogenous mixture of air and fuel before the reaction begins. Mixing fuel and air quickly and efficiently is difficult to accomplish in practice. Mixing performance is crucial because mixing times are limited to 1 ms due to the danger of autoignition. An elegant problem that reveals many of the key features associated with the fuel injection process is that of the plain liquid jet injected from a wall into a uniform crossflow. As a result of the simplicity of the configuration, this problem has received a tremendous amount of attention in the literature , , , 4 5 6 7 from both experimental and modeling perspectives. As a result, key dependencies on dimensionless groups such as momentum flux ratio, Weber number and Reynolds number have been identified and can be thus used to develop simple expressions to describe the resulting fuel distribution and penetration. Momentum flux ratio is quantitatively defined as the ratio of the jet to crossflow momentum (Equation 1). The interaction between density (ρ) and velocity (U) dictates how far the jet will penetrate into the flow; large penetration is signified by a large momentum flux ratio. This dimensionless quantity is extensively used in jet penetration correlation. The subscript j and c reference the jet and crossflow respectively.