Prediction of Gas Turbine Oil Scoop Capture Efficiency

Abstract

The lubrication system in a gas turbine engine is akin to the human blood circulatory system. Providing right quantities of oil to the right components for cooling and lubrication is the primary function of the lubrication system. In the current analysis, at the downstream end of the lube oil supply line, a stationary oil nozzle sprays a jet of oil to a high speed rotating component called an oil scoop. The function of the oil scoop which rotates at speeds usually greater than 10000 RPM is to ‘Scoop’ or capture the oil and provide an under race oil transfer mechanism to the bearings rotating especially at such high speeds. If the oil capture is less than required by the downstream bearing components, it could lead to diminished bearing lives in the gas turbine. The oil scoop consists of two or more blades that are angled with respect to the radius of the Scoop to provide an entry to the oil jet. The ‘window’ of open space between the blades is important to capture the oil. The ratio of quantity of oil captured to the total oil sprayed on to oil scoop is termed as the oil capture efficiency. Several parameters like oil nozzle distance from the blade tip, spray characteristics, jet velocity, number of blades, blade angle, window width, rotational speeds, oil temperature etc. are important factors that determine the capture efficiency of the oil scoop. Prior to the availability of efficient CFD methodologies, it was extremely difficult to develop an oil scoop capture efficiency predictive tool that involves a complex 3D fluid flow from a stationary to a rotating component. The typical Reynolds number of the jet is around 13000 and the oil scoop tip speeds of the order of Mach 0.2 to 0.4. To evaluate various scoop design configurations and enhancements, a transient CFD methodology was developed using multiphase Volume of Fluid approach available in FLUENT® software. In this paper a technique or process is described and demonstrated to simulate the right ‘periodic’ nature of the oil capture and transfer mechanism. It is shown that the CFD methodology described compares well with experimental data. This robust CFD methodology predicts the complex 3D flow with sufficient accuracy and has the potential to be used to optimize the geometry for maximum oil capture efficiency of oil scoops in gas turbine lubrication system. Significant reduction of costly experiments is also an important benefit of developing this predictive methodology.