Abstract

Aero-gas turbine engines with a mixed exhaust configuration offer significant benefits to the cycle efficiency relative to separate exhaust systems, such as increase in gross thrust and a reduction in fan pressure ratio required. A number of military and civil engines have a single mixed exhaust system designed to mix out the bypass and core streams. To reduce mixing losses, the two streams are designed to have similar total pressures. In design point whole engine performance solvers, a mixed exhaust is modelled using simple assumptions; momentum balance and a percentage total pressure loss. However at far off-design conditions such as windmilling and altitude relights, the bypass and core streams have very dissimilar total pressures and momentum, with the flow preferring to pass through the bypass duct, increasing drastically the bypass ratio. Mixing of highly dissimilar coaxial streams leads to complex turbulent flow fields for which the simple assumptions and models used in current performance solvers cease to be valid. The effect on simulation results is significant since the nozzle pressure affects critical aspects such as the fan operating point, and therefore the windmilling shaft speeds and air mass flow rates. This paper presents a numerical study on the performance of a lobed mixer under windmilling conditions. An analysis of the flow field is carried out at various total mixer pressure ratios, identifying the onset and nature of recirculation, the flow field characteristics, and the total pressure loss along the mixer as a function of the operating conditions. The data generated from the numerical simulations is used together with a probabilistic approach to generate a response surface in terms of the mass averaged percentage total pressure loss across the mixer, as a function of the engine operating point. This study offers an improved understanding on the complex flows that arise from mixing of highly dissimilar coaxial flows within an aero-gas turbine mixer environment. The total pressure response surface generated using this approach can be used as look-up data for the engine performance solver to include the effects of such turbulent mixing losses.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call