Assessment of a Set of Numerical Tools for the Design of Aero-Engines Combustors: Study of a Tubular Test Rig

Abstract

This paper reports the numerical study performed during the setup of a tubular laboratory combustor developed by Avio Group. The main purpose of this study is the assessment of some numerical tools which might be employed for the design of aero-engines combustors. Adopted methodologies pertains with heat transfer and exhaust emissions issues of combustor design, and they refer to both detailed three dimensional and simplified zero/one dimensional formulations. An exhaustive discussion about the obtained result and about the adopted modelling criteria is reported in the paper. The rig has a tubular geometry that allows to study the single burner of a typical annular combustor of jet engines. Considered operating condition for the tests refers to Max Take-Off (ICAO 100%) in the cycle of a typical jet engine. Three different numerical tools were considered, representing the fundamental steps during the preliminary and detailed design of combustors. The first one is a 1D procedure capable to analyze the cooling flow network of the combustor and to predict liner wall temperature and heat loads. The second is a classical chemical reactor network code for flame temperature and exhaust emissions prediction. Such preliminary design tools are usually supported by reactive CFD computations. In this work both two dimensional and three dimensional CFD models are considered with different goals. A 2D CFD model of flame tube with prescribed mass flow rates allows to quickly test various turbulence, combustion and emissions models, while a 3D model of the complete hardware (liner, burner and cooling network), permits to predict air flow splits and wall radiative and convective heat loads. Furthermore a final 3D reactive CFD computation with radiation conjugated with the thermal conduction solution across the liner, permits to estimate the wall temperature distribution. The whole set of numerical results has pointed out an appreciable agreement among the various tools representing a valid assessment of the validity and robustness of selected design methodologies. A more significant validation of codes accuracy will be possible as soon as the scheduled experimental results will be available.