| description abstract | Transpiration cooling represents the pinnacle of turbine cooling and is characterized by an intrinsic material porosity which achieves high internal convective cooling, and full coverage cooling films on the external surface subjected to the hot gases. Quasi-transpiration systems, such as the double-wall effusion system discussed here, attempt to replicate the cooling effect of transpiration systems. The double-wall system is characterized by a large internal wetted area providing high internal convective cooling performance, with a highly porous external wall allowing the formation of a protective film over the external surface. This paper presents a low-order thermal model of a double-wall system designed to rapidly ascertain cooling performance based solely on the geometry, solid thermal conductivity, and approximate surface heat transfer coefficients. The performance of the model is initially validated using experimental data with heat transfer coefficients for the low-order model obtained from fully conjugate computational fluid dynamics (CFD) simulations. Following this, a more controlled CFD study is undertaken with both fully conjugate and fluid-only simulations performed on several double-wall geometries to ascertain both overall effectiveness and film effectiveness data. Data from these simulations are used as inputs to the low-order thermal model developed and the results compared. The low-order model successfully captures both the trends and absolute cooling effectiveness achieved by the various double-wall geometries. The model therefore provides an extremely powerful tool in which the cooling performance of double-wall geometries can be near instantaneously predicted during the initial design stage, potentially allowing geometry optimization to rapidly occur prior to more in-depth, costly, and time-consuming analyses of the systems being performed. This potential benefit is demonstrated via the implementation of the model with input boundary conditions obtained using empirical correlations. | |
