A Loss Prediction Model for Manufacturing Variations on High-Pressure Turbine Blade Shroud Platforms

Abstract Manufacturing variability affects high-pressure turbine blade shroud platforms, causing platform steps and interplatform gaps. This article presents a physics-based shroud step sentencing model. The model predicts the increase in stagnation pressure loss coefficient from the local conditions at the wedge face and measurements of the step height from the production line. These local conditions can be obtained from a precomputed clean endwall simulation. It is shown that a quasi-2D (Q2D) approximation is sufficient to assess the loss caused by manufacturing variations to within ±2% of the clean endwall passage loss. The approximation relies on the fact that the step loss is mainly caused by the step-normal velocity. The model provides a local stagnation pressure loss coefficient, obtained from a 2D control volume, which requires a base pressure and a skin friction coefficient. These coefficients are found by correlation against a Q2D CFD database. The overall passage step loss can then be calculated by integrating the local loss coefficient along the platform edge. The sentencing model is validated against steps in a linear turbine cascade for which the steps are a significant source of loss and for which experimental data are available. The performance of the model is then demonstrated on engine-like geometry.