CFD Simulation of a Two-Stage High-Pressure Turbine with Varying Tip Clearances and Evaluation of Turbine Characteristics

Kurzfassung

ERRATA: The pressure loss coefficient comparison with the loss correlations (Fig. 5.13) is erroneous. Please refer to the internal report DLR-IB-AT-GO-2026-46 for a corrected version.

DLR aims to develop a comprehensive predictive maintenance strategy to lower maintenance costs and increase the availability and safety of aero-engines. This requires performance data from engine components that exhibit commonly observed degradation phenomena. In this work, the effect of tip clearance degradation on the performance of a cooled two-stage high-pressure turbine is investigated computationally. Three-dimensional steady-state Reynolds-Averaged Navier-Stokes equations are solved using the DLR Institute of Propulsion Technologies’ in-house flow solver TRACE. Tip clearance degradation is modeled by increasing the clearance from 0.5 % to 3.0 % of the blade height (τ/h), in increments of ∆τ/h = 0.5 %. A comprehensive meshing concept is applied to the turbine forming structured multiblock grids with highly resolved tip clearance gaps. Cooling is modeled using a panel-source approach. Additionally, a grid convergence study and sensitivity studies on turbine inlet turbulence and boundary layer transition are conducted to ensure numerical validity of the results.

The results are in agreement with tip clearances effects on turbine performance found in the literature. An average drop in efficiency of 1.068 %pt per ∆τ/h = 0.5 % of tip clearance increase is observed when tip clearances in both rotors are increased. A quadratic curve is fitted to the data to model the efficiency penalty associated with tip clearance growth. The results of this work are compared to correlation based models commonly used for predicting the tip clearance related efficiency penalty, tip clearance pressure loss coefficient, and the change in deviation. The results are in good agreement with the Kacker-Okapuu correlation for the efficiency penalty, and with Buske’s correlations for pressure loss coefficient and change in deviation.