Abstract

Abstract Ceramic Matrix Composite (CMC) with environmental barrier coating can tolerate significantly higher temperatures compared with the conventional metal alloy blades and thus can help achieve higher thermal efficiency in gas turbine engines. Due to the complex fabrication process of CMC blades, larger blade trailing edge thicknesses and larger leading-edge radii of curvature are expected. These features significantly alter aerodynamic performance especially near the trailing edge and thus the loss profile. In previous studies, the authors used large eddy simulation (LES) as well as the Reynolds Averaged Navier-Stokes (RANS) coupled with an intermittency function-based transition model to simulate aerodynamic performance of three different CMC blades and compared the predictions against the data acquired at NASA Glenn Transonic Turbine Blade Cascade Rig. Although these were able to successfully predict different characteristics of aerodynamic performance among three different blades (e.g., pressure loading and integrated losses), LES consistently predicted a small recirculation bubble at the end of the suction side. This was not inconsistent with the physics of the flow, but the pressure loading, locally, near the trailing edge, and as a result, the measured wake profiles did not match the experimental data. To delve further into these issues, a new LES grid with a much finer mesh near the trailing edge was generated. The spanwise extent of the grid was increased by 50% at the same time to accommodate large vortices which might not have been captured by the original domain due to size limitations. An LES simulation for the CMC blade with the largest trailing edge thicknesses at the high-freestream turbulence (Tu = 13%) condition showed that using such a mesh is necessary for LES to accurately capture the transition phenomena on the suction side. The resulting total pressure coefficient profile and the shift of the wake profile were greatly improved. The increase in the span was found to be inconsequential.

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