Experimental/Numerical Crossover Jet Impingement in an Airfoil Leading Edge Cooling Channel

Abstract

Technological advancement in the gas turbine field demands high temperature gases impacting on the turbine airfoils in order to increase the output power as well as thermal efficiency. The leading edge is one of the most critical and life limiting sections of the airfoil which requires intricate cooling schemes to maintain a robust design. In order to maintain coherence with a typical external aerodynamic blade profile, cooling processes usually take place in geometricallycomplex internal paths where analytical approaches may not provide a proper solution. In this study, experimental and numerical models simulating the leadingedge and its adjacent cavity were created. Cooling flow entered the leadingedge cavity through the crossover ports on the partition wall between the two cavities and impinged on the internal surface of the leading edge. Three flow arrangements were tested: (1) and (2) being flow entering from one side (root or tip) of the adjacent cavity and emerging from either the same side or the opposite side of the leadingedge cavity, and (3) flow entering from one side of the adjacent cavity and emerging from both sides of the leadingedge cavity. These flow arrangements were tested for five crossoverhole settings with a focus on studying the heat transfer rate dependency on the axial flow produced by upstream crossover holes (spent air). Numerical results were obtained from a threedimensional unstructured computational fluid dynamics model with 1.1 أ— 106 hexahedral elements. For turbulence modeling, the realizable kخµ was employed in combination with the enhanced wall treatment approach for the near wall regions. Other available RANS turbulence models with similar computational cost did not produce any results in better agreement with the measured data. Nusselt numbers on the nose area and the pressure/suction sides are reported for jet Reynolds numbers ranging from 8000 to 55,000 and a constant crossover hole to the leadingedge nose distance ratio, Z/Dh, of 2.81. Comparisons with experimental results were made in order to validate the employed turbulence model and the numericallyobtained results. Results show a significant dependency of Nusselt number on the axial flow introduced by upstream jets as it drastically diminishes the impingement effects on the leadingedge channel walls. Flow arrangement has immense effects on the heat transfer results. Discrepancies between the experimental and numerical results averaged between +0.3% and −24.5%; however, correlation between the two can be clearly observed.