HIGH-FIDELITY NUMERICAL INVESTIGATION OF A HIGH PRESSURE TURBINE COOLED VANE

A significant portion of a turbine blade/vane might experience a transitional boundary layer presumably on the suction surface, which affects the rate of heat transfer. Appropriate choice of the turbulence model on predicting a realistic heat transfer rate in the vicinity of the transition region is therefore imperative. Current study presents the numerical assessment effort of an industry-standard Energy Efficient Engine (E-3) high pressure turbine (HPT) second stage cooled vane in terms of flowfield and wall temperatures. In order to assess numerical predictions, wall temperature distributions at the cooling design point at 65% and 95% span locations reported by NASA, deduced from core engine tests accompanied with thermal analyses, are utilised. Conjugate heat transfer (CHT) analyses are carried out to reveal wall temperatures. External flow field is examined through the given normalized velocity distributions provided in the NASA reports. For these studies various turbulence models; k-omega SST, k-omega SST gamma-ReO transitional, Lag EB k-epsilon (EBL) and V2F were used. As an alternative, Large Eddy Simulation (LES) is conducted on external flow domain to validate the flow field around the vane by comparing the HTC distributions with the other turbulence models. However, LES analysis did not give promising results. It is found that near the end of leading edge region, metal temperatures are significantly overpredicted when the effect of laminar to turbulent transition is not taken into account. In all of the turbulence models, EBL model has captured the laminar-to-turbulent transition and give the best results. Finally, when the platform solid added to the CFD model, the simulation results get close to the NASA results due to conduction.