The Nozzle Guide Vane (NGV) of High Pressure (HP) turbine of modern military gas turbine is extensively cooled to reap the benefits of using higher cycle temperatures and to ensure longer life of the turbine components. NGV of the HP turbine encounters the maximum temperature from the combustion chamber. Hence, cooling of the NGV is essential from creep and thermal fatigue points of view. Compressor Bleed air is used for cooling the NGV and due to addition of mass, leads to decrease in turbine efficiency and increase in losses. The objective of the present study is to determine numerically using 3D Navier Stokes (NS) solver, the effect of Leading edge (LE) and Trailing Edge (TE) coolant ejection on the performance of the NGV via blade loading, efficiency and exit flow angles for different mass ratios
Sunday, 3 January 2021
LARGE EDDY SIMULATION OF FLOW OVER NOZZLE GUIDE VANE OF A TRANSONIC HIGH PRESSURE TURBINE
LARGE EDDY SIMULATION OF FLOW OVER NOZZLE GUIDE VANEOF A TRANSONIC HIGH PRESSURE TURBINE
Fifth generation fighter aircraft engines demand thrust to weight ratios in excess of 8 and moderate Specific Fuel Consumption. This calls for lower weight and higher component efficiencies. To improve turbine efficiency, it is necessary to understand the aerodynamic loss generation mechanism in the flow field. Since most such mechanisms are unsteady in nature, it is often necessary to carry out unsteady analysis of the turbine stage. It is desirable to study the loss generation mechanisms in the stationary Nozzle Guide Vanes (NGVs) and the rotor blades through unsteady analysis. In this paper, NGV of a transonic high pressure turbine has been analysed using Large Eddy Simulation (LES) and the results have been compared with Reynolds Averaged Navier Stokes (RANS) simulations and Unsteady RANS (URANS) simulations. The investigation is focused on the mechanisms of loss generation in the NGV at low Reynolds number condition of 1.5x106 (high altitude low Mach number condition) where shock boundary layer interactions are predominant.