The effect of hot-streaks on HP vane surface and endwall heat transfer: An experimental and numerical study

Pronounced non-uniformities in combustor exit flow temperature (hot-streaks), which arise because of discrete injection of fuel and dilution air jets within the combustor and because of end-wall cooling flows, affect both component life and aerodynamics. Because it is very difficult to quantitatively predict the affects of these temperature non-uniformities on the heat transfer rates, designers are forced to budget for hot-streaks in the cooling system design process. Consequently, components are designed for higher working temperatures than the mass-mean gas temperature, and this imposes a significant overall performance penalty. An inadequate cooling budget can lead to reduced component life. An improved understanding of hot-streak migration physics, or robust correlations based on reliable experimental data, would help designers minimise the overhead on cooling flow that is currently a necessity. A number of recent research projects sponsored by a range of industrial gas turbine and aero-engine manufacturers attest to the growing interest in hot-streak physics. This paper presents measurements of surface and end-wall heat transfer rate for an HP nozzle guide vane (NGV) operating as part of a full HP turbine stage in an annular transonic rotating turbine facility. Measurements were conducted with both uniform stage inlet temperature and with two non-uniform temperature profiles. The temperature profiles were non-dimensionally similar to profiles measured in an engine. A difference of one half of an NGV pitch in the circumferential (clocking) position of the hot-streak with respect to the NGV was used to investigate the affect of clocking on the vane surface and end-wall heat transfer rate. The vane surface pressure distributions, and the results of a flow-visualisation study, which are also given, are used to aid interpretation of the results. The results are compared to two-dimensional predictions conducted using two different boundary layer methods. Experiments were conducted in the Isentropic Light Piston Facility (ILPF) at QinetiQ Famborough, a short duration engine-size turbine facility. Mach number, Reynolds number and gas-to-wall temperature ratios were correctly modelled. It is believed that the heat transfer measurements presented in this paper are the first of their kind.