| description abstract | In order to test HP turbine stages under engine representative conditions on a heat transfer point of view, blow-down test rigs are often used. In these rigs the evolution of gas temperature, pressure, and density is similar to a step function. Hence, the use of hot-wires, which are sensitive to flow velocity, density, and temperature, is more difficult than in an incompressible flow at constant temperature. This investigation describes how the data reduction can be performed in such an environment in order to extract the velocity. The gas temperature is measured with a thermocouple and the gas density is derived from the measurement of the total pressure thanks to an iterative procedure. Once the velocity is derived, the turbulence can be computed. The effectiveness of the method is first demonstrated in a heated jet where both pressure and temperature are varied. Tests in the turbine facility are performed at turbine inlet temperatures of 480K. Thus, overheat ratios up to 1.9 had to be used, leading to a very high temperature of the tungsten platinum coated wire. The aging of the probe was very fast, causing a drift in the voltage output between the successive tests. A technique is proposed to minimize the aging effect. It consists in adapting the calibration based on the resistance of the wire measured before each test. Measurements were carried out at the turbine inlet and rotor outlet. At the turbine inlet, velocity radial profiles are obtained together with measurements of the turbulence intensity. The time-averaged data is compared with pneumatic probe measurements. At the rotor exit, the time-resolved periodic velocity fluctuations are analyzed using a phase-locked average technique. | |
