Objective
A better understanding of the mixing process between turbine blade rows is undoubtedly a key issue in one of the most challenging present day research fields, the investigation of the unsteady aerodynamic and heat transfer behaviour of gas turbines due to blade row interference effects.
The objective of the present research programme is to investigate the particularly interesting and important case of the flow mixing process behind a transonic inlet guide vane with trailing edge coolant flow ejection either from the pressure side or through the trailing edge.
Among the wealth of information from the two-dimensional tests, the information on the effect of Mach number and coolant flow rate on the trailing edge base pressure is most interesting because of the importance of the base pressure on the blade losses in the transonic domain on one side, and the entirely different evolution of the base pressure for the two trailing-edge geometries on the other side. The ability of the two-dimensional NS code to visualize the near wake flow pattern in the trailing-edge separation region contributed significantly to the understanding of the measured data. The inferior performance of the PS ejection blade is the result of the combined effect of the differences in the trailing-edge base pressures, the trailing-edge shock strength and the position of the maximum suction side velocity and the associated point of boundary layer transition. As regards the mixing of the coolant flow with the main flow one observes an extremely fast decrease in the maximum CO2 concentration in the near-wake, cm approximately = 10% at 10% of the axial chord, followed by a rather slow decrease in the far-wake region, cm = 4-5% at a downstream distance of 50% of the axial chord. Analogous tendencies are found for the wake temperature profiles in the annular cascade tests. The concentration and total pressure profiles appear to be similar to each other and can be approximated by a Gaussian distribution.
In comparing the straight and annular cascade tests it appears that the losses in the annular cascade are lower than what one might expect on the basis of the straight cascade tests. This different behaviour is explained by three-dimensional flow effects which are particularly strong in the trailing-edge region, as witnessed by three-dimensional NS visualizations.
The measurements with the fast rotating probe can be viewed as a success. However, the blockage effect of the probe appears to paly a dramatic role for the relative total pressure profile. This finding indicates enormous difficulties in the interpretation of the relative total pressure when measured with probes mounted on a rotor blade in a real turbine stage.
The mixing process will be studied in a straight cold flow suction type cascade tunnel and an annular compression tube cascade facility allowing the testing of full scale hp nozzle guide vanes at correctly simulated temperature ratios between main and coolant flows. The experimental programme requires the use of advanced measuring,testing and data acquisition techniques. Navier Stokes calculations are used to assist in the interpretation of the experimental data. In addition to a better physical understanding of the mixing process behind cooled guide vanes through the full integration of experimental and theoretical tools the project will provide additional valuable benchmark test cases for the validation of 2-dimensional and 3-dimensional Navier-Stokes codes.
Fields of science
Topic(s)
Data not availableCall for proposal
Data not availableFunding Scheme
CSC - Cost-sharing contractsCoordinator
1640 RHODE-ST-GENESE
Belgium