Detailed heat transfer distributions of narrow impingement channels for cast-in turbine airfoils

Gas turbine operation at elevated temperatures ensures increased thermal efficiency and useful power output. However, the industrial tendency to push further firing temperatures is limited by the capabilities of the current turbine blade cooling technologies which approach their limits. Nowadays, the capabilities of the foundry industry to produce high integrity and dimensionally accurate castings allow the production of integrally cast turbine airfoils with the so-called double-wall cooling technology. Cast-in turbine airfoils include a more distributed form of cooling, where the coolant can be injected in the form of impingement jets within the wall rather than the hollow of the airfoil increasing dramatically the heat exchanged capabilities. Narrow impingement cooling passages can be therefore generated close to the external hot gas flow. In this study detailed heat transfer distributions for all internal surfaces of narrow impingement channels are evaluated. The test models consist of a single row of five impingement jets investigated over a range of engine representative Reynolds numbers. Effects of jet-to-jet spacing, channel width and height and impingement jet pattern are independently examined composing a test matrix of 22 large-scale geometries. For the evaluation of the heat transfer coefficients, the transient liquid crystal technique is used considering also thermocouple thermal inertia effects. Given that the use of different impingement geometries could play an important role on the distribution of convection coefficients, and hence, on the homogeneity of blade metal temperatures, the experimental results of this study can be used along with the external heat load for the development of thermal design models for integrally cast turbine vanes and blades.

Application of the Transient Heater Foil Technique for Heat Transfer and Film Cooling Effectiveness Measurements on a Turbine Vane Endwall

Dominique Charbonnier, Magnus Jonsson, Peter Ott

The paper presents an application of the transient heater foil measurement technique using thermochromic liquid crystals (TLC) to endwall heat transfer and film cooling investigations in a transonic turbine vane cascade. The film cooling configuration consists of an upstream slot, representing the leakage flow area between the interface of the combustor and the turbine, and several rows of cylindrical and fan-shaped holes within the passage. With the transient method chosen, the heat transfer and adiabatic film cooling effectiveness distributions can be obtained simultaneously together with the local heat release within the heater foil. Therefore, the heat release in the foil is not required to be uniform, and the foil can contain discrete holes in the film cooling configuration. Some new developments are presented, which are directed towards improved application of the transient heater foil method for such a complex configuration. This includes tailoring the foil heat-release distribution towards the expected heat transfer patterns, supported by numerical Finite-Element computations and the use of a double-TLC mixture for improved time-wise TLC indications. Additionally, CFDsimulations were used to evaluate the recovery temperature distribution through the vane cascade without film cooling. The experiments were performed in the linear cascade facility at the EPFL-Lausanne. A compressor provides a continuous air flow at near-ambient temperature regulated with heat exchangers. Carbon dioxide is used as coolant in order to achieve engine-representative density ratio between coolant and main flow. Multiple experiments with the same main and coolant flow settings but varying heat flux levels and coolant injection temperatures have been performed and simultaneously analysed using nonlinear regression analysis. The time required between successive experiments to return to homogenous initial conditions, as required by the transient method, has been analysed using an analytical solution for heat-on-heat-off conditions. This permits the model assumption of onedimensional conduction within a semi-infinite wall with a heat releasing layer on the top. Example results for cases without cooling, with film cooling from rows of discrete holes and the addition of slot film cooling are used to illustrate the benefit of the new approaches for the investigated vane cascade.

American Soc Mechanical Engineers, Three Park Avenue, New York, Ny 10016-5990 Usa2008

Local Heat Transfer Distributions of Small Impingement Configurations: A Detailed Experimental Procedure Using Transient Heat Transfer Methods

Peter Ott, Alexandros Terzis, Bernhard Weigand

Impingement cooling is widely used in the cooling of turbine airfoils at critical regions where high heat transfer rates are required due to the large local convection coefficients that can be achieved. An accurate determination of local heat transfer distributions, is therefore, essential for reliable calculations of blade metal temperatures. In modern cast-in turbine airfoils, e.g. Lutum et al. [1], narrow impingement cooling channels can be formed, where the coolant is practically injected within the wall rather than the hollow of the airfoil, as shown in Figure 1. Narrow impingement channels consist of single or double rows of several cooling jets generating a relatively low amount of crossflow while the small wall thickness of a typical airfoil (~2mm) dictates significant heat transfer in all the interior surfaces of the cavity. A turbine airfoil can include a plurality of such small impingement configurations in the spanwise direction ensuring homogenous distribution of the material temperature. Contrary to the large amount of literature sources for multi-array impingement cooling systems, e.g. a review can be found in Weigand and Spring [2], little amount of research focused on narrow impingement channels with limited information for the sidewalls and impingement plate. A few studies can be found in Gillespie et al. [3], Chambers et al. [4], Ricklick et al. [5], Lamont et al. [6] and Terzis et al. [7]. This paper describes the overall experimental procedure and the heat transfer distributions for all the interior walls of a narrow impingement channel consisting of a single row of five inline jets. Heat transfer coefficients are evaluated with the transient liquid crystal technique considering also thermocouple thermal inertia effects.

2014

Experimental investigations of showerhead film cooling on the leading edge of a turbine blade

Guillaume Wagner

The turbine components of a gas turbine need extensive cooling to withstand the high heat loads generated by the flow of the hot combustion gases. The leading edge of a turbine blade is one of the areas that faces the hottest gas flow conditions and is thus one of the most critical areas to be cooled. The stagnation region of turbine airfoils is generally cooled with showerhead cooling scheme, which consist in injecting coolant onto the stagnation region through closely spaced rows of film cooling holes. This provides convective cooling inside the holes and produces a layer of coolant film on the external surface. The objective of this study is to investigate experimentally the influence of mainstream incidence on showerhead film cooling. The film cooling effectiveness and heat transfer coefficient were thus measured on a series of symmetric blunt bodies with different showerhead configurations simulating the leading edge of turbine airfoils. Tests have been performed at 0° on-design and 5° off-design mainstream incidences at an approach Mach number of 0.26. Besides the mainstream incidence angle, the investigated parameters were the blowing ratio, the wedge angle, the row location, the spanwise pitch between the holes and the hole shape. In addition to the film cooling measurements, baseline heat transfer tests were performed on test models without film cooling holes in order to evaluate the film cooling performances. To characterize the mainflow conditions and the velocity profile on the test models, aerodynamic measurements have also been carried out. The heat transfer measurements have been performed using the transient liquid crystal technique. The transient experiments were generated by a pre-conditioning system and a rapid insertion mechanism. The data analysis was adapted to measurements on highly curved surfaces using the approximate analytical solution of the heat conduction into a solid cylinder with convective boundary conditions on the surface. Time-varying adiabatic wall temperatures were accounted for using Duhamel's superposition theorem. The first tested showerhead geometry was a triangular blunt body representative of a turbine rotor blade showerhead with a circular leading edge, a 21.1° wedge angle and two film cooling rows. Because of jet lift-off, the film cooling performances of this two-row configuration were relatively low and were almost unaffected by the increase of blowing ratio from 1.5 to 3.9. The heat transfer coefficient was found to be greatly increased at the row location in comparison to the case without film cooling. At 5° off-design incidence angle, the stagnation line moves passed the pressure side row. The two rows were then located on the same side of the stagnation line and were both blowing towards the suction side. On the suction side the film cooling protection was thus very good due to film accumulation but the pressure side was left almost unprotected resulting in highly asymmetric film cooling performances. Two-dimensional FEM calculations of the metal temperature in this second geometry have been performed at engine representative conditions for design and off design mainstream incidences. It was found that the contribution of the film cooling to cool the leading edge region was rather small in comparison to the contribution of the internal convective cooling. Showerhead film cooling was however required to effectively cool the stagnation region and to obtain a uniform surface temperature. The second tested showerhead geometry was similar to the first geometry but with an increased wedge angle of 55° and cooling rows located further downstream. The film cooling performances were found to decrease when the blowing ratio was increased from 1.2 to 3.0 because of jet lift-off. At middle blowing ratio (2.0), the film cooling effectiveness had similar values as for the first geometry. However, the heat transfer increase due to the presence of the film cooling jets was much higher for the second geometry because of higher local mainstream velocities at the injection locations. The resulting film cooling performances were thus lower for the second geometry than for the first geometry. Nevertheless, the larger wedge angle and the cooling rows located further downstream assured that the stagnation line stayed in-between the two cooling rows even at 5° off-design incidence. The second geometry had thus the advantage to be much less sensitive to incidence angle and, unlike the first geometry, provided very similar film cooling performances at 5° and at 0° incidence. The use of shaped holes was found to greatly improve the film cooling performances of the second geometry due to a better lateral coverage and a reduced tendency of jet lift-off.