Computational Prediction of Heat Transfer to Gas Turbine Nozzle Guide Vanes With Roughened Surfaces

Abstract

The local Mach number and heat transfer coefficient over the aerofoil surfaces and endwalls of a transonic gas turbine nozzle guide vane have been calculated. The computations were performed by solving the time-averaged Navier–Stokes equations using a fully three-dimensional computational code (CFDS), which is well established at Rolls-Royce. A model to predict the effects of roughness has been incorporated into CFDS and heat transfer levels have been calculated for both hydraulically smooth and transitionally rough surfaces. The roughness influences the calculations in two ways; first the mixing length at a certain height above the surface is increased; second the wall function used to reconcile the wall condition with the first grid point above the wall is also altered. The first involves a relatively straightforward shift of the origin in the van Driest damping function description, the second requires an integration of the momentum equation across the wall layer. A similar treatment applies to the energy equation. The calculations are compared with experimental contours of heat transfer coefficient obtained using both thin-film gages and the transient liquid crystal technique. Measurements were performed using both hydraulically smooth and roughened surfaces, and at engine-representative Mach and Reynolds numbers. The heat transfer results are discussed and interpreted in terms of surface-shear flow visualization using oil and dye techniques.

Thermal Performance of Different Cooled Tips in A High-Pressure Turbine Cascade

“…The experimental data used to validate the computational fluid dynamic (CFD) results are taken from Newton [27]. The uncertainty in the temperature measurement was approximately 1%, which resulted in 5% uncertainty in the experimental value of heat-transfer coefficient and film-cooling effectiveness [28,29]. Figure 5 shows the predicted and measured static-pressure distribution on the end wall.…”

Roughness Coefficient of a Highly Calcinated Penstock

“…On the other hand, manufacturers use directly measurable quantities Ra, Rq and Rz to represent the surface texture of their products. Equations available in literature [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] that try to relate these parameters vary greatly and there is no consensus on a single relation on these parameters. Equations proposed by [19][20][21][22][23][24] mostly use sand grains, emery papers or machined surfaces to produce the empirical equations using Ra, while suggested equations for groove height or liquid crystal surfaces, Rq [26][27] is mostly used for surfaces with sprayed particles as an alternative.…”

“…Naturally occurring calcination deposits were not studied in any of these works; however, for a wide range of surfaces, a general trend is to use Rz, peak to valley roughness height, equal to surface roughness value (ε) (Rz ≈ ε) [16] [17]. This corresponds to 287.45 μm for the calcinated surface sample.…”

Effects of Surface Roughness on Heat Transfer and Aerodynamic Performance of Turbine Airfoils

“…Blair used Han's method (1985)for calculating the local friction coefficients as a common input to the four rough surface heat transfer correlations. Guo et al (1996) measured vane heat transfer in an annular rig using liquid crystals with a reported roughness R. of 25 micrometers, and also a smooth surface R. of less than 1 micrometer with thin film heat flux gages. R z is defined as the average peak-to-peak roughness height measured with a profilometer.…”

“…Blair (1992) and Boyle (1993) suggested that k s = R 2 . Dunn et al (1992) used Ic s = R / (13 and Guo et al (1996) found that k s = Rz for the best comparison of their experimental data. Bogard et al (1996) adopted a roughness shape / density parameter of Sigal and Danberg (1990) to characterize vane roughness, and then modeled large scale, flat plate surfaces to mimic this parameter value.…”