Turbine vanes experience high convective surface heat transfer as a consequence of the turbulent flow exiting the combustor. Before improvements to vane heat transfer predictions through boundary layer calculations can be made, we need to understand how the turbulent flow in the inviscid region of the passage reacts as it passes between two adjacent turbine vanes. In this study, a scaled-up turbine vane geometry was used in a low-speed wind tunnel simulation. The test section included a central airfoil with two adjacent vanes. To generate the 20 percent turbulence levels at the entrance to the cascade, which simulates levels exiting the combustor, an active grid was used. Three-component laser-Doppler velocimeter measurements of the mean and fluctuating quantities were measured in a plane at the vane midspan. Coincident velocity measurements were made to quantify Reynolds shear stress and correlation coefficients. The energy spectra and length scales were also measured to give a complete set of inlet boundary conditions that can be used for numerical simulations. The results show that the turbulent kinetic energy throughout the inviscid region remained relatively high. The surface heat transfer measurements indicated high augmentation near the leading edge as well as the pressure side of the vane as a result of the elevated turbulence levels. [S0889-504X(00)02302-3]

High free-stream turbulence along a gas turbine airfoil and strong secondary flows along the endwall have both been reported to increase convective heat transfer significantly. This study superimposes high free-stream turbulence on the naturally occurring secondary flow vortices to determine the effects on the flowfield and the endwall convective heat transfer. Measured flowfield and heat transfer data were compared between low free-stream turbulence levels (0.6 percent) and combustor simulated turbulence levels (19.5 percent) that were generated using an active grid. These experiments were conducted using a scaled-up, first-stage stator vane geometry. Infrared thermography was used to measure surface temperatures on a constant heat flux plate placed on the endwall surface. Laser-Doppler Velocimetry (LDV) measurements were performed of all three components of the mean and fluctuating velocities of the leading edge horseshoe vortex. The results indicate that the mean flowfields for the leading edge horseshoe vortex were similar between the low and high free-stream turbulence cases. High turbulence levels in the leading edge–endwall juncture were attributed to a vortex unsteadiness for both the low and high free-stream turbulence cases. While, in general, the high free-stream turbulence increased the endwall heat transfer, low augmentations were found to coincide with the regions having the most intense vortex motions. [S0889-504X(00)00704-2]

Gaining a good understanding of how high freestream turbulence augments heat transfer is important for predicting thermal loadings for turbine blades and vanes. This study was aimed at documenting the surface heat transfer and the highly turbulent flowfield around a stator vane. The effects of turbulence levels between 11% and 24% were studied. At the highest turbulence level, two different Reynolds numbers (Reex = 6 × 105 and 1.2 × 106) and two different length scales were also studied. Three-component laser Doppler velocimeter measurements of the velocity fluctuations indicated that downstream of the active grid there was an initial decay of the turbulent kinetic energy which then leveled off at about one leading edge radius upstream of the vane. Inside the vane passage the turbulent kinetic energy increased slightly and then decayed through the passage. The surface heat transfer showed the largest augmentations on the pressure side of the vane with higher augmentations at higher turbulence levels, smaller length scales, and higher Reynolds numbers.

No abstract

High freestream turbulence levels have been shown to greatly augment the heat transfer on a gas turbine airfoil. To better understand these effects, this study has examined the effects elevated freestream turbulence levels have on the boundary layer development along a stator vane airfoil. Low freestream turbulence measurements (0.6 percent) were performed as a baseline for comparison to measurements at combustor simulated turbulence levels (19.5 percent). A two-component LDV system was used for detailed boundary layer measurements of both the mean and fluctuating velocities on the pressure and suction surfaces. Although the mean velocity profiles appeared to be more consistent with laminar profiles, large velocity fluctuations were measured in the boundary layer along the pressure side at the high freestream turbulence conditions. Along the suction side, transition occurred further upstream due to freestream turbulence.