Search

The complex three-dimensional flow in the endwall region near the base of a turbine blade has an important impact on the local heat transfer. The initial horseshoe vortex, the passage vortex, and resulting corner vortices cause large variations in heat transfer over the entire endwall region. Due to these large surface gradients in heat transfer, conventional measurement techniques generally do not provide an accurate determination of the local heat transfer coefficients. In the present study, the heat/mass transfer analogy is used to examine the local transport coefficients for two different endwall boundary layer thicknesses and two free-stream Reynolds numbers. A linear turbine blade cascade is used in conjunction with a removable endwall plate. Naphthalene (C10H8) is cast into a mold on the plate and the rate of naphthalene sublimation is determined at 6000 + locations on the simulated endwall by employing a computer-aided data acquisition system. This technique allows one to obtain detailed contour plots of the local convection coefficient over the entire endwall. By examining the mass transfer contours, it is possible to infer information on the three-dimensional flow in the passage between the blades. Extremely high transport coefficients on the endwall indicate locations of potential overheating and failure in an actual turbine.

There is increasing evidence that reproducibility of the plasma spray process depends to a large degree on the reproducibility of the plasma jet which is utilised for spraying. Recent research using a broad range of diagnostic techniques, including, LDA, electric and acoustic probing, spectroscopy, and enthalpy probes, has greatly enhanced our present understanding of the thermal and fluid dynamic characteristics of plasma jets. Experimental results strongly imply that the entrainment mechanism in turbulent plasma jets is more of an engulfment type process as opposed to simple diffusion. Turbulence measurements indicate that the velocity fluctuations are non–isotropic, contrary to what has been assumed in most numerical work. Measurements of electric, acoustic, and light fluctuations demonstrate that there are correlations between some of these fluctuations and that turbulence in the plasma jet may be strongly affected by fluctuations in the torch nozzle. Temperature measurements using spectrometric and enthalpy probes indicate strong discrepancies, particularly in the jet fringes.

There is increasing evidence that the entrainment of cold gas surrounding a turbulent plasma jet is more of an engulfment type process rather than simple diffusion. A variety of diagnostic techniques have been employed to determine the development of turbulence in a plasma jet and to measure concentration and temperatures of the cold gas entrained into atmospheric-pressure argon plasma jets in ambient argon or air. The results indicate that the transition to turbulence causes a rapid drop of the axial jet velocity due to entrainment of the cold gas surrounding the plasma jet. Dissipation of the cold engulfed gas bubbles by molecular diffusion is relatively slow if molecular gases (for example air) are entrained, as indicated by conditional sampling and CARS measurements. Temperature measurements using emission spectroscopy and enthalpy probes show strong discrepancies in the jet fringes.