Surface roughness is known to have a significant impact on turbine heat loads and performance. As the exposure history of turbine flow path components is increased, many of the external surfaces become rougher, which results in the increase of heat loads and friction losses. While there have been several investigations that included surfaces with uniform or two dimensional roughness patterns there is now a clear need to measure the influence of real surface roughness on turbine blade flow and heat transfer.

The objective of the proposed investigation is to conduct measurements that will reveal the influence of real surface roughness on the boundary layer. The test surface would be a large-scale version of rough surfaces that are specified by the Air Force Research Laboratory/Propulsion Directorate. That is, geometric models of real surfaces would be employed but in much larger size so we can obtain high quality velocity and turbulence data in the near wall region, including the viscous layer. Thus, the measurements obtained would be related to convective heat transfer from the surface. Since the near wall region is still subsonic in a transonic boundary layer, such detailed measurements should be valuable to assess and guide development of computational fluid dynamics models proposed for predictions of flows over realistic surfaces in gas turbine passages at engine conditions.

The measurements will be conducted in the Matched-Index-of-Refraction (MIR) Facility at the Idaho National Engineering and Environmental Laboratory, the largest MIR facility in the world. We expect to obtain a grid of measurements of mean and turbulence quantities within the viscous layer and near wall region of a boundary layer over each real surface model. For the near wall region, the key to correct modeling is to match the non-dimensional parameters in terms of wall scaling and to have the same range of dimensionless pressure gradients plus the proper scaled geometry. The test section inlet section would be modified to induce elevated freestream turbulence, to simulate gas turbine conditions, and the streamwise pressure gradient would be adjusted to match operating conditions appropriately. Measurements would be conducted with a 2-component fiber optics LDV system. The locations where near-wall data are obtained would be fabricated of fused quartz having the same refractive index as the oil in the MIR facility. This approach will allow for high signal-to-noise measurements even within the part of the viscous layer that is buried in the terrain of the roughness elements. In addition, it will provide for improved spatial and temporal resolution compared to typical experiments in wind or water tunnels.

I.N.E.E.L. MATCHED INDEX OF REFRACTION OIL TUNNEL