While it is well understood that certain nondimensional parameters, such as freestream Reynolds number and turbulence intensity, must be matched for proper design of film cooling experiments, uncertainty continues on the ideal method to scale film cooling flow rate. This debate typically surrounds the influence of the coolant to freestream density ratio and whether mass flux ratio or momentum flux ratio properly accounts for the density effects. Unfortunately, density is not the only fluid property to differ between typical wind tunnel experiments and actual turbine conditions. Temperature differences account for the majority of the property differences; however, attempts to match density ratio through the use of alternative gases can exacerbate these property differences.

A computational study was conducted to determine the influence of other fluid properties besides density — namely specific heat, thermal conductivity, and dynamic viscosity. CFD simulations were performed by altering traditional film cooling nondimensional parameters as well as others such as the Reynolds number ratio, Prandtl number ratio, and heat capacity ratio to evaluate their effects on adiabatic effectiveness and heat transfer coefficient. A cylindrical leading edge with a flat afterbody was used to simulate a turbine blade leading edge region. A single coolant hole was located 21.5° from the leading edge, angled 20° to the surface and 90° from the streamwise direction. Results indicated that thermal properties can play a significant role in understanding and matching results in cooling performance. Density effects certainly dominate; however, variations in conductivity and heat capacity can result in 10% or higher changes in the resulting heat flux to the surface when scaling ambient rig configurations to engine representative conditions.

This content is only available via PDF.
You do not currently have access to this content.