The demand for more power is rapidly increasing worldwide. Attention is turned to increasing the efficiency of modern methods for power generation. Gas turbines provide 35% of the power demands within the United States. Efficiency of gas turbines is defined in an ideal sense by the thermal efficiency of the Brayton Cycle. The overall efficiency of a gas turbine can be increased while simultaneously maximizing specific work output, by increasing the turbine inlet temperature. However, even with the advancements in modern materials in terms of maximum operating temperature, various components are already subjected to temperatures higher than their melting temperatures. An increase in inlet temperature would subject various components to even higher temperatures, such that more effective cooling would be necessary, whilst ideally using the same (or less) amount of cooling air bled from compressor. Improvements in the performance of these cooling techniques is thus required. The focus of this thesis is on one such advanced cooling technique, namely film cooling. The objective of this study is to investigate the effects of coolant density on the jet structure for different multi-row film cooling configurations. As research is performed on improving the performance of film cooling, the available conditions during testing may not reflect actual engine-like conditions. Typical operating density ratio at engine conditions are between 1.5 and 2, while it is observed that a majority of the density ratios tested in literature are between 1 and 1.5. While these tests may be executed outside of engine-like conditions, it is important to understand how density ratio effects the flow physics and film cooling performance. The density ratio within this study is varied between 1.0 and 1.5 by alternating the injecting fluid between air and Carbon Dioxide, respectively. Both a simple cylindrical and fan-shape multi-row film cooling configuration are tested in the present study. In order to compare the results collected from these geometries, lateral and spanwise hole-to-hole spacing, metering hole diameter, hole length, and inclination angle are held constant between all testing configurations. The effect of fluid density upon injection is examined by independently holding either blowing, momentum flux, or velocity ratio constant whilst varying density ratio. Comparisons between both of the film cooling configurations are also made as similar ratios are tested between geometries. This allows the variation in flow structure and performance to be observed from alternating the film cooling hole shape. Particle Image Velocimetry (PIV) is implemented to obtain both streamwise and wall normal velocity measurements for the array centerline plane. This data is used to examine the interaction of the jet as it leaves the film cooling hole and the structure produced when the jet mixes with the boundary layer. Similarities in jet to jet interactions and surface attachment between density ratios are seen for the cylindrical configuration when momentum flux ratio is held constant. When observing constant blowing ratio comparisons of the cylindrical configurations, the lower density ratio is seen to begin detaching from the wall at M = 0.72 with little evidence of coolant in the near wall region. However, the higher density cylindrical injection retains its surface attachment at M = 0.74 with noticeably more coolant near the wall, because of significantly lower momentum flux ratio and lower "jetting" effect. The fan-shape film cooling configuration demonstrates improved performance, in terms of surface attachment, over a larger range of all ratios than that of the cylindrical cases. Additionally, the fan-shape configuration is shown to constantly retain a thicker layer of low velocity fluid in the near wall region when injected with the higher density coolant, suggesting improved performance at the higher density ratio. When tracking the jet trajectory, it is shown that the injection of CO2 through the cylindrical configuration yields a higher centerline wall normal height per downstream location than that of the lower density fluid. Comparing the results of the centerline tracking produced by the third and fifth rows for both the injection of air and CO2, it is confirmed that the fifth row of injection interacts with the boundary layer at a great wall normal height than that of the third row. Additionally, when observing the change in downstream trajectory between the fifth and seventh row of injection, a significant decrease in wall normal height is seen for the coolant produced by the seventh row. It is believed that the lack of a ninth row of injection allows the coolant from the seventh row of injection to remain closer to the target surface. This is further supported by the observation of the derived pressure gradient field and the path streamlines take while interacting with the recirculatory region produced by the injection of coolant into the boundary layer. Further conclusions are drawn by investigating the interaction between momentum thickness and the influence of blowing ratio. Relatively constant downstream momentum thickness is observed for the injection of lower density fluid for the blowing ratio range of M= 0.4 to 0.8 for the cylindrical configuration. It is suggested that a correlation exists between momentum thickness and film cooling performance, however further studies are needed to validate this hypothesis.

Graduation Date





Kapat, Jayanta


Master of Science in Mechanical Engineering (M.S.M.E.)


College of Engineering and Computer Science


Mechanical and Aerospace Engineering

Degree Program

Mechanical Engineering; Thermo-Fluids Track









Release Date

August 2017

Length of Campus-only Access


Access Status

Masters Thesis (Open Access)