Open Access
Gokce, Zeki Ozgur
Graduate Program:
Aerospace Engineering
Doctor of Philosophy
Document Type:
Date of Defense:
October 15, 2012
Committee Members:
  • Cengiz Camci, Dissertation Advisor
  • Cengiz Camci, Committee Chair
  • Sven Schmitz, Committee Member
  • Savas Yavuzkurt, Committee Member
  • Timothy Francis Miller, Committee Member
  • Vortex generator
  • heat transfer enhancement
  • gas turbine cooling channel flow
The gas turbine is one of the most important parts of the air-breathing jet engine. Hence, improving its efficiency and rendering it operable under high temperatures are constant goals for the aerospace industry. Two types of flow within the gas turbine are of critical relevance: The flow around the first row of stator blades (also known as the nozzle guide vane blade - NGV) and the cooling flow inside the turbine blade cooling channel. The behavior of the former flow type affects the total pressure level downstream of the NGV, thereby affecting the efficiency of the gas turbine. The behavior of the latter flow type affects the cooling performance of turbine blades, having a direct effect on the maximum total temperature the turbine material can withstand, as well as affecting the level of thrust produced via the total temperature. The flow in the vicinity of the turbine blades and the endwall boundary layer has a great effect on the behavior of the overall flow through the gas turbine. The flow near the pin-fins contained inside turbine blade cooling channels dictates the cooling performance of the blade. These two facts have prompted the aerospace industry to investigate the potential benefits of modifying the shape of the endwall. The results of various studies showed improvements in both flow types. Following this thought process, this thesis study focused on finding innovative ways of attaining even higher performance. We previously demonstrated that adding upstream endwall fences leads to beneficial changes in the flow near the NGV blade. Inspired by our prior findings, we decided to analyze the effects of endwall shape modifications on turbine cooling channel flow, in addition to the flow near the NGV. In short, the subject of this thesis work was to search for methods that could improve the characteristics of these two types of flows, thus enabling superior engine performance. The innovative aspect of our work was to apply an endwall shape modification previously employed by non-aerospace industries for cooling applications, to the gas turbine cooling flow which is vital to aerospace propulsion. Since the costs of investigating the possible benefits of any idea via extensive experiments could be quite high, we decided to use computational fluid dynamics (CFD) followed by experimentation as our methodology. We decided to analyze the potential benefits of using vortex generators (VGs) as well as the rectangular endwall fence. Since the pin-fins used in cooling flow are circular cylinders, and since the boundary layer flow is mainly characterized by the leading edge diameter of the NGV blade, we modeled both the pin-fins and the NGV blade as vertical circular cylinders. The baseline case consisted of the cylinder(s) being subjected to cross flow and a certain amount of freestream turbulence. The modifications we made on the endwall consisted of rectangular fences. In the case of the cooling flow, we used triangular shaped, common flow up oriented, delta winglet type vortex generators as well as rectangular endwall fences. The channel contained singular cylinders as well as staggered rows of multiple cylinders. For the NGV flow, a rectangular endwall fence and a singular cylinder were utilized. Using extensive CFD modeling and analysis, we confirmed that placing a rectangular endwall fence upstream of the cylinder created additional turbulent mixing in the domain. This led to increased mixing of the cooler flow in the freestream and the hotter flow near the endwall. As a result, we showed that adding a rectangular fence created a 10% mean heat transfer increase downstream of the cylinder. When vortex generators are used, as the flow passes over the sharp edges of the vortex generators, it separates and continues downstream in a rolling, helical pattern. Combined with the effect generated by the orientation of the vortex generators, this flow structure mixes the higher momentum fluid in the freestream with lower momentum fluid in the boundary layer. Similar turbulent mixing behavior is observed over the entire domain, near the cylinders and the side walls. As a result, the heat transfer levels over the wall surfaces are increased and improved cooling is achieved. The improvements in heat transfer are obtained at the expense of acceptable pressure losses across the cooling channel. When the vortex generators are used, the CFD modeling studies showed that overall heat transfer improvements as high as 27% compared to the baseline case are observed inside a domain containing multiple rows of cylinders. A price in the form of 13% pressure loss increase across the channel is paid for the heat transfer benefits. Experiments conducted in the open loop wind tunnel of the Turbomachinery Aero-Heat Transfer Laboratory of the Department of Aerospace Engineering of Penn State University supported the general positive trend of these findings, with a 14% overall increase in heat transfer over the constant heat flux surface when vortex generators are installed, accompanied by an 8% increase in pressure loss. To our knowledge, our study is the first to demonstrate the positive effects of vortex generators used in conjunction with circular pin-fins on the heat transfer properties of gas turbine blade cooling channel flow. The findings of our study may also have practical implications for other scientific and industrial fields using flows of similar Reynolds numbers.