Computational and Experimental Studies of Midpassage Gap Leakage and Misalignment for a Non-Axisymmetric Contoured Turbine Blade Endwall

Abstract

Turbine vanes and blades are generally manufactured as single or double airfoil sections that must each be installed onto a turbine disk. Between each section, a gap at the endwalls through the blade passage is present, through which high pressure coolant is leaked. Furthermore, sections can become misaligned due to thermal expansion or centrifugal forces. Flow and heat transfer around the gap is complicated due to the interaction of the mainstream and the leakage flow. An experimental and computational study was undertaken to determine the physics of the leakage flow interaction for a realistic turbine blade endwall, and assess whether steady RANS CFD, commonly used for non-axisymmetric endwall design, can be used to accurately model this interaction. Computational models were compared against experimental observations of endwall heat transfer on a contoured endwall with a midpassage gap. Endwall heat transfer coefficients were determined experimentally by using infrared thermography to capture spatially-resolved surface temperatures on a uniform heat flux surface (heater) attached to the endwall. Predictions and measurements both indicated an increase in endwall heat transfer with increasing gap leakage flow, although the distribution of heat transfer coefficients along the gap was not well captured by CFD. A misalignment of the blade endwall causing a forward-facing step for the near-endwall flow resulted in a large highly turbulent recirculation region downstream of the step and high local heat transfer that was overpredicted by CFD. Conversely, a backward-facing step reduced turbulence and local heat transfer. The misprediction of local heat transfer around the gap is thought to be caused by unsteady interaction of the passage secondary flow and gap leakage flow, which cannot be well-captured by a steady RANS approach.

Direct numerical simulation of slot film cooling downstream of misaligned plates

“…The lack of near-wall data motivated computational work. Lange, Lynch, and Lewis (2016) compared their Reynolds averaged Navier-Stokes (RANS) with Cardwell et al's (2005) and their own experiments and noted a number of RANS's inadequacies. In another RANS study, Kim, Chung, Rhee, and Cho (2016) reported the cooling effectiveness as a function of the misalignment's size.…”

“…Here, we are able to do DNS because the Reynolds number of the flow is not very high. In Cardwell et al (2005) and Lange et al (2016), the authors matched to engine conditions and the flow's Reynolds number is Re in = U 0 C/𝜈 = O(10 5 ), where U 0 is the free-stream velocity, C is the chord length and 𝜈 is the kinematic viscosity. The height of the incoming boundary layer is approximately 𝛿 = 0.1C, and therefore the above Reynolds number corresponds to a Reynolds number of Re b = U 0 𝛿/𝜈 = O(10 4 ) for the incoming boundary layer: this is quite moderate, and DNSs of boundary layers at this Reynolds numbers are reported in Schlatter and Örlü (2010) and Pirozzoli, Grasso, and Gatski (2004).…”