A 3D conjugate heat transfer simulation of a radially cooled gas turbine vane has been performed using STAR-CD™ code and the metal temperature distribution of the blade has been obtained. The study focused on the linear NASA-C3X cascade, for which experimental data are available; the blade is internally cooled by air through ten radially oriented circular cross section channels. According to the chosen approach, boundary conditions for the conjugate analysis were specified only at the inlet and outlet planes and on the openings of the internal cooling channels: neither temperature distribution nor heat flux profile were assigned along the walls. Static pressure, external temperature and heat transfer coefficient distributions along the vane were compared with experimental data. In addition, in order to asses the impact of transition on heat transfer profile, just the external flow (supposed fully turbulent in the conjugate approach) was separately simulated with TRAF code too and the behaviour of the transitional boundary layer has been analyzed and discussed. Loading distributions were found to be in good agreement with experiments for both conjugate and non conjugate approaches, but, since both pressure and suction side exhibit a typical transitional behavior, HTC profiles obtained without taking into account transition severely overestimate experimental data especially near the leading edge. Results confirm the significant role of transition in predicting heat transfer and, therefore, vane temperature field when a conjugate analysis is performed.
The use of pin fin arrays in channels is one of the best choices to enhance overall heat transfer in gas turbine trailing edge blade cooling. Furthermore, in this particular application, the use of cross-pins in the trailing edge section of a turbine blade is a good way for supplying structural integrity to the blade itself. In this paper, results of several 3D RANS calculations performed in channels with cross-pins disposition such as in a typical trailing edge of a gas turbine blade are shown. Numerical calculations were compared with experimental results obtained on the same geometries using a transient Thermochromic Liquid Crystals (TLC) based technique. Goals of this comparison are both the evaluation of the accuracy of CFD packages with standard two equation turbulence models in heat transfer problems with complex geometries and the analysis of flow details to complete and support experimental activity. Two computational domains have been considered: they both consist in a wedge shaped channel with a stream-wise normal pin fin or pedestal arrays. The aim of the numerical analysis is the evaluation of convective Heat Transfer Coefficient (HTC) on the planar bottom surface of the wedge-shaped duct: this surface is commonly named “endwall” surface. Detailed analysis of the flow field points out the coexistence of an horse-shoe vortex, a stagnant wake behind the pin and a mean flow acceleration due to convergent shape of the channel. Calculations reveal the presence of a weak jet-like flow field toward endwall surfaces caused by the strong recirculation behind each pin.
Goal of this work is to define the main issues and guidelines for an accurate heat transfer CFD simulation of internal ribbed ducts. To this aim, two different ribbed ducts (AR = 1,3) have been experimentally investigated to obtain a data set useful to validate numerical analyses. Experimental HTC contour maps have been obtained using unsteady TLC technique. CFD activity deals with numerical simulation using both a commercial (Star-CD™) and an “in house” solver (HybFlow). Four different variants of the well-known two-equation turbulence models have been considered. Low cost heat transfer predictions of internal ducts are highly demanded by industry despite the uncommon complexity of modern internal coolant system. Accordingly, the main aim of the work is to provide some indications for the numerical modelling and to evaluate the accuracy level of predicted heat transfer when commercial or research packages are employed along with different grid resolution levels. Overall results are in good agreement with experimental data even if some local discrepancies are present.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.