The use of purge flow in gas turbines allows for high turbine entry temperatures, which are essential to produce high cycle efficiency. Purge air is bled from the compressor and reintroduced in the turbine to cool vulnerable components. Wheel-spaces are formed between adjacent rotating and stationary discs, with purge air supplied at low radius before exiting into the mainstream gas-path through a rim-seal at the disc periphery. An aerodynamic penalty is incurred as the purge flow egress interacts with the mainstream. This study presents unparalleled three-dimensional velocity data from a single-stage turbine test rig, specifically designed to investigate egress-mainstream interaction using optical measurement techniques. Volumetric Velocimetry is applied to the rotating environment with phase-locked measurements used to identify and track the vortical secondary flow features through the blade passage. A baseline case without purge flow is compared to experiments with a 1.7% purge mass fraction; the latter was chosen to ensure a fully sealed wheel-space. A non-localised vortex tracking function is applied to the data to identify the position of the core centroids. The strength of the secondary-flow vortices was determined by using a circulation criterion on rotated planes aligned to the vortex filaments. The pressure-side leg of the horseshoe vortex and a second vortex associated with the egress flow were identified by the experimental campaign. In the absence of purge flow the two vortices merged, forming the passage vortex. With the addition of purge flow, the two cores remained independent to 40% of the blade axial chord, while also demonstrating an increased radial migration and intensification of the passage vortex. The egress core was shown to remain closer to the suction-surface with purge flow. Importantly, where the vortex filaments demonstrated strong radial or tangential components of velocity, the circulation level calculated from axial planes underpredicted the true circulation by up to 50%.

Typical turbomachinery aerothermal problems of practical interest are characterised by flow structures of wide-ranging scales, which interact with each other. Such multi-scale interactions can be observed between the flow structures produced by surface roughness and by the bulk flow patterns. Moreover, additive manufacturing may sooner or later open a new chapter in component designs by granting designers the ability to control the surface roughness patterns. As a result, surface finish, which so far has been treated largely as a stochastic trait, can be shifted to a set of design parameters that consist of repetitive, discrete micro-elements on a wall surface (‘manufacturable roughness’). Considering this prospective capability requirement, the question would arise regarding how surface micro-structures can be incorporated in computational analyses during a design phase in the future. Semi-empirical methods for predicting aerothermal characteristics and the impact of manufacturable roughness could be used to minimise computational cost. However, the lack of element-to-element resolution may lead to erroneous predictions, as the interactions among the roughness micro-elements have been shown to be significant for adequate performance predictions [1]. In this paper a new multi-scale approach based on the novel Block Spectral method is adopted. This method aims to provide efficient resolution of the detailed local flow variation in space and time of the large scale micro-structures. This resolution is provided without resorting to modelling every single ones in detail, as a conventional large scale CFD simulation would demand, but still demonstrating similar time-accurate and time-averaged flow properties. The main emphasis of the present work is to develop a parallelised solver of the method to enable tackling large problems. The work also includes a first of the kind verification and demonstration of the method for wall surfaces with a large number of micro-structured elements.

The University of Oxford has been working for a number of years on a conceptually new philosophy for HP vane design [1]. The desired function of the HP vane is to accept flow from a — generally — annular intake, and turn and accelerate it with minimum loss and heat transfer, and with due consideration to mechanical, manufacturing and integration issues. The conventional design process is well-defined: annulus line definition; through-flow analysis; 2D airfoil section definition; vane stack definition; lean/sweep treatments; secondary flow contouring and fillet design; vane and platform cooling system design; integrated aero/thermal/capacity optimization. The process is based on optimizing 2D airfoil stacks to control secondary flows. More recent research (last 20 years) has led to better understanding of airfoil lean, airfoil sweep, leading edge treatments and end-wall profiling. There may be some iterative optimization in which the airfoil sections themselves are modified, but these treatments (as implied by the terminology) are essentially modifications to a 2D airfoil stack which interrupts the upstream annular (most aero-engines) or can-partitioned (most land-based gas turbines) duct. It is proposed that a more natural design starting point for admitting and accelerating flow with minimum loss (etc.) is a trumpet-shaped flow-path which gradually turns to the desired angle. This approach rejects many traditional concepts in aerodynamics (the sacrosanct nature of the airfoil, concepts of good and bad lift distributions, concepts of vane and platform, etc.), and focuses our attention on the design of a highly-lofted flow-path. The purpose of this paper is to show that there may be merit in reappraising the starting point for HP vane design, and in investigating Fully Lofted Oval vane (FLOvane) forms of the type described. The corresponding author has not yet heard a good argument against the proposed design philosophy. Despite significantly diverging from conventional wisdom in terms of both process and final geometric form, the following advantages are claimed: improved aerodynamic loss characteristics, reduced heat load, improved cooling performance, improved thermal-mechanical life, improved stage/engine efficiency. The claimed advantages are justified with detailed numerical analyses of two current industrial gas turbines, compared back-to-back with modified turbines incorporating unoptimised FLOvane designs. Significant performance improvements were observed when FLOvane designs were used.

Rim seals are fitted in gas turbines at the periphery of the wheel-space formed between rotor discs and their adjacent casings. These seals, also called platform overlap seals, reduce the ingress of hot gases which can limit the life of highly-stressed components in the engine. This paper describes the development of a new, patented rim-seal concept showing improved performance relative to a reference engine design, using URANS computations of a turbine stage at engine conditions. The CFD study was limited to a small number of purge-flow rates due to computational time and cost, and the computations were validated experimentally at a lower rotational Reynolds number and in conditions under incompressible flow. The new rim seal features a stator-side angel wing and two buffer cavities between outer and inner seals: the angel-wing promotes a counter-rotating vortex to reduce the effect of the ingress on the stator; the two buffer cavities are shown to attenuate the circumferential pressure asymmetries of the fluid ingested from the mainstream annulus. Rotor disc pumping is exploited to reduce the sealing flow rate required to prevent ingress, with the rotor boundary layer also providing protective cooling. Measurements of gas concentration and swirl ratio, determined from static and total pressure, were used to assess the performance of the new seal concept relative to a bench-mark generic seal. The radial variation of concentration through the seal was measured in the experiments and these data captured the improvements due to the intermediate buffer cavities predicted by the CFD. This successful design approach is a potent combination of insight provided by computation, and the flexibility and expedience provided by experiment.

This paper deals with a novel configuration of axial compressor cascade with chordwise split. Contrast to tandem cascade and cascade with a slot from pressure surface to suction surface, the cascade with a chordwise split not only controls the separation of boundary layer, increases flow turning angle and stage load, thus the compressor length can be shortened or pressure ratio increased, but also the flow pattern in the split is different. The research work done shows that the cascade with chordwise split avoids or delays the separation of boundary layer on blade surface effectively thus may lead to a larger camber angle. The theoretical model and program for locating the split position and predicting flow field in cascade passage by using Finite Element Method are given. Experiments have verified the mathematical model and the preferable effects of the chordwise split. The results of calculation correspond to the test data satisfactorily.

An approach to the estimation of losses produced by the secondary flow near the hub and casing of turbine cascade with leaned blades is presented after discussing the generation of secondary flow and the effect of some important factors on losses. The loss system is a development of the method in [9] and [10], but results obtained by comparing with experimental data are encouraging.

An investigation into the pressure distribution in a turbine nozzle cascade with and without leaned blades has been made using a 3D subsonic variational FEM. We can conclude from our studies that the gas flow circulation is the essential factor of redistributing the secondary flow loss along the blade height in a leaned cascade. In addition, this paper recommends a direction for further research into the leaned stator cascade.

The commercial computational fluid dynamics code ANSYS CFX 12.1 has been employed to carry out Unsteady Reynolds Averaged Navier Stokes (URANS) computations to investigate the fluid mechanics of two different rim-seal geometries in a 3D model of a turbine stage. The mainstream annulus, seal and wheel-space geometries are based on an experimental test rig used at the University of Bath. The calculated peak-to-trough pressure difference in the annulus, which is the main driving mechanism for ingestion, is in good agreement with experimental measurements. There is also good agreement between the computed and measured swirl ratios in the wheel-space. Computed values of concentration-based sealing effectiveness are obtained over a range of sealing flow rates for both an axial-clearance and a radial clearance rim-seal. Good agreement with gas concentration measurements is found for the axial-clearance seal over a certain range of sealing flow rates. Some under-prediction of the amount of ingestion for the radial-clearance seal is obtained. The computed mainstream pressure coefficient increases progressively with mainstream Mach number in moving from quasi-incompressible experimental rig conditions to the compressible flow conditions encountered in engines. It is shown that the minimum sealing flow rate required to prevent ingestion increases as mainstream Mach number increases. A scaling method is proposed to allow sealing flow rates to prevent ingestion obtained from low Mach number experiments to be extrapolated to engine-representative conditions.