No abstract

The ingress of hot gas through the rim seal of a gas turbine depends on the pressure difference between the mainstream flow in the turbine annulus and that in the wheel-space radially inward of the seal. This paper describes experimental measurements which quantify the effect of ingress on both the stator and rotor discs in a wheel-space pressurised by sealing flow. Infrared sensors were developed and calibrated to accurately measure the temperature history of the rotating disc surface during a transient experiment, leading to an adiabatic effectiveness. The performance of four generic (though engine-representative) single- and double-clearance seals was assessed in terms of the variation of adiabatic effectiveness with sealing flow rate. The measurements identify a so-called thermal buffering effect, where the boundary layer on the rotor protects the disc from the effects of ingress. It was shown that the effectiveness on the rotor was significantly higher than the equivalent stator effectiveness for all rim seals tested. Although the ingress through the rim seal is a consequence of an unsteady, three-dimensional flow field, and the cause-effect relationship between pressure and the sealing effectiveness is complex, the time-averaged experimental data is shown to be successfully predicted by relatively simple semi-empirical models, which are described in a separate paper. Of particular interest to the designer, significant ingress can enter the wheel-space before its effect is sensed by the rotor.

Ingress is the penetration of a hot mainstream gas in a turbine annulus through the rim seal into the wheel-space between the rotating turbine disc (the rotor) and the adjacent stationary casing (the stator). Purge flow is used to prevent or reduce ingress, and the sealing effectiveness relates the flow rates of the purge and ingress. In this paper, an adiabatic effectiveness is used to relate the temperatures of a thermally-insulated rotor, the purge flow and the ingress. A non-dimensional buffer parameter, Ψ, is used to relate the sealing effectiveness on the stator and the adiabatic effectiveness on the rotor, respectively. This paper reports the first experimental study of the effect of ingress and purge flow on the adiabatic temperatures of both upstream and downstream surfaces of the rotor. Measurements of concentration and swirl over a range of purge have been obtained in wheel-spaces upstream and downstream of the rotor in a turbine rig. In transient heating tests, fast-response thermocouples were used to measure the temperature of the air in the wheel-space core; simultaneously, the temperatures of the upstream and downstream rotor surfaces were determined from infra-red sensors. The extrapolated steady-state temperatures (obtained using a maximum-likelihood estimation analysis) were used to determine the adiabatic effectiveness as a function of purge flow rate. The buffer effect of the purge flow for both wheel-spaces was quantified via comparisons between the variation of Ψ with purge flow rate. It was shown that the sealing effectiveness for the downstream wheel-space was larger than for the upstream. Consequently, and consistent with the theoretical model, the buffering effect of the purge flow was shown to be smaller downstream.

Use of Pressure M easu rem en ts to D eterm in e Effectiveness of Turbine R im S ealsThe ingress o f hot gas through the rim seal o f a gas turbine depends on the pressure dif ference between the mainstream flow in the turbine annulus and that in the wheel-space radially inward o f the rim seal. In this paper, a previously published orifice model is modified so that the sealing effectiveness ec determined from concentration measurements in a rig could be used to determine ep, the effectiveness determined from pressure meas urements in an engine. It is assumed that there is a hypothetical "sweet spot" on the vane platform where the measured pressures would ensure that the calculated value o f ep equals ec, the value determined from concentration measurements. Experimental meas urements fo r a radial-clearance seal show that, as predicted, the hypothetical pressure difference at the sweet spot is linearly related to the pressure difference measured at an arbitrary location on the vane platform. There is good agreement between the values o f ep determined using the theoretical model and values o fec determined from concentration measurements. Supporting computations, using a 3D steady computational fluid dynamics (CFD) code, show that the axial location o f the sweet spot is very close to the upstream edge o f the seal clearance. It is shown how parameters obtained from measurements o f pressure and concentration in a rig could, in principle, be used to calculate the sealing effectiveness in an engine.

The rise in temperature and pressure ratios for both turbine and compressor intrinsically require an increased secondary air system performance. The more demanding operating environment of the components to be cooled and the quest for improved efficiency, command a particular attention to the losses faced along the secondary flow path. One of the major pressure drop is experienced with the radial inflow in a rotating cavity, becoming a potential high gain branch of the system. The use of vortex reducers has shown a considerable improvement in performance and this paper presents the work done to minimize the pressure losses for an industrial gas turbine application using various vortex reducers. The numerical results and the set-up of a one-dimensional network model and Computational Fluid Dynamics (CFD) model for various configurations are discussed in detail showing an improved system performance with the proposed vortex reducer configuration.

Pre-swirl system is installed to minimize energy loss between the stationary and rotating parts of turbine secondary air system. Although various optimization studies were conducted to increase the pre-swirl efficiency, most of the studies were focused on a hole type pre-swirl nozzles. In this study, a vane type pre-swirl nozzle was optimized to increase mass flow rate and temperature drop for given boundary conditions. The system performance was analyzed by 3D CFD and the objective functions were used to maximize the discharge coefficient and the adiabatic effectiveness. After sensitivity analysis, seven design variables were chosen on the three planes of nozzle span-wise direction. The OLHD method was used to obtain the initial scattered test points, and the additional ones were supplied by the ALHD method. The Kriging model was constructed as the surrogate one, and refined iteratively until satisfying the convergence criteria between the estimated point and the CFD result. The optimized model improved the span-wise uniformity of the flow path and the discharge coefficient was increased by 2.57%, whilst the adiabatic effectiveness remained nearly constant. The performance was also analyzed for pressure ratio and off-design points, with the optimized model showing better performance at all boundary conditions.

No abstract

The ingress of hot gas through the rim seal of a gas turbine depends on the pressure difference between the mainstream flow in the turbine annulus and that in the wheel-space radially inward of the rim seal. In this paper, a previously published orifice model is modified so that the sealing effectiveness εc determined from concentration measurements in a rig could be used to determine εp the effectiveness determined from pressure measurements in an engine. It is assumed that there is a hypothetical ‘sweet spot’ on the vane platform where the measured pressures would ensure that the calculated value of εp equals εc, the value determined from concentration measurements. Experimental measurements for a radial-clearance seal show that, as predicted, the hypothetical pressure difference at the sweet spot is linearly related to the pressure difference measured at an arbitrary location on the vane platform. There is good agreement between the values of εp determined using the theoretical model and values of εc determined from concentration measurements. Supporting computations, using a 3D steady CFD code, show that the axial location of the sweet spot is very close to the upstream edge of the seal clearance. It is shown how parameters obtained from measurements of pressure and concentration in a rig could, in principle, be used to calculate the sealing effectiveness in an engine.