<p>The fundamental design of axial compressors has matured to an exceptional level of performance due to a century of research. With the improvements in efficiency becoming increasingly difficult, attention continues to be channeled towards understanding and reducing secondary losses such as hub or tip clearance leakages, seal leakages, etc. Studies detailing the impact of seal leakages are relatively scarce due to difficulties of obtaining data in the complex rotating geometries of a high-speed compressor cavity. While the impact of seal leakages on primary passage is readily available, details inside the cavity geometry is scarce in open literature because majority of the investigations have been performed on linear cascades with slots machined as cavities or standalone labyrinth seals that fail to provide a wholesome understanding of the leakage flow and windage heating in the rotating geometries.<br></p> <p> Therefore, the principal objective of this work is to investigate flow physics in the stator cavity wells for understanding the flow path of the leakage fluid and windage heating within the cavity. A parametric model of the Purdue 3-Stage Compressor (P3S) is used to allow for rapid geometric modifications to the seal clearances in a coupled stator-cavity system. The investigations presented here consist of a series of numerical simulations using ANSYS CFX as the primary Computational Fluid Dynamics (CFD) tool. Measurements performed by previous investigators are utilized to define the boundary conditions of this model. This study’s goal is to characterize the interdependence of parameters such as cavity leakage flow rate, circumferential velocity, and windage heating for understanding the flow structure inside the cavity wells and their impact on cavity temperatures. Data acquired is intended to reveal mechanisms through which cavity leakage flows affect the stator passage aerodynamics and the windage heating, both regarding their effect on the compressor performance and the details of the flow path within the cavity. Consequently, this will provide insight into how the complex cavity leakage flow influences the design considerations for optimizing stator passage aerodynamics and minimizing stator cavity heating.</p> <p>The compressor operating conditions of Nominal Loading (NL) is the focus of this CFD work since the flow field at High Loading (HL) has significant boundary layer separation. NL is closest to both the design and peak efficiency conditions where the compressor would spend the majority of its time in operation, understanding cavity flow physics at this operating condition would have a direct impact on enhancing the overall compressor performance. A CFD model of the standalone primary passage is developed first using the dataset available from experiments performed by previous investigators for establishing confidence in the primary passage flow physics. Therefore, detailed total pressure, total temperature, velocity, and flow angle data collected behind each blade row is utilized for validating the primary passage flow in the CFD model. After validating the primary passage model, measurements in the coupled cavity model are acquired to understand the flow variations as well as temperature development in the cavity due to the varying labyrinth seal clearance.</p> <p>The investigations in this work are divided into two distinct branches. First, to aid the aerodynamic research community, the flow structure inside the cavity wells is investigated to understand the impact cavity leakage flow has on the compressor efficiency and on its interactions with the primary flow path. Secondly, for understanding the development and rise of temperature in the cavity wells, i.e., the windage effect, are performed to aid the thermo-mechanical research community so that the material choices and stress analysis of the cavity components can be optimized. Hence, the trends in the data acquired provide the aerodynamic, mechanical, and secondary flow system designers an indication of the complexities of the flow within shrouded stator cavities and provide insight into designing and optimizing more complex geometries.</p><p>Results from this investigation describe how increasing seal clearance deteriorates the stator performance and enables the cross-passage migration of low momentum fluid to worsen hub corner separation. The simulations also state the case for re-ingestion at tight seal clearances as the 3D streamlines show heated efflux emerges from the upstream cavity interface, dwells near the hub, and gets recirculated back into the cavity inlet well. Radial variations inside the cavity wells show high cavity temperatures with excessive cavity due to re-ingestion, while the cases that avoid re-ingestion are observed at the lowest temperatures. These radial variations also identify the cavity leakage flow path and the development of circumferential velocity. Lastly, the total pressure loss, total temperature rise and windage heating, all show a strong dependence on circumferential velocity development, which is inherently dependent on the labyrinth seal clearances.<br></p>

The ability to quantify leakage flow and windage heating for labyrinth seals with honeycomb lands is critical in understanding gas turbine engine system performance and predicting its component lifes. Variety of labyrinth seal configurations (number of teeth, stepped or straight, honeycomb cell size) are in use in gas turbines, and for each configuration, there are many additional geometric factors that can impact a seal’s leakage and windage characteristics. To achieve high performance in modern gas turbine engines, the labyrinth seals operate at low clearances and high rotational speed which are generally deployed with honeycomb lands on the stator. During the transient operation of aircraft engines, the stator and rotor mechanical and thermal growths differ from one another and can often result in the rotor’s incursion into the stator honeycomb structure. The incursions create rub-grooves in the honeycomb lands that can subsequently enlarge as the engine undergoes various manoeuvres. However, the effects of different honeycomb cell size, rotation and presence of rub-groove have not been thoroughly investigated in previously published work. The objective of the present research is to numerically investigate the influence of the above three factors on seal leakage and windage heating. The present work focuses the development of a numerical methodology aimed at studying above effects. Specifically, a three-dimensional CFD model is developed utilizing commercial finite volume-based software incorporating the RNG k-ε turbulence model. Detail validation of the numerical model is performed by comparing the leakage and windage heating measurements of several rig tests. The turbulent Schmidt number is found to be an important parameter governing the leakage prediction. It depends on honeycomb cell size and clearance for honeycomb seals, and Reynolds number in the presence smooth lands. The present numerical model with the modified RNG k- turbulence model predicts seal leakage and windage heating within 3-10% with available experimental data. Using the validated numerical model, a broad parametric study is conducted by varying honeycomb cell size, radial clearance, pressure ratio and rotational speed for a four-tooth straight-through labyrinth seal with and without rub-grooves. They further indicate that presence of rub-grooves increases seal leakage and reduce windage heating, specifically at smaller clearance and for larger honeycomb cell size. Rotation significantly reduces leakage with smooth stator land and smaller honeycomb cells whereas the effect is minimal for larger (3.2mm) honeycomb cells. However, at very high rotational speed seal flow reduces in all seal configurations due to high temperature rise and Rayleigh line effects. At no rub condition and lower clearance, the larger honeycomb cells leak more flow due to high bypass flow through the honeycomb cells. This results into lower pocket swirl and higher windage. When the seal clearance increases the larger honeycomb cells offers more drag to the seal flow, therefore they leak less. At higher clearances the flow travels like a strong wall jet and isolates the pocket air from honeycomb cells. Hence, at open clearances labyrinth seals with any honeycomb cell size essentially produce the same pocket swirl and windage heating.