No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Numerical simulation of rim seal flows in axial turbines
Abstract
In a gas turbine, ingestion of hot gas into the high-pressure turbine disc cavities could cause metal overheat. To prevent this, cool air is taken from the compressor and ejected through the cavities. However, this sealing flow also reduces the overall efficiency, and a compromise has to be found between the level of ingestion tolerated and the losses. Recent advances made in applying Computational Fluid Dynamics to such configurations are presented, with the aim of better understanding the physical phenomena and providing reliable design tools. First, results showing the pumping effect of the rotating disc are presented, including the influence of flow instabilities observed in both computational and experimental results. Second, the influence of the main annulus pressure asymmetries are analysed on a simplified representation of an available experiment, showing the combined influence of asymmetries generated by vanes and struts. Finally, a rim seal geometry representative of aero-engine design is studied in comparison to experiment, exhibiting the coupled influence of the cavity instabilities and annulus asymmetries.
... The role of inherently unsteady rotating flow modes is an area of particular interest in current research. Figure 1, reproduced from reference [1], indicates where such flow structures have been identified in experimental and computational studies, for a wide range of seal types [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18]. The figure shows the frequency of the unsteadiness in a stationary frame of reference normalized by the rotor speed. ...
... All the simulations presented here are time-dependent and use a URANS approach employing the Spalart-Allmaras turbulence model [26]. For species concentration analyses, a passive scalar equation has been solved following the approach of Boudet et al. [3]. ...
... Perhaps the tendency of the circumferential pressure asymmetry to drive ingestion is balanced by the reduced tendency of the swirling annulus flow to migrate radially inwards. The present results are broadly consistent with the URANS study of Boudet et al. [3] who reported similar levels of predicted ingestion for a chute seal with an axisymmetric swirling annulus flow to that from a full turbine stage calculation. ...
A systematic study of sealing performance for a chute style turbine rim seal using URANS methods is reported. This extends previous studies from a configuration without external flow in the main annulus to cases with a circumferentially uniform axial flow and vane generated swirling annulus flow (but without rotor blades). The study includes variation of the mean seal-to-rotor velocity ratio, main annulus-to-rotor velocity ratio, and seal clearance. The effects on the unsteady flow structures and the degree of main annulus flow ingestion into the rim seal cavity are examined. Sealing effectiveness is quantified by modeling a passive scalar, and the timescales for the convergence of this solution are considered. It has been found that intrinsic flow unsteadiness occurs in most cases, with the presence of vanes and external flow modifying, the associated flow structures and frequencies. Some sensitivities to the annulus flow conditions are identified. The circumferential pressure asymmetry generated by the vanes has a clear influence on the flow structure but does not lead to higher ingestion rates than the other conditions studied.
... Boudet et al. [8,9] investigated ingress using unsteady CFD configurations with and without blades. Using a small 13.33 deg sector they computed rim-seal instabilities at 0.44 of the BPF, showing these rotating structures disappeared at higher sealing flow rates. ...
... Large eddy simulations of ingress were performed by O'Mahoney et al. [15,16] using the same geometry as earlier studies by Boudet et al. [8,9]. The work found that increasing the sector size from 13.3 deg to 40 deg had little effect on the average flow fields, while LES showed an improved prediction of ingress relative to URANS. ...
... The CFD model identified 8-22 pressure structures rotating at 82-93% of the disk speed and like previous studies (e.g., Refs. [7,8,10,14]), this lowfrequency unsteadiness was found to disappear at the highest purge rates. Schadler et al. also identified that the rim-seal unsteadiness could give rise to nonnegligible noise-emission within the human perception of sound. ...
In high-pressure turbines, cool air is purged through rim seals at the periphery of wheel-spaces between the stator and rotor disks. The purge suppresses the ingress of hot gas from the annulus but superfluous use is inefficient. In this paper, the interaction between the ingress, purge, and mainstream flow is studied through comparisons of newly acquired experimental results alongside unsteady numerical simulations based on the DLR TRACE solver. New experimental measurements were taken from a one-and-a-half stage axial-turbine rig operating with engine-representative blade and vane geometries, and overlapping rim seals. Radial traverses using a miniature CO2 concentration probe quantified the penetration of ingress into the rim seal and the outer portion of the wheel-space. Unsteady pressure measurements from circumferentially positioned transducers on the stator disk identified distinct frequencies in the wheel-space, and the computations reveal these are associated with large-scale flow structures near the outer periphery rotating at just less than the disk speed. It is hypothesized that the physical origin of such phenomenon is driven by Kelvin-Helmholtz instabilities caused by the tangential shear between the annulus and egress flows, as also postulated by previous authors. The presence and intensity of these rotating structures are strongly dependent on the purge flow rate. While there is general qualitative agreement between experiment and computation, it is speculated that the underprediction by the computations of the measured levels of ingress is caused by deficiencies in the turbulence modeling.
... Boudet et al. [8,9] investigated ingress using unsteady CFD configurations with and without blades. Using a small 13.33º sector they computed rim-seal instabilities at 0.44 of the blade passing frequency (BPF), showing these rotating structures disappeared at higher sealing flow rates. ...
... Large Eddy Simulations (LES) of ingress were performed by O'Mahoney et al. [15,16] using the same geometry as earlier studies by Boudet et al. [8,9]. The work found that increasing the sector size from 13.3º to 40º had little effect on the average flow fields, while LES showed an improved prediction of ingress relative to URANS. ...
... A combined experimental and computational study focussing on the influence of the unsteady rim-seal/cavity flow on the annulus flow was presented by Schadler et al. [25]. The CFD model identified 8-22 pressure structures rotating at 82-93% of the disc speed and like previous studies [e.g., 7,8,10,14] this low-frequency unsteadiness was found to disappear at the highest purge rates. Schadler et al. also identified that the rim-seal unsteadiness could give rise to non-negligible noise-emission within the human perception of sound. ...
In high-pressure turbines, cool air is purged through rim seals at the periphery of wheel-spaces between the stator and rotor discs. The purge suppresses the ingress of hot gas from the annulus but superfluous use is inefficient. In this paper the interaction between the ingress, purge and mainstream flow is studied using unsteady numerical simulations based on the DLR TRACE solver. The computations are compared to experimental measurements from a one-and-a-half stage axial-turbine rig operating with engine-representative blade and vane geometries, and overlapping rim seals. Radial traverses using a miniature CO2 concentration probe quantified the penetration of ingress into the rim seal and the outer portion of the wheel-space. Unsteady pressure measurements from circumferentially-positioned transducers on the stator disc identified distinct frequencies in the wheelspace, and the computations reveal these are associated with large-scale flow structures near the outer periphery rotating at just less than the disc speed. It is hypothesised that the physical origin of such phenomenon is driven by Kelvin-Helmholtz instabilities caused by the tangential shear between the annulus and egress flows, as also postulated by previous authors. The presence and intensity of these rotating structures are strongly dependent on the purge flow rate. While there is general qualitative agreement between experiment and computation, it is speculated that the underprediction by the computations of the measured levels of ingress is caused by deficiencies in the turbulence modelling.
... Counter to this argument is the observation in, for example, Cao et al. 11 and Jakoby et al. 24 that rotating cavity, seal and annulus flows may be subject to 3D rotating flow modes that significantly affect ingestion. For more extended seals involving overlapping rotor and stator parts, the rotation and surface drag may also affect the flow in the seals as, for example, suggested by Graber et al. and supported by computations for a chute rim seal in Boudet et al. 25 With similar functional dependence of ingestion on seal flow rate observed across a wide range of experimental conditions it is difficult to distinguish distinct regimes or to separate the effects of individual parameters in these complex flows. ...
... Such comparisons have shown mixed levels of agreement, as shown in the literature. 11,19,20,[24][25][26][27][28][29][30][31] While conventional Reynolds-averaged Navier-Stokes (RANS) models sometimes give quantitative or qualitative trends in reasonable agreement with measurements, considerable uncertainties remain. ...
... In comparison with the 2D axisymmetric RANS model, the 3D URANS model that predicted rotating flow structures in the rim seal achieved better agreement with an empirically based sealing effectiveness correlation for rotationally driven ingestion. 12 These results were later published in Boudet et al. 25 The rotating flow structure illustrated had 56 lobes rotating slowly and giving a distinct frequency of $1.5 for pressure fluctuations. Figure 10 shows pressure contours on the rotor surface with alternating low and HP regions appearing in the seal on the rotor lip. ...
This paper presents a review of research on turbine rim sealing with emphasis placed on the underlying flow physics and modelling capability. Rim seal flows play a crucial role in controlling engine disc temperatures but represent a loss from the main engine power cycle and are associated with spoiling losses in the turbine. Elementary models that rely on empirical validation and are currently used in design do not account for some of the known flow mechanisms, and prediction of sealing performance with computational fluid dynamics has proved challenging. Computational fluid dynamics and experimental studies have indicated important unsteady flow effects that explain some of the differences identified in comparing predicted and measure sealing effectiveness. This review reveals some consistency of investigations across a range of configurations, with inertial waves in the rotating flow apparently interacting with other flow mechanisms which include vane, blade and seal flow interactions; disc pumping and cavity flows; shear layer and other instabilities; and turbulent mixing.
... Prior to this, in 2002, Autef (2002) reported URANS solutions showing unsteady flow structures of a rim seal configuration without vanes and blades, as described by Chew et al. (2003). In 2004Jakoby et al. (2004 Boudet et al. (2005) reported URANS solutions for a chute seal geometry. These revealed that the flows in this configuration were also inherently 3-D and unsteady. ...
... This showed the significance of unsteady flow effects but does not represent the effects of the low frequency unsteadiness discussed above. The unsteady RANS model of Boudet et al. (2005), which shows inherent unsteady flow features of the rim seal flow, achieves significant improvement in agreement with measured sealing effectiveness to the steady RANS. Thus, it can be conjectured that correct modelling of unsteady rim sealing flow structures is essential for the accurate prediction of sealing effectiveness, and that current design methods do not capture some important flow physics. ...
... This suggests the presence of a Taylor-Couette vortex in the mean flow. A possible Taylor-Couette flow mechanism in the rim sealing flow was also suggested by Boudet et al. (2005). ...
Unsteady flow phenomena unrelated to the main gas-path blading have been identified in a number of turbine rim seal investigations. This unsteadiness has significant influence on the sealing effectiveness predicted by the conventional steady RANS (Reynolds-averaged Navier–Stokes) method, thus it is important for turbine stage design and optimisation. This paper presents CFD (computational fluid dynamics) modelling of a chute type rim seal that has been previously experimentally investigated. The study focuses on inherent large-scale unsteadiness rather than that imposed by vanes and blades or external flow. A large-eddy simulation (LES) solver is validated for a pipe flow test case and then applied to the chute rim seal rotor/stator cavity. LES, RANS and unsteady RANS (URANS) models all showed reasonable agreement with steady measurements within the disc cavity, but only the LES shows unsteadiness at a similar distinct peak frequency to that found in the experiment, at 23 times the rotational frequency. The boundary layer profile within the chute rim seal clearance has been scrutinised, which may explain the improvement of LES over RANS predictions for the pressure drop across the seal. LES results show a clockwise mean flow vortex. A more detailed sketch of the rim sealing flow unsteady flow structures is established with the help of the LES results. However, there are some significant differences between unsteadiness predicted and the measurements, and possible causes of these are discussed.
... Many studies have reported that CFD models of rim seal flows captured unsteady flow structures with dominant frequencies distinct from those due to the blade rotation, see for example Refs. [9,11,[15][16][17][18]. Boudet et al. [15], Jakoby et al. [16] and Julien et al. [19] reported that better prediction of sealing effectiveness is achievable by capturing these unsteady flow features using unsteady RANS (URANS). ...
... [9,11,[15][16][17][18]. Boudet et al. [15], Jakoby et al. [16] and Julien et al. [19] reported that better prediction of sealing effectiveness is achievable by capturing these unsteady flow features using unsteady RANS (URANS). Further, O'Mahoney et al. investigated a turbine stage with chute rim seal from Gentilhomme [20] using LES, and achieved closer agreement with measurements of sealing effectiveness, compared to URANS. ...
... Possible mechanisms causing the distinct unsteady flow structures have been discussed by several authors. For example, Boudet et al. [15] associated the unsteady flow phenomenon with Taylor-Couette flow instability. Chilla et al. [17] attributed the unsteady flow patterns to Kelvin-Helmholtz vortex shedding related to the interaction between the main annulus flow and rim seal flow. ...
This paper reports large-eddy simulations (LES) and unsteady Reynolds-averaged Navier-Stokes (URANS) calculations of a turbine rim seal configuration previously investigated experimentally. The configuration does not include any vanes, blades or external flows, but investigates inherent unsteady flow features and limitations of CFD modelling identified in engine representative studies. Compared to RANS and URANS CFD models, a sector LES model showed closer agreement with mean pressure measurements. LES models also showed agreement with measured pressure frequency spectra, but discrepancies were found between the LES and experiment in the speed and the circumferential lobe number of the unsteady flow structures. Sensitivity of predictions to modelling assumptions and differences with experimental data are investigated through CFD calculations considering sector size, interaction between the rim cavity and the inner cavity, outer annulus boundary conditions, and the coolant mass flow. Significant sensitivity to external flow conditions, which could contribute to differences with measurements, is shown, although some discrepancies remain. Further detailed analysis of the CFD solutions is given illustrating the complex flow physics. Possible improvement of a steady RANS model using a priori analysis of LES was investigated, but showed a rather small improvement in mean pressure prediction.
... With a gap between the rotor and stator at the outer radius, some or all of the flow in the disk boundary layer will leave the cavity and will be replaced by fluid drawn in through the gap from the surrounding environment. As reported by Chew et al. [1] and Boudet et al. [2], a 3D unsteady Reynolds-averaged Navier-Stokes (RANS) CFD model gave good agreement with an empirical correlation for sealing effectiveness in a rotor/stator disk cavity with a simple axial clearance seal at the periphery. The CFD solution was inherently unsteady and 3D, despite the geometry being axisymmetric and the boundary conditions being steady and axisymmetric. ...
... The rim seal gap in the chute seal is bounded by an inner conical rotor surface and an outer conical stator surface. Boudet et al. [2,4] compared the RANS CFD solutions with the measurements by Gentilhomme et al. [5] of sealing effectiveness and mean pressure from a turbine rig. The CFD solutions including both blades and vanes were obtained. ...
... In this paper, a relatively simple rotating flow configuration, related to earlier experimental and numerical research by Boudet et al. [2], Gentilhomme et al. [5], and O'Mahoney et al. [6], is investigated experimentally. Focusing on the inherent unsteadiness of rim seal flows, rather than unsteadiness imposed by rotating blades, the experiments consider a rotor/stator disk cavity without external flow. ...
While turbine rim sealing flows are an important aspect of turbomachinery design, affecting turbine aerodynamic performance and turbine disk temperatures, the present understanding and predictive capability for such flows is limited. The aim of the present study is to clarify the flow physics involved in rim sealing flows and to provide high-quality experimental data for use in evaluation of computational fluid dynamics (CFD) models. The seal considered is similar to a chute seal previously investigated by other workers, and the study focuses on the inherent unsteadiness of rim seal flows, rather than unsteadiness imposed by the rotating blades. Unsteady pressure measurements from radially and circumferentially distributed transducers are presented for flow in a rotor-stator disk cavity and the rim seal without imposed external flow. The test matrix covered ranges in rotational Reynolds number, Reø, and nondimensional flow rate, Cw, of 2.2-3.0 × 10⁶ and 0-3.5 × 10³, respectively. Distinct frequencies are identified in the cavity flow, and detailed analysis of the pressure data associates these with large-scale flow structures rotating about the axis. This confirms the occurrence of such structures as predicted in previously published CFD studies and provides new data for detailed assessment of CFD models.
... With a gap between the rotor and stator at the outer radius, some or all of the flow in the disc boundary layer will leave the cavity and will be replaced by fluid drawn in through the gap from the surrounding environment. As reported by Chew et al. [1] and Boudet et al. [2], a 3D unsteady Reynolds-averaged Navier-Stokes (RANS) CFD model gave good agreement with an empirical correlation for sealing effectiveness in a rotor/stator disc cavity with a simple axial clearance seal at the periphery. The CFD solution was inherently unsteady and 3D, despite the geometry being axisymmetric and the boundary conditions being steady and axisymmetric. ...
... The rim seal gap in the chute seal is bounded by an inner conical rotor surface and an outer conical stator surface. Boudet et al. [2], [4] compared RANS CFD solutions with measurements by Gentilhomme et al. [5] of sealing effectiveness and mean pressure from a turbine rig. CFD solutions including both blades and vanes were obtained. ...
... In this paper a relatively simple rotating flow configuration, related to earlier experimental and numerical research by Boudet et al. [2], Gentilhomme et al. [5] and O'Mahoney et al. [6], is investigated experimentally. Focusing on the inherent unsteadiness of rim seal flows, rather than unsteadiness imposed by rotating blades, the experiments consider a rotor/stator disc cavity without external flow. ...
While turbine rim sealing flows are an important aspect of turbomachinery design, affecting turbine aerodynamic performance and turbine disc temperatures, the present understanding and predictive capability for such flows is limited. The aim of the present study is to clarify the flow physics involved in rim sealing flows and to provide high quality experimental data for use in evaluation of CFD models. The seal considered is similar to a chute seal previously investigated by other workers, and the study focuses on the inherent unsteadiness of rim seal flows, rather than unsteadiness imposed by the rotating blades. Unsteady pressure measurements from radially and circumferentially distributed transducers are presented for flow in a rotor-stator disc cavity and the rim seal without imposed external flow. The test matrix covered ranges in rotational Reynolds number, Reø, and non-dimensional flow rate, Cw, of 2.2–3.0×10⁶ and 0–3.5×10³ respectively. Distinct frequencies are identified in the cavity flow and detailed analysis of the pressure data associates these with large scale flow structures rotating about the axis. This confirms the occurrence of such structures as predicted in previously published CFD studies and provides new data for detailed assessment of CFD models.
Copyright © 2016 by Rolls-Royce plc Country-Specific Mortality and Growth Failure in Infancy and Yound Children and Association With Material Stature
Use interactive graphics and maps to view and sort country-specific infant and early dhildhood mortality and growth failure data and their association with maternal
... However, the validity of CFD models is obviously of general interest, and CFD models examining both effects can be envisaged. Recent CFD and experimental studies [14,15,16,17,18] have shown that at low sealing flow rates with ingestion into the disc cavities occurring, unsteady flow features, unrelated to the blade passing frequency, can be important. In some cases [14,15], k-ε model calculations have given some agreement with experiment in the level of ingestion obtained. ...
... In some cases [14,15], k-ε model calculations have given some agreement with experiment in the level of ingestion obtained. However, Boudet et al. [16,17] note discrepancies between their Spalart-Allmaras model predictions and experiments for an aero-engine representative chute rim seal geometry. More work is clearly needed to clarify rim seal flows in such conditions. ...
... This test case is a partial model of a two stage turbine, as studied experimentally by Gentilhomme [35]. It has previously been studied numerically by Boudet et al. [16,17] for low coolant flow using the original Spalart-Allmaras model. ...
The main objectives of the paper are to test widely used turbulence models against selected benchmark problems identifying range of applicability and limitations of the models used and to help accurately predict flow in turbine blade passages and disc cavities. The following models are considered: the k–ε model with and without a Kato–Launder correction and a Richardson number correction, the two-layer k–ε /k–l model, and the Spalart–Allmaras model with and without correction for rotation. Weaknesses in the models are identified and suggestions made for possible improvements. Numerical implementation of the wall functions approach is also considered. The test cases considered include flat plate boundary layer, flat plate heat transfer, enclosed rotating disc, and a combined turbine blade/disc cavity model. Comparisons are made with experimental data and computations from different CFD codes.
... Counter to this argument is the observation in, for example, Cao et al. 11 and Jakoby et al. 24 that rotating cavity, seal and annulus flows may be subject to 3D rotating flow modes that significantly affect ingestion. For more extended seals involving overlapping rotor and stator parts, the rotation and surface drag may also affect the flow in the seals as, for example, suggested by Graber et al. and supported by computations for a chute rim seal in Boudet et al. 25 With similar func- tional dependence of ingestion on seal flow rate observed across a wide range of experimental condi- tions it is difficult to distinguish distinct regimes or to separate the effects of individual parameters in these complex flows. ...
... Such comparisons have shown mixed levels of agreement, as shown in the literature. 11,19,20,[24][25][26][27][28][29][30][31] While conventional Reynolds-averaged Navier- Stokes (RANS) models sometimes give quantitative or qualitative trends in reasonable agreement with measurements, considerable uncertainties remain. ...
... In com- parison with the 2D axisymmetric RANS model, the 3D URANS model that predicted rotating flow struc- tures in the rim seal achieved better agreement with an empirically based sealing effectiveness correlation for rotationally driven ingestion. 12 These results were later published in Boudet et al. 25 The rotating flow structure illustrated had 56 lobes rotating slowly and giving a distinct frequency of $1.5 for pressure fluc- tuations. Figure 10 shows pressure contours on the rotor surface with alternating low and HP regions appearing in the seal on the rotor lip. ...
In order to gain a better knowledge of the mechanisms of corner stall and to calibrate computational-fluid-dynamics (CFD) tools including both Reynolds-averaged Navier-stokes and large eddy simulation, a detailed and accurate experiment of three-dimensional flow field through a linear compressor cascade has been set up. Experimental data were acquired for a Reynolds number of 3.82 × 105 based on blade chord and inlet flow conditions. First, inlet flow conditions were surveyed by hot-wire anemometry in boundary layers. Second, in order to investigate the effects of incidence, measurements then were acquired at five incidences from −2° to 6°. The results included the outlet flow variables of the cascade, measured by a five-hole pressure probe, and static pressures on both blade and endwall surfaces, measured by pressure taps. Third, the flow field details were measured at an incidence angle of 4°. In this configuration the corner stall region was large enough to be investigated, and without two-dimensional (2D) separation at mid-span on the blade suction side near the trailing edge. The velocity field was then measured by 2D Particle Image Velocimetry in cross-sections parallel to the endwall. And the velocity field in the vicinity of the blade suction side was measured with 2D Laser Dropper Anemometry. In order to test the performance of CFD and also to validate the experimental results, a series of numerical simulations were carried out and compared with the experimental results. We thus obtained a set of detailed measurements which constitute an original and complete data base and in good agreement with the published experimental results in literature. These data were also compared with CFD results and showed that the improvements needed in turbulence modeling in order to accurately simulate the three-dimensional separation configuration of corner stall.
... Computational studies of ingress have also been performed [13][14][15][16][17][18]. Cao et al. [13] studied how sealing flow interacts with the hot gas flow in the annulus. ...
... Cao et al. [13] studied how sealing flow interacts with the hot gas flow in the annulus. Boudet et al. [14] studied the same configuration as Cao et al. [15] and observed nonlinear coupling among the flow features. Jakoby et al. [16] showed a large-scale structure forming in the wheelspace and how it strongly influences ingestion. ...
In gas turbines, the hot gas exiting the combustor can have temperatures as high as 2000 °C, and some of this hot gas enter into the space between the stator and rotor disks (wheelspace). Since the entering hot gas could damage the disks, its ingestion must be minimized. This is carried out by rim seals and by introducing a cooler flow from the compressor (sealing flow) into the wheelspace. Ingress and egress into rim seals are driven by the stator vanes, the rotor and its rotation, and the rotor blades. This study focuses on the ingress and egress driven by the rotor and its rotation. This is carried out by performing wall-resolved large eddy simulation (LES) around an axial seal in a rotor–stator configuration without vanes and blades. Results obtained show the mechanisms by which the rotor and its rotation induce ingress, egress, and flow trajectories. Kelvin–Helmholtz instability was found to create a wavy shear layer and displacement thickness that produces alternating regions of high and low pressures around the rotor side of the seal. Vortex shedding on the backward-facing side of the seal and its impingement on the rotor side of the seal also produces alternating regions of high and low pressures. The locations of the alternating regions of high and low pressures were found to be statistically stationary and to cause ingress to start on the rotor side of the seal. Vortex shedding and recirculating flow in the seal clearance also cause ingress by entrainment. With the effects of the rotor and its rotation on ingress and egress isolated, this study enables the effects of stator vanes and rotor blades to be assessed.
... In the simulation process, unsteady simulation using URANS method is carried out, and the unsteady simulation is initialized by a steady simulation result using RANS method. All the simulations are performed by a commercial solver ANSYS CFX v. 19. ...
... Structured mesh is generated using ANSYS ICEM CFD v. 19, and the total mesh number is 4 million. In the simulation, the Shear Stress Transport (SST) turbulent model is used. ...
Unsteady flow structures have been observed and reported in a number of recent rim-sealing investigations. These unsteady flow structures will influence the cavity pressure distribution, therefore influence the sealing efficiency. As a result, it is important to determine the mechanisms of these unsteady flow structures and how they influence the hot gas ingestion and sealing efficiency. A two-sector axial rim seal model is used to carry out the numerical investigation. The simulation is performed using the URANS method by the commercial CFD code ANSYS CFX, in which the SST turbulent model is applied. The mechanism and influence of the unsteady flow structures are analyzed. It was found that two different types of unsteadiness are observed inside the wheel space cavity: radial large flow structures dominated by the mainstream pressure distribution and inertia wave, and circumferential Kelvin-Helmholtz vortexes induced by circumferential velocity discontinuous distribution. The number and rotating speed of the radial and circumferential flow structures can be calculated using a cross-correlation method, and it was found that they can lead to a deeper ingress. By increasing the sealing flow rate, the pressure fluctuation inside the wheel space cavity is suppressed and the rotating speed of the flow structures is deaccelerated; thus, the sealing flow stabilizes the flow inside the wheel space cavity. Meanwhile, the K-H vortices’ position is lifted by the increased sealing flow rate, and the strength of the K-H vortices is suppressed, thus the sealing efficiency inside the wheel space cavity is also improved.
... To validate the passive scalar equation comparisons are made with available analytical and URANS solutions following Boudet [20]. ...
... Unsteady computations were carried out, using inviscid wall boundary conditions on the sides ( , , and ), and a periodic condition between and . An ILES was conducted using an explicit three-step Runge-Kutta scheme with CFL = 1, and results are compared with analytical solutions and URANS data obtained by Boudet [20]. ...
This paper presents WMLES simulations of a chute type turbine rim seal. Configurations with an axisymmetric annulus flow and with nozzle guide vanes fitted (but without rotor blades) are considered. The passive scalar concentration solution and WMLES are validated against available data in the literature for uniform convection and a rotor-stator cavity flow. The WMLES approach is shown to be effective, giving significant improvements over an eddy viscosity turbulence model, in prediction of rim seal effectiveness compared to research rig measurements. WMLES requires considerably less computational time than wall-resolved LES, and has the potential for extension to engine conditions. All WMLES solutions show rotating inertial waves in the chute seal. Good agreement between WMLES and measurements for sealing effectiveness in the configuration without vanes is found. For cases with vanes fitted the WMLES simulation shows less ingestion than the measurements, and possible reasons are discussed.
... To validate the passive scalar equation comparisons are made with available analytical and URANS solutions following Boudet [20]. ...
... Unsteady computations were carried out, using inviscid wall boundary conditions on the sides ( , , and ), and a periodic condition between and . An ILES was conducted using an explicit three-step Runge-Kutta scheme with CFL 1, and results are compared with analytical solutions and URANS data obtained by Boudet [20]. ...
This paper presents WMLES simulations of a chute type turbine rim seal. Configurations with an axisymmetric annulus flow and with nozzle guide vanes fitted (but without rotor blades) are considered. The passive scalar concentration solution and WMLES are validated against available data in the literature for uniform convection and a rotor-stator cavity flow. The WMLES approach is shown to be effective, giving significant improvements over an eddy viscosity turbulence model, in prediction of rim seal effectiveness compared to research rig measurements. WMLES requires considerably less computational time than wall-resolved LES, and has the potential for extension to engine conditions. All WMLES solutions show rotating inertial waves in the chute seal. Good agreement between WMLES and measurements for sealing effectiveness in the configuration without vanes is found. For cases with vanes fitted the WMLES simulation shows less ingestion than the measurements, and possible reasons are discussed.
... Despite a wide variety of rim seal geometries having been investigated, there are relatively few studies with chute seals as considered in this paper. Hence the results for chute seals presented by Horwood et al. [17], Gentilhomme et al. [18] and Boudet et al. [19] are of direct relevance to the current study. An earlier investigation by Phadke and Owen [4] included rotationallydriven ingestion measurements for a 'mitered' seal which may be regarded as a 'high angle chute' seal. ...
... However An extraordinary good agreement is evident despite the different test conditions. Sealing effectiveness measurements at the University of Sussex for a chute seal by Gentilhomme [18], as presented in references [19] and [22], also show some agreement with the present results. The Sussex experiments were conducted in a turbine rig with a flow coefficient ~0.6, and included a significant density difference between the purge and main annulus flows. ...
This experimental study considered the performance of a chute rim seal downstream of turbine inlet guide vanes (but without rotor blades). The experimental set up reproduced rotationally-driven ingestion without vanes and conditions of pressure-driven ingestion with vanes. The maximum rotor speed was 9000 rpm corresponding to a rotational Reynolds number of 3x106 with a flow coefficient of 0.48. Measurements of mean pressures in the annulus and the disc rim cavity as well as values of sealing effectiveness deduced from gas concentration data are presented. At high values of flow coefficient (low rotational speeds), the circumferential pressure variation generated by the vanes drove relatively high levels of ingestion into the disc rim cavity. For a given purge flow rate, increasing the disc rotation speed led to a reduction in ingestion shown by higher values of sealing effectiveness despite the presence of upstream vanes. At U_ax⁄((Ωb))="0.48" , the sealing effectiveness approached that associated with purely rotationally-driven ingestion. A map of sealing effectiveness against non-dimensional purge flow summarises the results and illustrates the combined rotational and pressure-driven effects on the ingestion mechanism. The results imply that flow coefficient is an important parameter in rim sealing and that rotational effects are important in many applications, especially turbines with low flow coefficient.
... In addition to these mechanisms many recent studies, reviewed in Ref. [11], report large-scale unsteady flow modes with characteristic frequencies unrelated to those of the rotating blade in the main annulus. In the presence of the intrinsic unsteady flow modes empirical correlations and Reynolds-averaged Navier-Stokes (RANS) models give solutions for sealing effectiveness that depart from the experimental measurements [14,15]. ...
... Schematics and some geometrical parameters are shown in Figure 2. The chute seal configuration reproduces the experimental rig studied by Beard et al. [27]. The axial seal geometry considers the studies by Gentilhomme et al. [28] and Boudet et al. [14], but with a larger disc rim radius identical to the chute seal configuration. The radial seal has the same disc rim radius as for the other two seals. ...
Rotating fluids are well-known to be susceptible to waves. This has received much attention from the geophysics, oceanographic and atmospheric research communities. Inertial waves, which are driven by restoring forces, for example the Coriolis force, have been detected in the research fields mentioned above. This paper investigates inertial waves in turbine rim seal flows in turbomachinery. These are associated with the large-scale unsteady flow structures having distinct frequencies, unrelated to the main annulus blading, identified in many experimental and numerical studies. These unsteady flow structures have been shown in some cases to reduce sealing effectiveness and are difficult to predict with conventional steady Reynolds-averaged Navier-Stokes (RANS) approaches. Improved understanding of the underlying flow mechanisms and how these could be controlled is needed to improve the efficiency and stability of gas turbines. This study presents large-eddy simulations for three rim seal configurations-chute, axial and radial rim seals-representative of those used in gas turbines. Evidence of inertial waves is shown in the axial and chute seals, with characteristic wave frequencies limited within the threshold for inertial waves given by classic linear theory (i.e. |f * /f rel | ≤ 2), and instantaneous flow fields showing helical characteristics. The radial seal, which limits the radial fluid motion with the seal geometry, restricts the Coriolis force and suppresses the inertial wave.
... Many parameters can influence the ingress of gaspath flow through the rim seal and inside the cavity, such as unsteady vane-blade interactions, geometry details of the rim seal, operating conditions or centrifugal pumping. In addition, some recent numerical studies (Jakoby et al. [2], Cao et al. [3], Boudet et al. [4]) of disk cavity flows performed on large sectors with an empty gaspath (without vanes nor blades) have suggested the possible existence of large scale flow structures within the cavity. Experimentally, energetic low frequencies were observed on pressure spectrums by Jakoby et al. [2], Cao et al. [3] and by Schuepbach et al. [5], who also suggested the existence of large scale structures. ...
... Having a characteristic wave length greater than that associated with the pitch of vanes and blades [3], these structures are susceptible to induce deep ingress of hot gases [2]. Their azimuthal count seems very much case dependent and it is known that they rotate with an angular velocity slightly less than that of the rotor (80% in [2] and [4], 90 to 97% in [3]). Nevertheless, our current knowledge of the underlying phenomena driving ingestion is still incomplete and a better global understanding of that complex flow physics is necessary, especially with respect to the dynamic influence of high-frequency vane-blade interactions with possible large scale flow structures inside the cavity. ...
Preliminary results of unsteady numerical simulations of disk cavity flow in interaction with the main gaspath flow in an axial turbine are presented in this article. A large periodic sector including vanes, blades and disk cavity of approximately 74° has been used in order to allow for the formation of large scale flow structures within the cavity. Three purge flow rates have been tested, namely no purge, low purge and high purge flow rates. Energetic large scale flow structures are detected through flow visualizations for the two lowest purge flow rates. They are found to rotate at an angular velocity slightly less than the rotor speed. The presence of the large scale structures involves important pressure perturbations inside the cavity that may lead to deep mass flow ingress, whereas the unsteady vane-blade interaction seems to cause only shallow ingress. Increasing purge flow rate appears to have a stabilizing effect on the pressure fluctuations inside the cavity and to reduce the intensity of the large scale flow structures.
... Cavity flows between a rotating disc and a fixed stator [2][3][4][5] and more particularly the interaction between the main annulus flow and the cavity rim seal [6][7][8][9][10][11] have been the subject of many studies. Hot air from the annulus is ingested into the cavity, and will reduce the disc life. ...
... This may be linked to unsteadiness in the rim seal. As identified in earlier studies [7,8], rim seal flows may well be inherently unsteady for low net rim seal throughflow rates, as occurs at this coolant flow rate. ...
In axial gas turbines, hot air from the main annulus path tends to be ingested into the turbine disc cavities. This leads to overheating which will reduce the disc's life time or lead to serious damage. Often, to overcome this problem, some air is extracted from the compressor to cool the rotor discs. This also helps seal the rim seals and to protect the disc from the hot annulus gas. However, this will deteriorate the overall efficiency. A detailed knowledge of the flow interaction between the main gas path and the disc cavities is necessary in order to optimise thermal effectiveness against overall efficiency due to losses of the cooling air from the main gas path. The aim of this study is to provide better understanding of the flow in a turbine stator-well, and evaluate the use of different CFD methods for this complex, 3-dimensional unsteady flow. This study presents CFD results for a 2-stage turbine. The stator-well cavity for the second row of stationary vanes is included in the calculation and results for both turbine performance and stator-well sealing efficiency are presented.
... The results showed periodic large-scale structures in the seal gap, which were attributed to the interaction between the annulus and rim seal flows. Boudet et al. 13 conducted both steady and unsteady Reynolds-Averaged Navier-Stokes simulations in a rotor-stator system with an axial seal under axis-symmetrical annulus flow conditions. Unsteady results tended to show a decreased sealing effectiveness. ...
The presence of a rotating disk adjacent to a stationary disk forms a rotor–stator cavity known as a wheel-space. It is necessary for gas turbine wheel-spaces to be purged with sealing flow bled from the compressor to counteract the harmful effects of ingress. This paper presents a combined experimental, theoretical, and computational study of rotationally induced ingress in rotor–stator systems. Measurements were made in a wheel-space with an axial clearance rim seal under axisymmetric conditions in the absence of a mainstream annulus through-flow. Ingress was quantified using a gas concentration technique and the flow structure in the cavity was explored with static and total pressure measurements to determine the swirl ratio. A low-order theoretical model was developed based on the boundary layer momentum-integral equations. The theory gave excellent results when predicting the effects of ingress and purge flows on the radial pressure and swirl gradients. Unsteady Reynolds-Averaged Navier–Stokes computations were conducted to provide greater fluid dynamic insight into the wheel-space flow structure and ingress through the rim seal. The computational results demonstrated some of the closest agreement with experimental measurements of ingress available in the literature, showing that rotationally induced ingress is dominated by unsteady large-scale structures in the rim seal gap instead of the previously ascribed disk-pumping effect. The study serves as an important validation case for investigations of ingress in rotor–stator systems in more complex environments.
... This kind of flows are influenced by disk pumping, main gas path pressure asymmetry due to vane wake flows, and inherent unsteady flow modes. 22 Steady RANS was reported unable to accurately reproduce the ingestion level, 23 while LES 24 and WMLES 25 show better agreement with measured sealing effectiveness. Some recent work shows reasonable agreement between unsteady RANS (URANS) and experiment in turbine rim seal cavities, 26,27 indicating URANS as a viable approach to simulate disk cavity flows for a good level of accuracy. ...
Inter-shaft fuel supply systems (ISFSSs) consisting of a shaft annulus and a fuel slinger are usually used in micro turbojet engines to pump the fuel to the combustor, aiming at reducing engine's size and weight. Differing from the configuration with stationary fuel pipework, the fast-rotating shaft in the annulus would induce significant drag jeopardizing the fuel supply during engine acceleration. This study investigates the flow physics inside a micro turbojet engine ISFSS. Air–fuel two-phase flow is found in the slinger cavity and the shaft annulus, with main gas path air ingress into the ISFSS under disk pumping of the slinger, and the ingress is enhanced by introducing blades on the slinger disk. Results show that the axial flow drag in the shaft annulus is reduced as air being ingested in. Further investigation indicates that the best supplied fuel volume fraction is 0.8, and this can reduce the axial drag in the shaft annulus by ∼63%, compared with the single-phase fuel flow. Therefore, two-phase air–fuel mixture is proposed for the ISFSS drag reduction.
... Otherwise, the amplitude of RBPF is dominant in cavity, which indicates that the unsteady characteristics of gas ingress mainly come from the rotating effect of rotor blade. Boudet et al. [18] performed unsteady numerical simulations and the low-frequency was found at the seal rim. They attributed this frequency to Taylor-Couette instability. ...
This paper presents an unsteady numerical study of the unsteady flow in the unscalloped radial turbine cavity considering the effect of computational sector size and sealing flow rate. A simplified U-shaped cavity with vanes and blades was simulated and the sealing efficiency was obtained using additional variables method. Unsteady low-pressure structures exist under a certain sealing flow rate, which is similar with axial turbine cavity, while the impact area is much larger. A comparison of computational domains sizes for 60° and 360° shows that whether the sector model can accurately predict the flow field depends highly on the number of low-pressure structures. As the sealing flow rate increases, the number and the speed of low-pressure structures at high radius decrease, and the corresponding low-frequency amplitude increases initially but decreases subsequently. The speed of low-pressure structure is positively correlated with its number, which seems contrary to the phenomenon observed in the axial turbine cavity.
... Experimental data are presented at two differing rotational speeds, largely collapsing with an insensitivity to Reynolds number. There is an inflection in the experimental data for 0.06 < U 0 < 0.12, which is qualitatively similar to the experimental data reported elsewhere [1,22,34,35]. Here, the data have been collected at a flow coefficient C F ¼ 0.35, which is the design point for the stage; the degree of inflection is sensitive to flow coefficient and has not been accurately captured computationally, leading to the mismatch in sealing effectiveness in Fig. 11(c). ...
The ingress of hot annulus gas into stator-rotor cavities is an important topic to engine designers. Rim-seals reduce the pressurised purge required to protect highly-stressed components. This paper describes an experimental and computational study of flow through a turbine chute seal. The computations - which include a 360º domain - were undertaken using DLR TRACE's time-marching solver. The experiments used a low Reynolds number turbine rig operating with an engine-representative flow structure. The simulations provide an excellent prediction of cavity pressure and swirl, and good overall agreement of sealing effectiveness when compared to experiment. Computation of flow within the chute seal showed strong shear gradients which influence the pressure distribution and secondary-flow field near the blade leading edge. High levels of shear across the rim-seal promote the formation of largescale structures at the wheel-space periphery; the number and speed of which were measured experimentally and captured, qualitatively and quantitatively, by computations. A comparison of computational domains ranging from 30º to 360º indicate that steady features of the flow are largely unaffected by sector size. However, differences in large-scale flow structures were pronounced with a 60º sector and suggest that modelling an even number of blades in small sector simulations should be avoided.
... The inflection is consistent with the previously published data by Horwood et al. [17] for the same configuration as vane-P1. Similar inflected curves have been published by Boudet et al. [21], Gentilhomme et al. [22], and Clark et al. [23], though the authors do not present an explanation for the phenomenon. Here, the data have been collected at a flow coefficient of C F ¼ 0.407 corresponding to the design point of the stage. ...
This paper presents experimental and computational results using a 1.5-stage test rig designed to investigate the effects of ingress through a double radial overlap rim-seal. The effect of the vanes and blades on ingress was investigated by a series of carefully-controlled experiments: firstly, the position of the vane relative to the rim seal was varied; secondly, the effect of the rotor blades was isolated using a disc with and without blades.
Measurements of steady pressure in the annulus show a strong influence of the vane position. The relationship between sealing effectiveness and purge flow-rate exhibited a pronounced inflexion for intermediate levels of purge; the inflexion did not occur for experiments with a bladeless rotor. Shifting the vane closer to the rim-seal, and therefore the blade, caused a local increase in ingress in the inflexion region; again this effect was not observed for the bladeless experiments.
Unsteady pressure measurements at the periphery of the wheel-space revealed the existence of large-scale pressure structures (or instabilities) which depended weakly on the vane position and sealing flow rate. These were measured with and without the blades on the rotor disc. In all cases these structures rotated close to the disc speed.
... Similar, more distinct, behavior has been reported by Horwood et al. [39]; the behavior effect of Re / on axial distribution of C p,n over the NGV taps at 50% span; C F 5 0.38; U 0 5 0 was associated with a magnification of low-frequency unsteadiness in the wheel-space. Boudet et al. [40], Gentilhomme et al. [41] and Clark et al. [42] also report a similar inflection phenomenon but the authors do not discuss the driving mechanism. Further work by Hualca et al. [43] has shown that inflection in the variation of effectiveness disappears when the blades are removed. ...
In modern gas turbines, endwall contouring (EWC) is employed to modify the static pressure field downstream of the vanes and minimise the growth of secondary flow structures developed in the blade passage. Purge flow (or egress) from the upstream rim-seal interferes with the mainstream flow, adding to the loss generated in the rotor. Despite this, EWC is typically designed without consideration of mainstream-egress interactions. The performance gains offered by EWC can be reduced, or in the limit eliminated, when purge air is considered. In addition, EWC can result in a reduction in sealing effectiveness across the rim seal. Consequently, industry is pursuing a combined design approach that encompasses the rim-seal, seal-clearance profile and EWC on the rotor endwall. This paper presents the design of, and preliminary results from a new single-stage axial turbine facility developed to investigate the fundamental fluid dynamics of egress-mainstream flow interactions. To the authors' knowledge this is the only test facility in the world capable of investigating the interaction effects between cavity flows, rim seals and EWC. The design of optical measurement capabilities for future studies, employing volumetric velocimetry and planar laser induced fluorescence are also presented. The fluid-dynamically scaled rig operates at benign pressures and temperatures suited to these techniques and is modular. The facility enables expedient interchange of EWC (integrated into the rotor bling), blade-fillet and rim-seals geometries.
... Experimental data are presented at two differing rotational speeds, largely collapsing with an insensitivity to Reynolds number. There is an inflexion in the experimental data for 0.06 < Φ0 < 0.12, which is qualitatively similar to experimental data reported elsewhere [1,22,34,35]. Here the data has been collected at a flow coefficient CF = 0.35, which is the design point for the stage; the degree of inflexion is sensitive to flow coefficient, and has not been accurately captured computationally, leading to the mismatch in sealing effectiveness in Figure 11 (c). ...
The ingress of hot annulus gas into stator-rotor cavities is an important topic to engine designers. Rim-seals reduce the pressurised purge required to protect highly-stressed components. This paper describes an experimental and computational study of flow through a turbine chute seal. The computations — which include a 360° domain — were undertaken using DLR TRACE’s time-marching solver. The experiments used a low Reynolds number turbine rig operating with an engine-representative flow structure. The simulations provide an excellent prediction of cavity pressure and swirl, and good overall agreement of sealing effectiveness when compared to experiment.
Computation of flow within the chute seal showed strong shear gradients which influence the pressure distribution and secondary-flow field near the blade leading edge. High levels of shear across the rim-seal promote the formation of large-scale structures at the wheel-space periphery; the number and speed of which were measured experimentally and captured, qualitatively and quantitatively, by computations.
A comparison of computational domains ranging from 30° to 360° indicate that steady features of the flow are largely unaffected by sector size. However, differences in large-scale flow structures were pronounced with a 60° sector and suggest that modelling an even number of blades in small sector simulations should be avoided.
... The inflexion is consistent with previously published data by Horwood et al. [17] for the same configuration as vane-P1. Similar inflected curves have been published by Boudet et al. [21], Gentilhomme et al. [22] and Clark et al. [23], though the authors do not present an explanation for the phenomenon. ...
This paper presents experimental and computational results using a 1.5-stage test rig designed to investigate the effects of ingress through a double radial overlap rim-seal. The effect of the vanes and blades on ingress was investigated by a series of carefully-controlled experiments: firstly, the position of the vane relative to the rim seal was varied; secondly, the effect of the rotor blades was isolated using a disc with and without blades.
Measurements of steady pressure in the annulus show a strong influence of the vane position. The relationship between sealing effectiveness and purge flow-rate exhibited a pronounced inflexion for intermediate levels of purge; the inflexion did not occur for experiments with a bladeless rotor. Shifting the vane closer to the rim-seal, and therefore the blade, caused a local increase in ingress in the inflexion region; again this effect was not observed for the bladeless experiments.
Unsteady pressure measurements at the periphery of the wheel-space revealed the existence of large-scale pressure structures (or instabilities) which depended weakly on the vane position and sealing flow rate. These were measured with and without the blades on the rotor disc. In all cases these structures rotated close to the disc speed.
... Similar, more distinct, behaviour has been reported by Horwood et al. [37]; the behaviour was associated with a magnification of low-frequency unsteadiness in the wheel-space. Boudet et al. [38], Gentilhomme et al. [39] and Clark et al. [40] also report a similar inflexion phenomenon but the authors do not discuss the driving mechanism; this research group believe that the turn in the RBs influences this behaviour. ...
In modern gas turbines, endwall contouring (EWC) is employed to modify the static pressure field downstream of the vanes and minimise the growth of secondary flow structures developed in the blade passage. Purge flow (or egress) from the upstream rim-seal interferes with the mainstream flow, adding to the loss generated in the rotor. Despite this, EWC is typically designed without consideration of mainstream-egress interactions. The performance gains offered by EWC can be reduced, or in the limit eliminated, when purge air is considered. In addition, EWC can result in a reduction in sealing effectiveness across the rim seal. Consequently, industry is pursuing a combined design approach that encompasses the rim-seal, seal-clearance profile and EWC on the rotor endwall.
This paper presents the design of, and preliminary results from a new single-stage axial turbine facility developed to investigate the fundamental fluid dynamics of egress-mainstream flow interactions. To the authors’ knowledge this is the only test facility in the world capable of investigating the interaction effects between cavity flows, rim seals and EWC. The design of optical measurement capabilities for future studies, employing volumetric velocimetry and planar laser induced fluorescence are also presented. The fluid-dynamically scaled rig operates at benign pressures and temperatures suited to these techniques and is modular. The facility enables expedient interchange of EWC (integrated into the rotor bling), blade-fillet and rim-seals geometries.
The measurements presented in this paper include: gas concentration effectiveness and swirl measurements on the stator wall and in the wheel-space core; pressure distributions around the nozzle guide vanes at three different spanwise locations; pitchwise static pressure distributions downstream of the nozzle guide vane at four axial locations on the stator platform.
... In addition to the mechanisms stated above, Smout et al. [2002] and Chew et al. [2003] reported large-scale low-frequency flow structures, even in an axisymmetric geometry without vanes and blades. The URANS simulations on rim seal geometries of Boudet et al. [2005] revealed that those structures are inherently 3-D and unsteady, and that reasonable predictions of those features might improve the estimation of sealing effectiveness. They attributed this phenomenon to the possible Taylor-Couette instability. ...
... method is found to improve the numerical solutions in simulating turbomachinery flows, compared with steady RANS approach. [53][54][55] Thus, the idea is to investigate the sensitivity of the corner separation to realistic inflow perturbations in URANS solutions. Perturbations are imposed on the inlet plane by varying the inflow angle. ...
Large-eddy simulation (LES) is compared with experiment and Reynolds-averaged Navier-Stokes (RANS), and LES is shown to be superior to RANS in reproducing corner separation in the LMFA-NACA65 linear compressor cascade, in terms of surface limiting streamlines, blade pressure coefficient, total pressure losses and blade suction side boundary layer profiles. However, LES is too expensive to conduct an influencing parameter study of the corner separation. RANS approach, despite over-predicting the corner separation, gives reasonable descriptions of the corner separated flow, and is thus selected to conduct a parametric study in this paper. Two kinds of influencing parameters on corner separation, numerical and physical parameters, are analyzed and discussed: second order spatial scheme is necessary for a RANS simulation; incidence angle and inflow boundary layer thickness are found to show the most significant influences on the corner separation among the parameters studied; unsteady RANS with the imposed inflow unsteadiness does not show any non-linear effect on the corner separation.
... Instabilities and flow unsteadiness in rim-seal flows are discussed further by Boudet et al. [44]. CFD solutions for the unsteady equations were obtained for the three experimental configurations including that of Gentilhomme et al. [28]. ...
This review summarizes research concerned with the ingress of hot mainstream gas through the rim seals of gas turbines. It includes experimental, theoretical, and computational studies conducted by many institutions, and the ingress is classified as externally induced (EI), rotationally induced (RI), and combined ingress (CI). Although EI ingress (which is caused by the circumferential distribution of pressure created by the vanes and blades in the turbine annulus) occurs in all turbines, RI and CI ingress can be important at off-design conditions and for the inner seal of a double-seal geometry. For all three types of ingress, the equations from a simple orifice model are shown to be useful for relating the sealing effectiveness (and therefore the amount of hot gas ingested into the wheel-space of a turbine) to the sealing flow rate. In this paper, experimental data obtained from different research groups have been transformed into a consistent format and reviewed using the orifice model equations. Most of the published results for sealing effectiveness have been made using concentration measurements of a tracer gas (usually CO2) on the surface of the stator, and - for a large number of tests with single and double seals - the measured distributions of effectiveness with sealing flow rate are shown to be consistent with those predicted by the model. Although the flow through the rim seal can be treated as inviscid, the flow inside the wheel-space is controlled by the boundary layers on the rotor and stator. Using boundary-layer theory and the similarity between the transfer of mass and energy, a theoretical model has been developed to relate the adiabatic effectiveness on the rotor to the sealing effectiveness of the rim seal. Concentration measurements on the stator and infrared (IR) measurements on the rotor have confirmed that, even when ingress occurs, the sealing flow will help to protect the rotor from the effect of hot-gas ingestion. Despite the improved understanding of the "ingress problem," there are still many unanswered questions to be addressed.
... As reported by Hills et al. [7], Gentilhomme et al. [8], Roy et al. [9] and Chew [10], turbulent, three dimensional and unsteady effects have to be properly taken into account to correctly model the flow dynamics in a rim seal environment. In more recent numerical and experimental studies, namely by Julien et al. [11], Jakoby et al. [12], Cao et al. [13] and Boudet et al. [14], the existence of large-scale structures has been reported and their pressure traces have been correlated to ingress paths. As shown in [11], these structures are not necessarily related to the vane or the blade counts and they can account for an important part of the ingestion phenomenon. ...
This paper reports the first phase of an investigation aiming to determine the validity of using a CO2 marker in cold rig experiments to characterize the thermal performances of turbine rim seals under actual engine operating conditions. For comparison purposes, simulations are carried out for two sets of operating conditions, namely cold rig (with uniform low temperature) and real turbine thermal conditions (high temperature gaspath and cold purge flow). Sealing effectiveness based on the CO2 diagnostic under cold rig operating conditions is compared to sealing effectiveness based on the computed temperature field under real engine temperature conditions. Unsteady RANS simulations with different purge flow rates are performed. Tested geometries include a 180° domain presenting a simplified rim seal geometry with no vanes nor blades in the gaspath, and a 24° sector of a complete turbine stage including 3 vanes and 4 blades. Three-dimensional flow structures known to affect ingestion are found with both geometries but appear to be sensitive to the differences in operating conditions. Indeed, their circumferential number and strength differ between the two scenarios of conditions. Furthermore, it is found that the cold rig predictor tends to slightly overestimate the sealing effectiveness, while providing nonetheless the right trends and reasonably accurate average values in levels of actual sealing. At this stage of the investigation, we conclude that it seems adequate to use a passive tracer in cold rig experiments to compare performances of rim seal designs.
... As illustrated by Boudet et al. [29,30], turbine rim seal flows (in common with compressor disc cavities) can exhibit inherent large scale unsteadiness. It has yet to be established if URANS models can accurately predict hot gas ingestion through the rim seals that can severely affect disc heat transfer. ...
Use of large-scale computational fluid dynamics (CFD) models in aeroengine design has grown rapidly in recent years as parallel computing hardware has become available. This has reached the point where research aimed at the development of CFD-based ‘virtual engine test cells’ is underway, with considerable debate of the subject within the industrial and research communities. The present article considers and illustrates the state-of-the art and prospects for advances in this field. Limitations to CFD model accuracy, the need for aero-thermo-mechanical analysis through an engine flight cycle, coupling of numerical solutions for solid and fluid domains, and timescales for capability development are considered. While the fidelity of large-scale CFD models will remain limited by turbulence modelling and other issues for the foreseeable future, it is clear that use of multi-scale, multi-physics modelling in engine design will expand considerably. Development of user-friendly, versatile, efficient programs and systems for use in a massively parallel computing environment is considered a key issue.
... Instabilities and flow unsteadiness in rim seal flows are discussed further by Boudet et al. (2005Boudet et al. ( , 2006. For a rim seal geometry representative of aeroengines and previously studied experimentally by , unsteady CFD solutions including both blades and vanes were obtained. ...
Considerable progress in development and application of computational fluid dynamics (CFD) for aeroengine internal flow systems has been made in recent years. CFD is regularly used in industry for assessment of air systems, and the performance of CFD for basic axisymmetric rotor/rotor and stator/rotor disc cavities with radial throughflow is largely understood and documented. Incorporation of three-dimensional geometrical features and calculation of unsteady flows are becoming commonplace. Automation of CFD, coupling with thermal models of the solid components, and extension of CFD models to include both air system and main gas path flows are current areas of development. CFD is also being used as a research tool to investigate a number of flow phenomena that are not yet fully understood. These include buoyancy-affected flows in rotating cavities, rim seal flows and mixed air/oil flows. Large eddy simulation has shown considerable promise for the buoyancy-driven flows and its use for air system flows is expected to expand in the future.
This experimental study considered the performance of a chute rim seal downstream of turbine inlet guide vanes (but without rotor blades). The experimental set up reproduced rotationally-driven ingestion without vanes and conditions of pressure-driven ingestion with vanes. The maximum rotor speed was 9000 rpm corresponding to a rotational Reynolds number of 3.3x106 with a flow coefficient of 0.485. Measurements of mean pressures in the annulus and the disc rim cavity as well as values of sealing effectiveness deduced from gas concentration data are presented. At high values of flow coefficient (low rotational speeds), the circumferential pressure variation generated by the vanes drove relatively high levels of ingestion into the disc rim cavity. For a given purge flow rate, increasing the disc rotational speed led to a reduction in ingestion, shown by higher values of sealing effectiveness, despite the presence of upstream vanes. At U_ax/((Ob))="0.485" , the sealing effectiveness approached that associated with purely rotationally-driven ingestion. A map of sealing effectiveness against non-dimensional purge flow summarises the results and illustrates the combined rotational and pressure-driven effects on the ingestion mechanism. The results imply that flow coefficient is an important parameter in rim sealing and that rotational effects are important in many applications, especially turbines with low flow coefficient
The existence and causes of the deep ingress into the core region of a turbine rotor-stator disc cavity, or core penetration flow, generated by a rotating non-axisymmetric geometry have been investigated experimentally. In a low-speed, low-expansion ratio, single-stage, cold turbine test facility, time-resolved tangential and radial velocities in the cavity have been measured with 2-D hot-wire anemometers. In addition, time-resolved static pressures on the stator disc have been measured with fast response pressure transducers, and unsteady cavity velocity field in the absolute frame has been measured using Particle Image Velocimetry (PIV). Rotating geometric non-axisymmetry leads to an unsteady radial pressure gradient in the disc cavity. A time lag in the tangential velocity adjustment to the variation in the radial pressure gradient results in a net radial force, causing core penetration flow. A first-harmonic rotating geometric non-axisymmetry makes the core penetration flow to occur three times per revolution (twice when the cavity exit pressure increases and once when the cavity exit pressure decreases) and to revolve at the disc’s rotational speed. Core penetration flow due to the rotating geometric asymmetry is not affected by variations in the annulus flow coefficient or rotational Reynolds number but is weakened by increasing purge air flow rate.
This paper presents a numerical investigation on the steady and unsteady flow characteristics of rim seal for the first stage in gas turbine. The Reynolds-averaged Navier-Stokes equations, coupled with Shear Stress Transport turbulence model, and a scalar equation are solved. In the steady-state conditions, the influences of rim seal structures and rotational speeds on the sealing effectiveness and non-dimensional sealing air mass flow have been numerically investigated. In the unsteady-state conditions, the pressure distribution, pressure-frequency and sealing effectiveness are discussed in detail. The results obtained in this study indicate that the sealing effectiveness increases with the non-dimensional sealing air mass flow. The minimum non-dimensional sealing mass flow for the axial rim seals and radial seals ranges from 9.62×103 to 9.63×103 and from 5.07×103 to 5.75×103, respectively. It means that the radial seals perform better than the axial ones in sealing effectiveness. The minimum non-dimensional sealing mass flow increases with the increase of rotational speeds. Compared to Cw,seal=9840, there is an additional pressure frequency of f/fbld=0.125 for the case of Cw,seal=4070. The increase of sealing air is conducive to reducing instability in the rim seal cavity, which has small influences on the circumferential pressure downstream the vane trailing edge. The unsteady-state sealing effectiveness is lower than that calculated in steady state.
This paper describes the work done to achieve high parallel performance for an unstructured, unsteady turbomachinery computational fluid dynamics (CFD) code. The aim of the work described here is to be able to scale problems to the thousands of processors that current and future machine architectures will provide. The CFD code is in design use in industry and is also used as a research tool at a number of universities. High parallel scalability has been achieved for a range of turbomachinery test cases, from steady-state hexahedral mesh cases to fully unsteady unstructured mesh cases. This has been achieved by a combination of code modification and consideration of the parallel partitioning strategy and resulting load balancing. A sliding plane option is necessary to run fully unsteady multistage turbomachinery test cases and this has been implemented within the CFD code. Sample CFD calculations of a full turbine including parts of the internal air system are presented.
Accurate prediction of turbine blade channel and disc cavity flows remains a chal-lenging task despite considerable work in this area and the acceptance of CFD as a design tool. The quality of the CFD calculations of the flows in turbomachinery applications strongly depends on the proper prediction of turbulence phenomena. In-vestigations of heat transfer, skin friction, secondary flows, flow separation and re-attachment effects demand a reliable simulation of the turbulence, reliable methods, accurate programming, and robust working practices. The study addresses some ques-tions related to development, verification and validation of turbulence models, and focuses on development of best practice for combined blade passage and disc cavity flow calculations. The study involves some basic validation studies for the k-e and Spalart-Allmaras turbulence models. The k-e model with or without Kato-Launder correction and Richardson number correction for curvature of streamlines, standard and modified Spalart-Allmaras model and two-layer model are validated for rotating disc cavity systems. The test cases considered include benchmark cases for flat plate flow and heat transfer, rotating disc flow, a combined turbine blade/disc cavity model, and low-speed compressor blade flow. Comparisons are made with experimental data and computations from different CFD codes.
Large-Eddy Simulations of wall bounded, low Mach number turbulent flows are conducted using an unstructured finite-volume solver of the compressible flow equations. The numerical method employs linear reconstructions of the primitive variables based on the least-squares approach of Barth. The standard Smagorinsky model is adopted as the subgrid term. The artificial viscosity inherent to the spatial discretization is maintained as low as possible reducing the dissipative contribution embedded in the approximate Riemann solver to the minimum necessary. Comparisons are also discussed with the results obtained using the implicit LES procedure.
Two canonical test-cases are described: a fully developed pipe flow at a bulk Reynolds number Reb = 44 × 103 based on the pipe diameter, and a confined rotor-stator flow at the rotational Reynolds number ReΩ = 4 × 105 based on the outer radius. In both cases the mean flow and the turbulent statistics agree well with existing DNS or experimental data.
Numerical simulations of turbine rim seal experiments are conducted with a time-dependent, 360-degree CFD model of the complete turbine stage with a rim seal and cavity. The turbine stage has 22 vanes and 28 blades and is modeled with a uniform flow upstream of the vane inlet, a pressure condition downstream of the blades and three coolant flow conditions previously employed during experiments at Arizona State University. The simulations show the pressure fields downstream of the vanes and upstream of the blades interacting to form a complex pressure pattern above the rim seal. Circumferential distributions of 15 and 17 sets of ingress and egress velocities flow through the rim seal at the two modest coolant flow rate conditions. These flow distributions rotate at wheel speed and are not associated with the numbers of blades or vanes. The seal velocity distribution for a high coolant flow rate with little or no ingestion into the stator wall boundary layer is associated with the blade pressure field. These pressure field characteristics and the rim seal ingress/egress pattern provide new insight to the physics of rim seal ingestion. Flow patterns within the rim cavity have large cells that rotate in the wheel direction at a slightly slower speed. These secondary flows are similar to structures noted in previous a 360-degree model and large sector models but not obtained in a single blade or vane sector model with periodic boundary condition at sector boundaries. The predictions of pressure profiles, sealing effectiveness and cavity velocity components are compared with experimental data.
This paper presents numerical simulations of the unsteady flow interactions between the main annulus and the disc cavity for an axial turbine. The simulations show the influence of the main annulus asymmetries (vane wakes, blade potential effect), and the appearance of rim seal flow instabilities. The generation of secondary frequencies due to non-linear interactions is observed, and the possibility of further low frequency effects and resonance is noted. The computations are compared to experimental results, looking at tracer gas concentration and mass-flows. Results are further analysed to investigate the influence of the rim seal flow on the blading aerodynamics. The flow that is ejected through the rim seal influences the unsteady flow impinging the blades. The influence of this rim-seal flow is even observed downstream of the blades, where it distorts the radial profile of stagnation temperature.
Large-Eddy Simulations (LES) were carried out for a turbine rim seal and
the sensitivity of the results to changes in grid resolution and the
size of the computational domain are investigated. Ingestion of hot
annulus gas into the rotor-stator cavity is compared between LES results
and against experiments and Unsteady Reynolds-Averaged
Navier-Stokes (URANS) calculations. The LES calculations show
greater ingestion than the URANS calculation and show better agreement
with experiments. Increased grid resolution shows a small improvement in
ingestion predictions whereas increasing the sector model size has
little effect on the results. The contrast between the different CFD
models is most stark in the inner cavity, where the URANS shows almost
no ingestion. Particular attention is also paid to the presence of low
frequency oscillations in the disc cavity. URANS calculations show such
low frequency oscillations at different frequencies than the LES. The
oscillations also take a very long time to develop in the LES. The
results show that the difficult problem of estimating ingestion through
rim seals could be overcome by using LES but that the computational
requirements were still restrictive.
Unsteady flow dynamics in turbine rim seals are known to be complex and attempts accurately to predict the interaction of the mainstream flow with the secondary air system cooling flows using computational fluid dynamics (CFD) with Reynolds-averaged Navier–Stokes (RANS) turbulence models have proved difficult. In particular, published results from RANS models have over-predicted the sealing effectiveness of the rim seal, although their use in this context continues to be common. Previous studies have ascribed this discrepancy to the failure to model flow structures with a scale greater than the one which can be captured in the small-sector models typically used. This article presents results from a series of Large-Eddy Simulations (LES) of a turbine stage including a rim seal and rim cavity for, it is believed by the authors, the first time. The simulations were run at a rotational Reynolds number Reθ = 2.2 × 106 and a main annulus axial Reynolds number Rex = 1.3 × 106 and with varying levels of coolant mass flow. Comparison is made with previously published experimental data and with unsteady RANS simulations. The LES models are shown to be in closer agreement with the experimental sealing effectiveness than the unsteady RANS simulations. The result indicates that the previous failure to predict rim seal effectiveness was due to turbulence model limitations in the turbine rim seal flow. Consideration is given to the flow structure in this region.
This paper considers the influence of the mass flow of the gas ingested from the cavity formed by the rotor-to-stator clearance
on the formation and structure of secondary flows in the blade passage of the gas-turbine stage. The flow is described by
the Reynolds-averaged Navier-Stokes equations, to close which the Spalart-Allmaras model and the k-ε model of turbulence with
corrections for the rotation and curvature of the streamlines are used. Comparison of the results of the numerical simulation
obtained from the point of view of different turbulence models is made.
A transport equation for the turbulent viscosity is assembled, using empiricism and arguments of dimensional analysis, Galilean invariance, and selective dependence on the molecular viscosity. It has similarities with the models of Nee and Kovaszany, Secundov et al., and Baldwin and Barth. The equation includes a non-viscous destruction term that depends on the distance of the wall.
The transitional turbulent regime in confined flow between a rotating and a stationary disc is studied using direct numerical
simulation. Besides its fundamental importance as a three-dimensional prototype flow, such flows frequently arise in many
industrial devices, especially in turbomachinary applications. The present contribution extends the DNS simulation into the
turbulent flow regime, to a rotational Reynolds number Re =3 × 105. An annular rotor-stator cavity of radial extension ΔR and height H, is considered with L = 4.72(L = ΔR/H) and Rm = 2.33 (Rm = (R
1+ R
0)/ΔR). The direct numerical simulation is performed by integrating the time-dependent Navier–Stokes equations until a statistically
steady state is reached. A three-dimensional spectral method is used with the aim of providing both very accurate instantaneous
fields and reliable statistical data. The instantaneous quantities are analysed in order to enhance our knowledge of the physics
of turbulent rotating flows. Also, the results have been averaged so as to provide target turbulence data for any subsequent
modelling attempts at reproducing the flow.
This paper describes an experimental study of an air-cooled gas turbine disk using the model of a disk rotating near a shrouded stator. Measurements of pressure distribution, frictional moment, and the cooling air flow necessary to prevent the ingress of hot gases over the turbine disk are described for a range of rotational speeds, mass flow rates, and different geometries. The pressure distribution is shown to be calculable by the super-position of the pressure drop due to the shroud and the unshrouded distribution. Moment coefficients are shown to increase with increasing mass flow rate anddecreasing shroud clearance, but are little affected by the rotor/ stator gap. Applying Reynolds analogy to the moment coefficients, it is estimated that heat transfer from the rotor will be controlled primarily by rate of radial cooling flow at low rotational Reynolds numbers, and will be governed primarily by Reynolds number at high rotational speeds.
A simple method, for both estimating circumferential pressure variation in a turbine, and using this method to calculate the annulus flow ingested across a rim seal into the disc cavity, is presented. This method is compared with test data from 2 separate experimental programmes. The model is shown to collapse the majority of the test data well; with calibration it could form the basis of a preliminary design methodology. The model does not collapse data where unsteady pressure fluctuations were measured in the cavity, suggesting these fluctuations, when present, play an important role in determining how much annulus gas is ingested into the cavity.
Experimental measurements from a new single stage turbine are presented. The turbine has 26 vanes and 59 rotating blades with a design point stage expansion ratio of 2.5 and vane exit Mach number of 0.96. A variable sealing flow is supplied to the disc cavity upstream of the rotor and then enters the annulus through a simple axial clearance seal situated on the hub between the stator and rotor. Measurements at the annulus hub wall just downstream of the vanes show the degree of circumferential pressure variation. Further pressure measurements in the disc cavity indicate the strength of the swirling flow in the cavity, and show the effects of mainstream gas ingestion at low sealing flows. Ingestion is further quantified through seeding of the sealing air with nitrous oxide or carbon dioxide and measurement of gas concentrations in the cavity. Interpretation of the measurements is aided by steady and unsteady computational fluid dynamics solutions, and comparison with an elementary model of ingestion.
Sealing of the cavity formed between a stationary disk and a rotating disk under axisymmetric conditions is considered. A mathematical model of the flow in the cavity based on momentum integral methods is described and this is coupled to a simple model of the seal for the case when no ingress occurs. Predictions of the minimum imposed flow required to prevent ingress are obtained and shown to be in reasonable agreement with the data of Bayley and Owen (1970), Owen and Phadke (1980), Phadke (1982), and Phadke and Owen (1983a, 1983b, 1988). With an empirical constant in the model chosen to match these data, predictions for the minimum sealing flow are shown to be in good agreement with the measurements of Graber et al. (1987). The analysis of Phadke's data also indicates the measurements for small seal clearances must be viewed with caution due to errors in setting the seal clearance. These errors are estimated to be twice the minimum clearance considered. Seal behavior when ingress occurs is also considered and estimates of the amount of ingress are made from the available data.
A combined computational fluid dynamics (CFD) and experimental study of interaction of main gas path and rim sealing flow is reported. The experiments were conducted on a two stage axial turbine and included pressure measurements for the cavity formed between the stage 2 rotor disc and the upstream diaphragm for two values of the diaphragm-to-rotor axial clearance. The pressure measurements indicate that ingestion of the highly swirling annulus flow leads to increased vortex strength within the cavity. This effect is particularly strong for the larger axial clearance. Results from a number of steady and unsteady CFD models have been compared to the measured results. Good agreement between measurement and calculation for time-averaged pressures was obtained using unsteady CFD models, which predicted previously unknown unsteady flow features. This led to fast response pressure transducer measurements being made on the rig, and these confirmed the CFD prediction.
This note describes one- and two-parameter families of solutions of steady rotationally-symmetric viscous flow. The solutions are such that the Navier-Stokes equations reduce to ordinary differential equations in a single position variable. The one-parameter family represents flow which is rigid-body rotation at infinity and over a plane through the origin; the solution given by von Kármán in 1921 is one member of this family. The two-parameter family represents flow which is rigid body rotation over each of two planes at a finite distance apart. The case of large Reynolds number is particularly interesting, since the two bounding planes are then separated by a region of rigid-body rotation and translation in which viscous effects are negligible.
The Kolmogorov-Prandtl turbulence energy hypothesis is formulated in a way which is valid for the laminar sublayer as well as the fully turbulent region of a one-dimensional flow. The necessary constants are fitted to available experimental data. Numerical solutions are obtained for Couette flow with turbulence augmentation and pressure gradient and for turbulent duct flow. Reasonable agreement with available experimental data is obtained. Some new dimensionless groups are used and shown to be superior to the ones based on the friction velocity. The effects of turbulence augmentation and pressure gradient on the velocity and temperature distribution are studied. It is found that the solutions tend to approach solutions for limiting cases. The results are plotted in some figures in Section 5.
The sealing characteristics of a shrouded rotor-stator system have been studied using flow visualization, pressure, and concentration measurements. Seven shroud geometries, incorporating axial clearance, radial clearance, or mitered seals, have been tested for a range of clearance ratios and rotational Reynolds numbers up to Reθ = 1.2 × 106. For all axial clearance seals, the superimposed airflow rate necessary to prevent the ingress of external fluid into the rotor-stator wheel space increased with rotational speed and with seal clearance. Owing to a “pressure inversion effect,” where the pressure in the wheel space increased rather than decreased with rotational speed, the increase of sealing flow rate with rotational speed for two of the radial clearance seals was less than that for the other seals. As expected, the mitered seal had a performance intermediate between the purely axial clearance seals and the radial clearance seals exhibiting the pressure inversion effect. The tests referred to above were conducted in a quiescent environment. In parts 2 and 3, the effect of an external axial flow of air at the periphery of the system is studied.
The sealing characteristics of a shrouded rotor-stator system with an external flow of air have been studied using flow visualization, pressure, and concentration measurements. Three shroud geometries, incorporating axial clearance or radial clearance seals, have been tested for a range of clearance ratios for both rotational and axial Reynolds numbers Reθ and Rew respectively, up to 1.2 × 106. It was found that there were two regimes: a rotation-dominated regime at low values of Rew/Reθ, and an external-flow-dominated regime at high values. For Rew = 0, Cw, min (the dimensionless flow rate of sealing air necessary to prevent the ingress of external fluid into the rotor-stator wheel space) increased with increasing Reθ. For small values of Rew/Reθ, Cw, min decreased with increasing Rew; for large values of Rew/Reθ, Cw, min was proportional to Rew and was independent of Reθ. The latter effect is attributed to the nonaxisymmetric pressure distribution in the external flow: fluid moved transversely across the wheel space from high-pressure to low-pressure regions of the external flow.
This book addresses rotor-stator systems. Topics covered include: Basic Equations; Laminar flow over a single disc; Turbulent flow over a single disc; Heat transfer from a single disc; Rotor-stator systems with no superposed flow; Rotor-stator systems with a superposed flow; Heat transfer in rotor-stator systems; and Sealing rotor-stator systems: The ingress Problem.
The paper provides a review of fluid flow and heat transfer in the
rotating disk systems that are relevant to designers of turbomachinery.
Starting with the free disk, the review includes rotor-stator systems
(which are used to simulate turbine disks rotating near stationary
casings) and rotating cavities (which are used to simulate corotating
turbine disks). Although there are many papers devoted to these systems,
and some design information does exist, there are still a number of
important areas that need further theoretical and experimental research.
These areas include the study of mainstream gas ingress (in a
rotor-stator system) and turbulent buoyancy-driven flows (inside
rotating cavities).
Turbine Rim Seal Ingestion
- O Gentilhomme
GENTILHOMME, O. Turbine Rim Seal Ingestion, PhD Thesis, University
of Sussex, 2004.
Rim sealing of rotor stator wheelspaces in the presence of external flow
- J W Chew
- T Green
- A B Turner
CHEW, J.W., GREEN, T. and TURNER, A.B. Rim sealing of rotor stator
wheelspaces in the presence of external flow, ASME Paper 94-GT-126,
1994.