Experimental and numerical investigations of tip injection on tip clearance flow in an axial turbine cascade

This investigation presents detailed experimental measurements of an active tip-clearance control method based on tip injection in a high-turning axial turbine cascade. Besides that, numerical investigations are also conducted to study phenomena which are not easily measured in the experiments. It aims to study the influence of tip injection on tip clearance flow, with emphasis on the effects of injection locations. Detailed flow field measurements were made downstream of the cascade using a three-hole probe. Static pressure distributions were also measured on the blade surface at 50% and 97.5% span, respectively. The results suggest that tip injection can weaken tip clearance flow, reducing the tip clearance mass flow and its associated losses. Meanwhile, the heat transfer condition on the blade tip surface can be also improved significantly. It also can be found that injection chordwise location plays an important role in the redistribution of secondary flow within the cascade passage. When the same number of injection holes and injection mass flow are applied, holes located in the aft part of blade can perform much better than that in the front part.     

A Numerical Study of the Flow Around a Model Winged Seed in Auto-Rotation

In this study the flow around a winged-seed in auto-rotation is characterized using direct numerical simulations (DNS) at Reynolds number in the range 80–240, based on the descent speed and a characteristic chord length. In this range, the flow is approximately steady when observed from a reference frame fixed to the seed. For all cases, the flow structure consists of a wing tip vortex which describes a helical path, a vortex shed behind the nut of the seed and a stable leading edge vortex above the wing surface which merges with the tip vortex. With increasing Reynolds number, the leading edge vortex becomes more intense and gets closer to the wing surface. The simulation results also show the formation of a spanwise flow on the upper surface of the wing, moving fluid towards the wing tip in a region downstream and beneath the leading edge vortex. This spanwise flow is rather weak inside the core of the leading edge vortex, and the analysis of the streamlines show a very weak transport of vorticity along the vortex for the cases under consideration. The analysis of the flow suggests that the stabilization of the leading edge vortex is mainly due to non-inertial accelerations, although viscous effects may contribute, specially at lower Re. Furthermore, the leading edge vortex has been characterized by analysing the flow variables averaged along cross-sections of the vortex. While some quantities, like the spanwise velocity or the pressure inside the vortex, are rather insensitive to the threshold used to define the leading edge vortex, the same is not true for the circulation of the vortex or its averaged spanwise vorticity, due to the viscous nature of the vortex. Finally, it is observed that the spanwise vorticity scales with the angular rotation of the seed for the different Re.     

Heat transfer in film-cooled turbulent boundary layers at different blowing ratios as affected by longitudinal vortices

Effects of embedded longitudinal vortices on heat transfer in film-cooled turbulent boundary layers at different blowing ratios are discussed. These results were obtained in boundary layers at free-stream velocities of 10 and 15 m/s. Film coolant was injected from a single row of holes at blowing ratios of 0.47–1.26. A single longitudinal vortex was induced upstream of the film-cooling holes using a half-delta wing attached to the wind tunnel floor. Heat transfer measurements were made downstream of injection using a constant heat flux surface with 126 thermocouples for surface temperature measurements. For all blowing ratios examined, the embedded vortices cause significant alterations to wall heat transfer and to film cooling distributions. Measurrments of mean temperatures and mean velocities in spanwise planes show that high wall heat transfer regions are associated with regions of high near-wall longitudinal velocity where very little film coolant is present. In addition to high heat transfer regions associated with the vortex downwash, there are also secondary heat transfer peaks. These secondary peaks develop due to shear layer mixing and interaction between the vortex and cooling jets and become higher in magnitude and more persistent with downstream distance as the blowing ratio increases from 0.47 to 1.26.     

Effect of multiple DBD plasma actuators on the tip leakage flow structure and loss of a turbine cascade

In this paper, the effects of multiple dielectric barrier discharge (DBD) plasma actuators on the leakage flow structures and loss conditions have been numerically studied in an axial turbine cascade. Kriging surrogate model is adopted to obtain the optimal cases. The physical mechanism of flow structures inside the gap that control leakage flow is presented, which is obtained by analyzing the flow topology, the evolution of the flow structures and its influence on the secondary velocity and loss conditions in the passage as well. The results show that the induced vortex caused by DBD actuators can change the leakage flow direction inside the tip gap and make the separation bubble break earlier, leading to a new type of the flow pattern. When the actuators are applied, the speed of leakage flow is significantly reduced and the angle between leakage flow and main flow has an obviously diminution, causing the reduction of mixing losses in the passage compared with the Baseline case. Furthermore, the comparison of secondary velocity shows that the tip leakage vortex (TLV) approaches the suction surface, resulting in reduced affected area and weakened loss strength. Plasma actuators can diminish the loss coefficient in both TLV and passage vortex near the casing (PVC) zones. The actuators arranged near the trailing edge mainly affect the strength of TLV, while the actuators in the leading edge area contribute to the loss reduction in the zone of PVC.     

Tip gap height effects on the aerodynamic performance of a cavity squealer tip in a turbine cascade in comparison with plane tip results: part 2—aerodynamic losses

Tip gap height effects on aerodynamic losses downstream of a cavity squealer tip have been investigated in a linear turbine cascade for power generation, in comparison with plane tip results. Three-dimensional flow fields are measured with a five-hole probe for tip gap height-to-chord ratios of h/c = 0.5, 1.0, 1.5 and 2.0%. The cavity squealer tip has a full length squealer with its rim height-to-chord ratio of 5.51%. For a fixed value of h/c, the tip leakage vortex for the cavity squealer tip is always weaker than that for the plane tip, and the flow field in the passage vortex region for the cavity squealer tip is less influenced by the tip leakage flow than that for the plane tip. For the cavity squealer tip, there is no appreciable change in local aerodynamic loss with h/c in the passage vortex region, but local aerodynamic loss in the tip leakage vortex region increases with h/c. The roles of the cavity squealer tip in reducing aerodynamic loss in comparison with the plane tip case are twofold: (1) the cavity squealer tip decreases the leakage flow discharge in the region from the leading edge to the mid-chord, which leads to an aerodynamic loss reduction in the passage vortex region and (2) it also decreases the leakage flow discharge downstream of the mid-chord, which results in an aerodynamic loss reduction in the tip leakage vortex region.     

Three-dimensional flow structures and associated turbulence in the tip region of a waterjet pump rotor blade

Stereo particle image velocimetry measurements focus on the flow structure and turbulence within the tip leakage vortex (TLV) of an axial waterjet pump rotor. Unobstructed optical access to the sample area is achieved by matching the optical refractive index of the transparent pump with that of the fluid. Data obtained in closely spaced planes enable us to reconstruct the 3D TLV structure, including all components of the mean vorticity and strain-rate tensor along with the Reynolds stresses and associated turbulence production rates. The flow in the tip region is highly three-dimensional, and the characteristics of the TLV and leakage flow vary significantly along the blade tip chordwise direction. The TLV starts to roll up along the suction side tip corner of the blade, and it propagates within the passage toward the pressure side of the neighboring blade. A shear layer with increasing length connects the TLV to the blade tip and initially feeds vorticity into it. During initial rollup, the TLV involves entrainment of a few vortex filaments with predominantly circumferential vorticity from the blade tip. Being shed from the blade, these filaments also have high circumferential velocity and appear as swirling jets. The circumferential velocity in the TLV core is also substantially higher than that in the surrounding passage flow, but the velocity peak does not coincide with the point of maximum vorticity. When entrainment of filaments stops in the aft part of the passage, newly forming filaments wrap around the core in helical trajectories. In ensemble-averaged data, these filaments generate a vortical region that surrounds the TLV with vorticity that is perpendicular to that in the vortex core. Turbulence within the TLV is highly anisotropic and spatially non-uniform. Trends can be traced to high turbulent kinetic energy and turbulent shear stresses, e.g., in the shear layer containing the vortex filaments and the contraction region situated along the line where the leakage backflow meets the throughflow, causing separation of the boundary layer at the pump casing. Upon exposure to adverse pressure gradients in the aft part of the passage, at 0.65–0.7 chord fraction in the present conditions, the TLV bursts into a broad turbulent array of widely distributed vortex filaments.     

The visualization of the flowfield about a delta wing with spanwise blowing

This paper describes a nonintrusive method for the visualization of the flow about a delta wing with spanwise blowing jets, based on the schlieren technique. The effects of the jet/leading-edge vortex interference are visualized by using both air and helium for the jets. The visualization of the leading-edge vortex trajectories and their breakdown, as well as the influence of the jets on them is achieved by spanwise blowing of air. The visualization of the jets' paths and the effects of the leading-edge vortices on these paths is achieved by spanwise blowing of helium.     

Fluid–structure interaction of a splitter plate in a convergent channel

The fluid–structure interaction (FSI) of a splitter plate in a convergent channel flow is studied by measuring both the flow field and the plate vibration. Particle Image Velocimetry (PIV) measurements show that the wake generated by the plate is characterized by cellular vortex shedding. Mean and RMS velocities presented in the plane normal to the main flow direction visualize clearly the cellular structure and related secondary flows. To evaluate the energy and spatial organization of the vortex shedding, spectral and correlation estimation methods are adapted to the PIV data. By presenting the spanwise variation of the streamwise spectra along the trailing edge, the nature of the cellular vortex shedding becomes evident. 2D space-correlation function reveals that the shedding in two neighboring cells occurs in a 180-degree phase shift. The vibration of the plate is studied with Digital Imaging (DI) and Laser Vibrometer (LV). The DI is based on images measured by the PIV system. An image-processing algorithm is used to detect the plate tip location and velocity simultaneously with the estimation of the fluid velocity field. The LV is used for the time-resolved measurement of the plate vibration. The results show that the plate vibrates in a very distinct mode characterized by a spanwise standing wave along the plate-trailing edge. This mode, in turn, causes the cellular vortex shedding.     

Vortex breakdown and its control on delta wings

In order to investigate the breakdown of vortices generated by the leading edge of delta wings, LDA-measurements have been performed in the flow on the suction side of a delta wing of aspect ratio A = 2. The measurements describe the growth of the vortex along the leading edge and reveal a certain radial structure upstream of the breakdown point. Moreover they shed light on the mechanism responsible for the onset of vortex breakdown on the suction side of a wing.

The occurrence of the breakdown phenomenon on a delta wing may be prevented or at least retarded by the use of spanwise blowing jets. The interaction of vortex and jets giving rise to these effects will also be discussed with the help of measured velocity profiles.     

Experimental study of three-dimensional vortex structures in translating and rotating plates

Motivated by the unsteady force generation of flying animals, vortex formation and vorticity transport processes around small aspect-ratio translating and rotating plates with a high angle of attack are investigated. Defocusing Digital Particle Image Velocimetry was employed to explore the structure and dynamics of the vortex generated by the plates. For both translating and rotating cases, we observe the presence of a spanwise flow over the plate and the consequent effect of vorticity transport due to the tilting of the leading-edge vortex. While the spanwise flow is confined inside the leading-edge vortex for the translating case, it is widely present over the plate and the wake region of the rotating case. The distribution of the spanwise flow is a prominent distinction between the vortex structures of these two cases. As the Reynolds number decreases, due to the increase in viscosity, the leading-edge and tip vortices tend to spread inside the area swept by the rotating plate. The different vorticity distributions of the low and high Reynolds number cases are consistent with the difference in measured lift forces, which is confirmed using the vorticity moment theory.     

Large aerodynamic forces on a sweeping wing at low Reynolds number

The aerodynamic forces and flow structure of a model insect wing is studied by solving the Navier-Stokes equations numerically. After an initial start from rest, the wing is made to execute an azimuthal rotation (sweeping) at a large angle of attack and constant angular velocity. The Reynolds number (Re) considered in the present note is 480 (Re is based on the mean chord length of the wing and the speed at 60% wing length from the wing root). During the constant-speed sweeping motion, the stall is absent and large and approximately constant lift and drag coefficients can be maintained. The mechanism for the absence of the stall or the maintenance of large aerodynamic force coefficients is as follows. Soon after the initial start, a vortex ring, which consists of the leading-edge vortex (LEV), the starting vortex, and the two wing-tip vortices, is formed in the wake of the wing. During the subsequent motion of the wing, a base-to-tip spanwise flow converts the vorticity in the LEV to the wing tip and the LEV keeps an approximately constant strength. This prevents the LEV from shedding. As a result, the size of the vortex ring increases approximately linearly with time, resulting in an approximately constant time rate of the first moment of vorticity, or approximately constant lift and drag coefficients. The variation of the relative velocity along the wing span causes a pressure gradient along the wingspan. The base-to-tip spanwise flow is mainly maintained by the pressure-gradient force. The project supported by the National Natural Science Foundation of China (10232010)     

Vortical structures on a flapping wing

A wing in the form of a rectangular flat plate is subjected to periodic flapping motion. Space–time imaging provides quantitative representations of the flow structure along the wing. Regions of spanwise flow exist along the wing surface; and depending on the location along the span, the flow is either toward or away from the tip of the wing. Onset and development of large-scale, streamwise-oriented vortical structures occur at locations inboard of the tip of the wing, and they can attain values of circulation of the order of one-half the circulation of the tip vortex. Time-shifted images indicate that these streamwise vortical structures persist over a major share of the wing chord. Space–time volume constructions define the form and duration of these structures, relative to the tip vortex.     

Three-component velocity field measurements of propeller wake using a stereoscopic PIV technique

A stereoscopic PIV (Particle Image Velocimetry) technique was used to measure the three-dimensional flow structure of the turbulent wake behind a marine propeller with five blades. The out-of-plane velocity component was determined using two CCD cameras with an angular displacement configuration. Four hundred instantaneous velocity fields were measured for each of four different blade phases, and ensemble averaged in order to find the spatial evolution of the propeller wake in the region from the trailing edge up to one propeller diameter (D) downstream. The influence of propeller loading conditions on the wake structure was also investigated by measuring the velocity fields at three advance ratios (J=0.59, 0.72 and 0.88). The phase-averaged velocity fields revealed that a viscous wake formed by the boundary layers developed along the blade surfaces. Tip vortices were generated periodically and the slipstream contracted in the near-wake region. The out-of-plane velocity component and strain rate had large values at the locations of the tip and trailing vortices. As the flow moved downstream, the turbulence intensity, the strength of the tip vortices, and the magnitude of the out-of-plane velocity component at trailing vortices all decreased due to effects such as viscous dissipation, turbulence diffusion, and blade-to-blade interaction.     

Effect of perforation on flow past a conic cylinder at <Emphasis Type="Italic">Re</Emphasis> = 100: vortex-shedding pattern and force history

The flow past a circular-section cylinder with a conic shroud perforated with four holes at the peak was simulated numerically at \(Re=100\), considering two factors, viz. the angle of attack and the diameter of the holes. The effects of the perforated conic shroud on the vortex shedding pattern in the near wake was mainly investigated, as well as the time history of the drag and lift forces. In the investigated parameter space, three flow regimes were generally identified, corresponding to weak, moderate, and strong disturbance effects. In regime I, the wake can mainly be described by alternately shedding Kármán or Kármán-like vortices. In regime II, the spanwise vortices are obviously disturbed along the span due to the appearance of additional vorticity components and their interactions with the spanwise vortices, but still shed in synchronization along the spanwise direction. In regime III, the typical Kármán vortices partially or totally disappear, and some new vortex shedding patterns appear, such as \(\Omega \)-type, obliquely shedding, and crossed spanwise vortices with opposite sign. Corresponding to these complex vortex shedding patterns in the near wake, the fluid forces no longer oscillate regularly at a single vortex shedding frequency, but rather with a lower modulation frequency and multiple amplitudes. An overview of these flow regimes is presented.     

Effect of film cooling on the aerodynamic performance of an airfoil

The effect of film cooling on the aerodynamic performance of turbine blades is becoming increasingly important as the gas turbine operating temperature is being increased in order to increase the performance. The current paper investigates the effect of blowing ratio on the aerodynamic losses of a symmetric airfoil by pressure measurements and Particle Image Velocimetry (PIV). The test model features 4 rows of holes located on the suction side at 5%, 10%, 15% and 50% of the chord length. The Reynolds number based on the airfoil chord is 1.2 × 10⁵. Experiments are performed by varying the location of air injection, the angle of attack, and the mainstream velocity. The coolant air is injected at ambient temperature and the blowing ratio is varied from 0 to 1.91. It is observed that the losses due to film cooling increase with blowing ratio of 0 to 0.48, and the wake is shifted towards the suction side. Conversely, the aerodynamic losses decrease when the blowing ratio is increased further from 0.64 to 1.91. This trend has been observed for all the experimental configurations. The effect of blowing ratio on flow separation is investigated with the time-averaged velocity fields obtained from PIV measurements. It is observed that low blowing ratios, the separation point shifts upstream and at high blowing ratios the ejected coolant energizes the flow and delays separation. The pressure field around the airfoil is reconstructed from the integration of the Poisson equation based on the PIV velocity fields. The experimental results can be used for validation of numerical models for predicting losses due to film cooling.     

Flow structure on a rotating plate

The flow structure on a rotating plate of low aspect ratio is characterized well after the onset of motion, such that transient effects are not significant, and only centripetal and Coriolis accelerations are present. Patterns of vorticity, velocity contours, and streamline topology are determined via quantitative imaging, in order to characterize the leading-edge vortex in relation to the overall flow structure. A stable leading-edge vortex is maintained over effective angles of attack from 30° to 75°, and at each angle of attack, its sectional structure at midspan is relatively insensitive to Reynolds number over the range from 3,600 to 14,500. The streamline topology, vorticity distribution, and circulation of the leading-edge vortex are determined as a function of angle of attack, and related to the velocity field oriented toward, and extending along, the leeward surface of the plate. The structure of the leading-edge vortex is classified into basic regimes along the span of the plate. Images of these regimes are complemented by patterns on crossflow planes, which indicate the influence of root and tip swirl, and spanwise flow along the leeward surface of the plate. Comparison with the equivalent of the purely translating plate, which does not induce the foregoing flow structure, further clarifies the effects of rotation.     

An experimental study of a spanwise structure around a reattachment region of a two-dimensional backward-facing step

 The flow field downstream of a two-dimensional backward-facing step is usually assumed to be independent of the direction along the span of the step. This assumption is made even though it is well known that the flow exhibits a three-dimensional vortex structure. This state of affairs is no doubt due to the lack of detailed information concerning the characteristics of the vortex structure. In this paper, we report our investigations of the flow structure around a reattachment region using an ultrasound velocity profiler to measure the spanwise velocity component as a function of the spanwise coordinate and time. The flow field is found to be very complex both in space and time. The low-frequency component of the spanwise velocity fluctuation becomes dominant in the near-wall region, with peaks in the power spectrum at frequencies fh/Uc=0.05 and fh/Uc=0.012. Using multiple ultrasound transducers, we also find that a streamwise vortex exists in the flow. Received: 20 March 2000 / Accepted: 15 January 2001 Published online: 29 November 2001     

Numerical study on film cooling effectiveness from shaped and crescent holes

This paper presents a comparative numerical investigation on film cooling from a row of holes injected at 35° on a flat plate with three film cooling configurations, including cylindrical hole, 15° forward diffused shaped hole, and new crescent hole. All simulations are conducted at blowing ratio of 0.6 and 1.25, length-to-diameter ratio of four and pitch-to-diameter ratio of three. Computational solutions of the steady, Reynolds averaged Navier–Stokes equations are obtained using a finite volume method. Previous successful application of a two-layer turbulence model to cylindrical hole is extended to predict film cooling for the different hole geometries. It has been found that the film cooling effectiveness of cylindrical holes obviously declined along with increasing the blowing ratio. While the forward diffused shaped hole presents a marked improvement, with a higher effectiveness at the lateral area between adjacent holes. By comparison, the crescent hole exhibits the highest film cooling effectiveness among the three configurations both in spanwise and streamwise especially downstream of the intersection of the two holes. Also, the crescent hole can restrain the vortex intensity, and then enhance the film cooling effectiveness.     

A broad reconsideration of anti-vortex film cooling method using numerical optimization and an improved heat-flux model

This paper represents the detailed results of an evolutionary optimization framework towards the exploration of vortex mechanisms leading to effective anti-vortex film cooling. In this regards, several arrangements of triple cooling holes were studied on flat and curved geometries using differential-evolution optimization algorithm and a modified Reynolds-stress based flow solver. Depending on the flow and geometric parameters, four distinct types of vortex interaction with different cooling mechanisms were identified. The vortex-trapping mechanism, observed in the optimized upstream arrangement acts through imposing a mild downwash over the main counter-rotating vortex pair and provides the best cooling effectiveness for the low injection angle (less than 30°) cases. The vortex-suppression and -balancing are the optimal possible solutions of the adjacent arrangement. The latter is the classic well-known type of anti-vortex cooling, while the former provides a sudden strong controlling potential for high blowing ratios (higher than 1.0) and high injection angle film cooling. For the non-flat surfaces the triple holes effectively perform up to blowing-ratio of 2.0. However, the reverse-vortex-trapping mechanism occurring in the downstream arrangement is recommended for convex surfaces, while the adjacent arrangement is the choice for concave regions. In general, there is a possibility of reducing the coolant consumption about 30% through increasing the pitch-to-diameter ratio, while the values of cooling-effectiveness still remain in an acceptable range.     

In-hole and mainflow velocity measurements of low-momentum jets in crossflow emanating from short holes

Discrete hole film cooling utilizes jet-in-crossflow geometry where the jet is supplied through a short hole which may be pitched relative to the main flow. Typically, the velocity ratio is near one. Under these conditions, the mean flow structure of the jet/mainstream interaction may be strongly affected by the characteristics of the flow within the hole. Magnetic resonance velocimetry (MRV) is used to measure the 3-dimensional mean velocity field for various jets in crossflow with short holes of varied inclination angles and blowing ratios typically of gas turbine applications. Novel measurements of the flow within inclined feed holes are captured using MRV. Secondary flows within the hole are found to be strongly dependent on the inclination of the hole. The traditional counter-rotating vortex pair is observed in the mainstream, as well as high levels of wall-normal vorticity. The 3D vorticity field is used to modify traditional jet-in-crossflow vortex ring theory to apply to low-momentum jets which remain attached to the ejection surface.