Stability Enhancement With Self-Recirculating Injection in Axial Flow Compressor

A low speed axial compressor with casing treatment of axial slots was numerically investigated. Time-accurate numerical calculations were performed to simulate unsteady flow in the rotor tip region and the effects of casing treatment on the flow. Since the compressor rotor had a large tip clearance, it was found that the tip leakage flow had an inherent unsteady feature that was not associated with rotor rotation. The unsteadiness of the tip leakage flow was induced by changes in the blade loading due to the pressure distribution formed by the tip leakage flow. This characteristic is called rotating instability or self-induced unsteadiness. The frequency of the flow oscillation was found to decrease as the flow rate was reduced. On the other hand, as expected, the operating range was improved by casing treatment, as shown by calculations in good agreement with the experimentally measured data. The unsteadiness of the tip leakage flow was alleviated by the casing treatment. The interaction between the flow in the tip region and the re-circulated flow through the axial slots was observed in detail. The removal and injection of flow through the axial slots were responsible not only for the extension of the operating range but also for the alleviation of the unsteadiness. Analyses of instantaneous flow fields explained the mechanism of the interaction between the casing treatment and the unsteady oscillation of the tip leakage flow. Furthermore, the effects of changes in the amount of re-circulation and the location of the removal and injection flow on the unsteadiness of the tip leakage flow were examined.

Effect of inlet radial distortion on aerodynamic stability in a multi-stage axial flow compressor

As the first step of the study of cavitation in the tip leakage flow of turbomachinery, experimental and numerical studies were carried out to investigate the characteristics of cavitation in the 2-D unsteady tip leakage flow. The numerical method was constructed by combining the vortex method with taking into account the effects of cavity growth. Three types of cavity models are examined. It is shown that one of them simulates the location and the size of the cavity fairly well. Comparisons with experiments are made for three values of initial pressure and for three values of tip clearance. The model can predict those effects and it is expected that the present model is useful for the estimation of cavitation characteristics of 3-D tip leakage flow.

Micro tip injection is a stall control technique in which engineers inject small-flowrate yet high-velocity jets into the tip region of axial compressor rotor blades to extend the stable operation range, while keeping the entire compressor characteristic line nearly unchanged. This paper summarizes the related research performed in Chinese Academy of Sciences in the past several years, ranging from understanding the fundamental mechanisms to demonstrating the technology in a laboratory compressor. A brief review in tip injection research indicates that for each compressor, there exists a critical injection momentum ratio, below which the injection flows may work differently from those over the critical value. Detailed casing unsteady static pressure measurements and the related sophisticated data processing proved the existence of self-induced unsteady tip leakage flow, which was also confirmed by unsteady CFD simulations. It is showed that this kind of unsteadiness in tip leakage flow might be responsible to make the auto-correlation coefficient drop significantly as the compressor operated at lower flowrate close to stall limit. This warning signal, which was first successfully used by other research groups with little care taken on its flow mechanism, appears much earlier than the classical modal-wave or spike stall precursors. Micro tip injection was effective if it interacts with the unsteady tip leakage flow long before the stall precursor emerges. As a technology demonstration, a DSP board with a build-in auto-correlation early-stall-warning algorithm and an on-off controller of injection valve worked well to drive a micro tip injection system and successfully delayed the stall in a low-speed compressor with much less jet flows than the continuous micro tip injection.

A study has been conducted, using an unsteady three-dimensional Reynolds-averaged Navier-Stokes simulation, to define the effect of impeller-diffuser interaction on the performance of a centrifugal compressor stage. The principal finding from the study was that the most influential aspect of this unsteady interaction was the effect on impeller tip leakage flow. In particular, the unsteadiness due to the upstream potential effect of the diffuser vanes led to larger viscous losses associated with the impeller tip leakage flow. The consequent changes at the impeller exit with increasing interaction were identified as reduced slip, reduced blockage, and increased loss. The first two were beneficial to pressure rise while the third one was detrimental. The magnitudes of the effects were examined using different impeller-diffuser spacings and it was shown that there was an optimal radial gap size for maximum impeller pressure rise. The physical mechanism was also determined: when the diffuser was placed closer to the impeller than the optimum, increased loss overcame the benefits of reduced slip and blockage. The findings provide a rigorous explanation for experimental observations made on centrifugal compressors. The success of a simple flow model in capturing the pressure rise trend indicated that although the changes in loss, blockage and slip were due largely to unsteadiness, the consequent impacts on performance were mainly one-dimensional. The influence of flow unsteadiness on diffuser performance was found to be less important than the upstream effect, by a factor of seven in terms of stage pressure rise in the present geometry. It is thus concluded that the beneficial effects of impeller-diffuser interaction on overall stage performance come mainly from the reduced blockage and reduced slip associated with the unsteady tip leakage flow in the impeller.

The effects of water ingestion on the performance of an axial flow compressor are experimentally studied with and without endwall treatment. The background to the work is derived from the assessment of airworthiness for an aero-engine. The stability-enhancing effects with endwall treatments under rain ingestion are not previously known. Moreover, all the endwall treatments are designed under dry air conditions in the compressor. Water ingestion at 3% and 5% relative to the design mass flow proposed in the airworthiness standard are applied to initially investigate the effects on the performance under smooth casing (SC). Results show that the water ingestions are mainly located near the casing wall after they move through the rotor blade row. The pressure rise coefficient increases, efficiency declines, and torque increases under the proposed water ingestion. The increase of the inlet water increases the thickness of the water film downstream the rotor blade row and aggravates the adverse effects on the performances. Subsequently, three endwall treatments, namely circumferential grooves, axial slots, and hybrid slots–grooves, are tested with and without water ingestion. Compared with no water ingestion, the circumferential grooves basically have no resistance to the water ingestion. The axial slots best prevent the drop of the pressure rise coefficient induced by water ingestion, and hybrid slots–grooves are the second-best place owing to the contribution of the front axial slots. Therefore, the hybrid slots–grooves can not only extend the stall margin with less efficiency penalty compared with axial slots, but also prevent rain ingestion from worsening the compressor performance.

This paper describes the impact of axial slots on the flow field in a transonic rotor blade row. The presented results are completely based on time-accurate 3-dimensional numerical simulations of a high pressure compressor front stage with and without casing treatment. Two different axial positions of a casing treatment consisting of axial slots were tested for their impact on flow stability and efficiency. The first tested position (configuration 1) was chosen in a conventional way. The slots extend approximately from the leading up to the trailing edge of the rotor blades. As expected, the simulations of the compressor stage with this configuration showed a significant increase in flow stability near surge compared to the solid wall case. However, a non negligible decrease in efficiency is also observed. Analyses of flow interactions between casing treatment and rotor blade rows under transonic conditions lead to the general conclusion that the stabilizing effect of circumferential grooves or axial slots mainly results from their impact on the tip leakage flow and its resulting vortex. A characteristic vortex inside the slots is observed in the simulations with the conventionally positioned casing treatment. This vortex removes fluid out of downstream parts of the blade passage and feeds it back into the main flow further upstream. The resulting impact on the tip leakage flow is responsible for the increased flow stability. However, the interaction between the configuration 1 casing treatment flow and the blade passage flow results in a significant relocation of the blade passage shock in the downstream direction. This explains the observed decrease in compressor efficiency. A second slot position (configuration 2) was tested with the objective to improve compressor efficiency. The casing treatment was shifted upstream, so that only 25% of the blade chord remained under the slots. The simulations carried out demonstrate that this shift positively effects the resulting efficiency, but maintains the increased level of flow stability. A time-accurate analysis of the flow shows clearly that the modified casing treatment stabilizes the tip leakage vortex and reduces the influence on the flow inside the blade passage.

To extend the current understanding of the circumferential groove casing suction applied to a counter-rotating axial flow compressor, the impact of different axial locations of the circumferential suction groove on the characteristics of the tip leakage flow (TLF) and the corresponding physical mechanisms producing the stability enhancement have been studied based on validated numerical simulations. The results show that the optimal location for the suction groove is at around 20% axial chord, which demonstrated a high potential for reducing additional stall mass flow coefficient with about 8.4% increment in the stall margin. After the casing suction groove was applied, the interface between the incoming main flow and TLF was pushed significantly downstream in the second rotor. The blade loading in the region below the groove, the tip leakage flow angle and the reversed axial momentum flux injected into main flow passage through the tip gap were all reduced, which contributed to the stall margin improvement. Detailed analysis of the tip leakage flow structures showed that the TLF originating from different chord locations played different roles in the stall inception process. It was found to be more effective to improve stall margin and adiabatic efficiency by controlling the front part of the TLF, which was most sensitive.

A three-dimensional computation was conducted to understand effects of the inlet boundary layer thickness on the internal flow and the loss characteristics in a low-speed axial compressor operating at the design condition (φ = 85%) and near stall condition (φ = 65%). At the design condition, independent of the inlet boundary layer thickness, flows in the axial compressor show similar characteristics such as the pressure distribution, size of hub corner-stall, tip leakage flow trajectory, limiting streamlines on the blade suction surface, etc. But, as the load is increased, for the thick inlet boundary layer at hub and casing, the hub corner stall grows to make a large separation region between the hub and suction surface, and the tip leakage flow is more vortical than that observed in the case with thin inlet boundary layer and has the critical point where the trajectory of the tip leakage flow is suddenly turned to the downstream. For the thin inlet boundary layer, the hub corner stall decays to form the thick boundary layer from hub to midspan on the suction surface owing to the blockage of the tip leakage flow and the tip leakage flow leans to the circumferential direction more than at the design condition. In addition to these, the severe reverse flow, induced by both boundary layers on the blade surface and the tip leakage flow, can be found to act as the blockage of flows near the casing, resulting in a heavy loss. As a result of these differences of the internal flow made by the different inlet boundary layer thickness, the spanwise distribution of the total loss is changed dramatically. At the design condition, total pressure losses for two different boundary layers are almost alike in the core flow region but the larger losses are generated at both hub and tip when the inlet boundary layer is thin. At the near stall condition, however, total loss for thick inlet boundary layer is found to be greater than that for thin inlet boundary layer on most of the span except the region near the hub and casing. In order to analyze effects of inlet boundary layer thickness on total loss in detail, total loss is scrutinized through three major loss categories available in a subsonic axial compressor such as profile loss, tip leakage loss and endwall loss.

Numerical simulations were carried out with the purpose of investigating the effect of applying circumferential grooves at axial compressor casing passive wall treatment to enhance the stall margin and change the tip leakage flow. The tip leakage flow is pointed out as one of the main contributors to stall inception in axial compressors. Hence, it is of major importance to treat appropriately the flow in this region. Circumferential grooves have shown a good performance in enhancing the stall margin in previous researches by changing the flow path in the tip clearance region. In this work, a passive wall treatment with four circumferential grooves was applied in the transonic axial compressor NASA Rotor 37. Its effect on the axial compressor performance and the flow in the tip clearance region was analyzed and set against the results attained for the smooth wall case. A 2.63% increase in the operational range of the axial compressor running at 100%N, was achieved, when compared with the original smooth wall casing configuration. The grooves installed at compressor casing, causes an increase in the flow entropy generation due to the high viscous effects in this gap region, between the rotor tip surface and casing with grooves. These viscous effects cause a drop in the turbomachine efficiency. For the grooves configurations used in this work, an efficiency drop of 0.7% was observed, compared with the original smooth wall. All the simulations were performed based on 3D turbulent flow calculations using Reynolds Averaged Navier-Stokes equations, and the flow eddy viscosity was determined using the two-equation SST turbulence model. The details of the grooves geometrical dimensions and its implementation are described in the paper.

Pneumatic probes are widely used to measure flow properties in turbomachinery, including downstream of rotor blade rows. This paper investigates the aerodynamic interaction between a row of compressor blades and a downstream cylindrical probe, using unsteady, three-dimensional simulations. The study demonstrates the probe influence on rotor performance as a function of probe size and rotor proximity, identifies the physical mechanisms driving rotor-probe interactions, and quantifies the resulting probe measurement errors. A probe downstream of a rotor row causes each blade passage to deviate periodically from the steady, axi-symmetric characteristic, as they pass the probe. In this study, the largest rotor disturbance occurs for a probe of diameter 14% blade chord, located 30% axial chord downstream of the rotor trailing edges. The flow coefficient, ϕ, moving with the rotor passage, varies between −10.4% and +5.5% compared to a case with no downstream probe. The total-to-total pressure rise coefficient, ѱtt, increases by up to 7.7%. These rotor-probe interaction effects are driven by the unsteady response of the rotor passage to the potential field of the probe. In the stationary frame, measurement error occurs because the probe is exposed to the disturbed flow field and this leads to maximum average errors of Δϕ = −15.1% and Δѱtt = 8%. The magnitude of the rotor disturbance decreases as the probe is moved away from the trailing edge of the blades and when probe size is reduced. Reducing probe size close to the blade row reduces the rotor disturbance more than moving a large probe downstream. Including a stator blade row downstream of the probe does not have a large impact on the sensitivity to probe size and proximity and the physical mechanisms are unchanged.

Aerothermal behavior and performance of turbine blade squealer tip has been extensively studied, but previously mainly confined to very low speed flows. Some recent research efforts have revealed some distinctive tip heat transfer as well as aerodynamic characteristics in a transonic flow, many of which are confined to a linear cascade setting. There is a need to identify driving mechanisms for such configurations and to quantify corresponding performance characteristics in general, and in the process address how those drivers would relate to some specific issues of interest such as: 1) the differences between a cascade and a rotor tip, thus the impact of relative casing movement? 2) The difference between a transonic and a high subsonic flow condition? 3) the difference between a squealer tip and flat tip? In this study, the vortical structures and the loss mechanism of the Over Tip Leakage (OTL) flow in a high-subsonic stage and a transonic stage are analyzed using a proposed vortex model and Denton’s loss model. The two stages are specifically designed to work in the same Reynolds number with the efficiency comparable to the one predicted in the Smith’s chart. Two counter-rotating vortices, the Pressure-side Separation Vortex (PSV) and the Casing-driven Cavity Vortex (CCV), are identified as key drivers as inside a squealer cavity strongly impacting aerothermal performance (OTL mass flow and the heat transfer distribution). The cavity flow pattern results from the balancing between the CCV primarily driven by the relative casing movement, and the PSV originating from the flow separation around the pressure side rim. Consequently a detoured flow path is created within the squealer cavity when the leakage flow separating on the pressure side is entrained by the two vortices to the squealer bath and finally leaves the cavity along the suction side of the squealer wall. The entrainment of the leakage flow induces a signature of a high HTC strip on the squealer floor. Without the relative casing movement, the CCV is absent in the majority of the cavity so that the leakage flow can directly cross the tip which increases the leakage mass flow rate and reduce the HTC value on the cavity floor. Due to the distinct difference in vortical structures, the change of the HTC with and without relative casing movement can be interpreted by the vortex model. The vortex model is also able to explain the variation of the leakage mass flow rate and the HTC contours with the increasing squealer height. From the best knowledge of the authors, this is the first-of-its kind effort in using in the cavity vortices to systematically and quantitatively study the effect of the relative casing movement and the aerothermal performance of the tip with varying squealer height. In counting the aero losses for the two stages, three sources are attributed to the total loss generation: the baseline loss which assumes there is no leakage, the in-tip loss which is generated within the tip gap and the out-tip loss which accounts for the mixing loss between the OTL flow and the main passage flow. The out-tip loss is found to be the major source of the loss caused by the tip leakage flow. Denton’s loss model is used to calculate the out-tip loss quantitatively. The model predicted loss agrees in trend with the one calculated from the CFD results. According to the model, the out-tip loss is related to the main flow Mach number, the OTL mass flow rate and the velocity difference between the main flow and the OTL flow. From the distribution of these terms along the tip, the loss is mainly generated at the rear portion of the tip. The reduction of the leakage mass flow rate by the squealer is mainly at the zones where the CCV and the PSV are leaving the squealer cavity. The present analysis and results indicate that the squealer in the transonic stage tends to be less effective in reducing the out-tip loss compared to the subsonic one. The region of the tip where the CCV has already left the cavity with the PSV remaining shows the highest entropy generation rate, due to locally both the high mainstream velocity and the high OTL mass flow given the PSV being left as the only sealing vortex there.

The study of flow field characteristics inside the tip clearance is essential for understanding tip leakage flow (TLF) generation mechanisms. This paper investigates the TLF flow field in a single-stage axial compressor using a particle image velocimetry visualization system, alongside proper orthogonal decomposition and omega vortex identification methods, to analyze the flow at the mid-axial cross section under varying rotational speeds. The results reveal that the time-averaged flow field inside the clearance can be divided into four distinct regions, largely unaffected by speed variations: the high-speed region at the blade leading edge, the diverting and steering region of TLF, the TLF and main stream convergence region, and the trailing edge low-velocity region. Notably, two distinct directions of TLF are observed. At the blade leading edge, TLF flows directly from the upstream field to the lower blade, with a maximum velocity of approximately 2.2 times that of the main flow. Between 30% and 40% of the blade length, radial vortices dominate the diverting and steering of the main flow, forming a precessing vortex structure. Subsequently, TLF from the pressure side accelerates upstream, intersecting with the main flow to form a low-velocity tip leakage vortex region, obstructing the upstream flow. As rotational speed increases, the diverging and steering region shifts downward, and the disturbances in the internal and upstream flow fields further intensify. Furthermore, TLF from the downstream field bypassing the tip is observed between blades. These findings offer valuable data to improve understanding and optimize control strategies of TLF.

This paper presents a study of the three-dimensional e owe eld within the blade rows of a single-stage highpressure axial turbine (low-speed, large-scale ). Measurements have been performed in the stationary and rotating frames of reference. Time-mean data have been obtained using e ve-hole pneumatic probes. The transport mechanisms of the stator wake and passage vortices through the rotor blade row have been studied using smoke e ow visualization. Furthermore, unsteady measurements have been carried out using a three-axis hot wire. Steady and unsteady numerical simulations have been performed using a structured three-dimensional Navier ‐Stokes solver to further understand the blade-row interactions. The transport of the stator viscous e ow through the rotor blade-row is described. The rotor passage vortices are affected by the transport of the stator secondary e ow. It is observed that the stator secondary e ow vortices are convected through the downstream rotor blade-row in a similar but not identical way to the wake. At the hub the kinematic interaction between the stator and the rotor passage vorticeshas two effects. First,thesuction sideleg of the statorpassagevortex isdisplaced radially upwards over the developing rotor passage vortex at the hub. Additionally, the pressure side leg of the stator passage vortex is entrained into the rotor passage vortex. The predicted e owe eld was interrogated from the perspective of loss production. The contribution of the unsteady e ow to the stage loss has been evaluated using unsteady numerical simulations. The effect of stator viscous e ow transport on the rotor e ow angles is also discussed in brief. Finally, a kinematic model is proposed for the transport of the secondary-e ow vortices in the downstream blade-row based on the understanding obtained from the measurements and numerical simulations.

In this paper, an unsteady tip leakage flow phenomenon is identified and investigated in a centrifugal compressor with a vaneless diffuser at near-stall conditions. This phenomenon is associated with the inception of a rotating instability in the compressor. The study is based on numerical simulations which are supported by experimental measurements. The study confirms that the unstable flow is primarily driven by the Kelvin-Helmholtz instability inherent in the shear layer between the main-stream flow and the tip leakage flow. The shear layer instability induces large scale vortex shedding which propagates circumferentially, leading to pressure perturbations with short wavelength and high amplitude which rotates at about half of the blade speed. The shed vortex is also found to interact with the main blade leading edge, resulting in the reduction of the blade loading identified in the experiment. The paper also reveals that the downstream volute imposes a long wavelength circumferential non-uniform back pressure at the impeller exit, which is found to either excite or supress the vortex shedding within the blade passage depending on the relative circumferential position of the impeller and the volutes.