News

Jul 01, 2025

Experimental and numerical investigation of endwall flow control through air injection in a contra rotating fan | Scientific Reports

Scientific Reports volume 15, Article number: 21080 (2025) Cite this article This study reports on an experimental and numerical investigation of tip leakage flow control and its effects on the

Scientific Reports volume 15, Article number: 21080 (2025) Cite this article

This study reports on an experimental and numerical investigation of tip leakage flow control and its effects on the endwall flow structure and stall evolution of a contra-rotating fan. The control method involves the annular injection of high-momentum fluid in the axial direction, upstream of the front rotor. Five-hole probe measurements have been performed at different blade span locations to obtain the hub-to-shroud velocity distribution and also detect stall. Half-annulus transient simulations have been further conducted and validated against experimental tests. Two injection mass flow rates, 1.6% and 3.2% of the main mass flow rate, have been evaluated. Fast Fourier Transform and wavelet transform analyses have been applied to the experimental data. These analyses show that stall disturbances are intensified near the endwall, indicating the tip-critical nature of the studied fan. Under stall conditions, a modal wave in the tip region has been identified, rotating at 40% of the rotor speed. Numerical contours confirm the presence of circumferential disturbances. Also, the results reveal that rotor-2 is responsible for stall initiation. The findings further show that a 3.2% injection has sufficient momentum to eliminate the tip leakage vortex in the endwall flow. With 3.2% tip air injection, the total pressure rise coefficient and the stable operating range increased by 21% and 59%, respectively.

Contra-rotating fans/compressors are of interest in the turbomachinery field and the gas turbine industry. They can enhance the performance, increase the mass flow rate, and also reduce the weight compared to conventional fans/compressors. The contra-rotating fan/compressor was first introduced by Lesley1,2 between 1930 and 1940 and has been followed by several researchers. Finding the effective parameters of contra-rotating fans/compressors, recognizing the type of instabilities and their behavior, and also improving their performance are the most important attempts of researchers in this field. The effects of axial spacing and rotor speed combinations were studied by Mistry and Pradeep3. It was found that the higher rotational speed of the second rotor provides more suction effect and improves the whole performance. Also, axial spacing equal to 0.9 of the chord was found to be optimal. However, the study by Nouri et al.4 showed that the axial spacing does not have any considerable effect on the stage performance. In another investigation reported by Mistry and Pradeep5, the performance of a contra-rotating fan under a hub and tip strong inflow distortion was studied. Results showed lower performance compared to the clean inflow. Furthermore, the flow direction after each distortion plane and its effect on performance were analysed. Wang et al.6 studied the effect of tip clearance on the performance of a contra-rotating fan. It was shown that total pressure rise and efficiency are decreased by increasing tip clearance. In addition, the tip clearance of the first rotor was found to be more critical for the performance of the stage.

Flow instabilities (including stall and surge) limit the performance of both axial and centrifugal fans/compressors. Several researchers tried to identify the type of flow instabilities and analyze their behavior in conventional compressors7,8,9,10,11. Flow instabilities have also been studied in contra-rotating fans and compressors in both experimental and numerical investigations12,13,14,15,16. In the study conducted by Toge and Pradeep12, casing wall sensors were used in order to recognize stall inception. Results showed that for both rotors, stall originated through long-length scale disturbances. Also, strong modal disturbances with a frequency equivalent to 25–80% of that of the first rotor were reported. Wang et al.13, investigated stall inception in a contra-rotating fan through high-response pressure sensors mounted at the leading edge of each rotor as well as the trailing edge of the second rotor. A rotating stall was incepted from the tip region of the second rotor with a speed of 35% of the rotor. In the experimental investigation carried out by Manas and Pradeep14, stall characteristics have been studied under distorted and clean inflow in a contra-rotating fan. The endwall of the front rotor was found to be the most critical region. Inlet distortion caused a pressure rise drop (most felt by rotor-1) and an efficiency decrement (sensed by rotor-2). Furthermore, a fluid phenomenon was observed on the blade surface, which has been shed in fully stall condition and rotates much slower than the rotor. Khaleghi et al.15 developed a model to predict the transient stall cell speed and its acceleration. Furthermore, they introduced the parameters that affect the stall cell speed. In the study carried out by Shahriyari et al.16 a model was developed based on Moore-Greitzer theory17, which can predict the post-stall behaviour of low-speed contra-rotating fans.

Many researchers focused on the stall-control methods, including casing treatment and endwall injection, to improve the fan/compressor stability and performance. The casing treatment method has been investigated in conventional compressor stages by several researchers18,19,20,21,22. Nonetheless, fewer studies have been reported on contra-rotating stages. Mao et al.23,24, performed a parametric study to find the best location of a circumferential groove to enhance the stability of a contra-rotating fan. The optimal location of the groove was found to be near the blade’s leading edge. The performance enhancement was achieved due to the blade tip unloading effect, redirection effect, and compound effect of suction-injection caused by the groove. In the investigation reported by Guo et al.25, a self-recirculation pipe has been designed to enhance the stability range of a contra-rotating fan. The optimum location of the suction port has been found, which improves the stability by 7.73%. Dong et al.26, evaluated the effect of microperforated-panel (MPP) casing treatment in a low-speed small contra-rotating fan. It has been shown that suction and injection via sub-millimetre holes on the treated casing postpone tip leakage vortex formation and improve rear rotor inflow. It is further shown that pressure unsteadiness has decreased by 14.5%. In the field of active stall control using endwall injection, many investigations have been carried out in common fans/compressors to evaluate the injection effect on stall suppression27,28,29, find the best injection velocity and mass flow rate29,30, and also optimize the injector location and configuration angle31. However, to the best of the author’s knowledge, the endwall injection has not been studied on contra-rotating fans/compressors.

In the present study, the effect of endwall injection as the flow control method is investigated on a contra-rotating fan. Total pressure rise coefficient, rotor efficiency, endwall flow, and stalling mechanisms are evaluated. Flow measurements are performed via a pitot tube and a five-hole pressure probe to obtain the characteristic curves and hub-to-tip velocity distributions behind the second rotor. Fast Fourier Transform (FFT) and wavelet power transform analyses are further performed on the signals captured downstream of the rear rotor to identify the occurrence of instabilities. To study the flow structure in detail, transient half-annulus simulations are performed.

The test case in the present study is a single-stage contra-rotating fan that was designed and tested based on the ISO-5801 standard in the research laboratory at the aerospace engineering department of Amirkabir University of Technology32,33. The schematic of the system is demonstrated in Fig. 1. As shown, the flow passes through a bell-mouth, contra-rotating rotors, a straightener, and is finally discharged to the atmosphere through a throttle valve.

In order to measure the fan pressure and mass flow rate, pressure taps (at four circumferential locations) and a pitot tube (with a traverse system) are used. In addition, a five-hole pressure probe is placed after the second rotor. It measures five pressures and correlates the pressure changes to the flow direction. Consequently, the velocity components can be calculated using the calibration curves and mathematical relations. The five-hole probe is traversed to determine the hub-to-tip pressure and velocity distributions. The probe pressures are recorded using fast-response pressure sensors with a sampling rate of 2 kHz. In the following section, the five-hole probe correlations and calibration curves are described in detail.

Fan test setup.

The front and rear rotors consist of 10 and 8 blades, respectively. The blade profile in both of the rotors is C4. Each rotor is driven by a separate electromotor. Both rotors rotate with their design speed equivalent to 2885 rpm. The axial distance between the rotors is kept constant, with a value equivalent to 0.9 of the axial chord at the hub. Furthermore, the tip clearance of both rotors is 4% of the blade height. Some design parameters are given in Table 1. More details are available in the previous study performed on this test stand34.

The fan has been equipped with an annular endwall air injector, located at about 16% of the blade chord upstream of the front rotor. Figure 2 illustrates the injector configuration, which is designed by Catia v6-3DX R24 (https://www.3ds.com/products/catia). The injector is able to inject high momentum air in the axial direction. It consists of two main parts. The larger one is a part of the fan casing. Special curves have been machined on both parts, to guide the air flow in the proper direction. As illustrated, a gasket is employed to prevent air leakage. Furthermore, by varying the thickness of the gasket, the injection cross-section can be adjusted, allowing for injection at a constant mass flow rate with different injector exit velocities. A thinner gasket results in a smaller injection cross-section, which produces a higher exit velocity, while a thicker gasket leads to a larger cross-section and, consequently, a lower exit velocity.

Injector configuration (Created by Catia v6-3DX R24 https://www.3ds.com/products/catia). a) 3D view. b) Cross-sectional view with different gasket thickness.

The injection air is provided by a piston compressor having a 350 L storage tank, a maximum mass flow rate of 350 lit/min, and a maximum pressure of 8 bar. As shown in Fig. 3, the injection air passes through a regulator, a rotameter, an air manifold and is finally connected to the injector via 8 pipes. In this study, two injection cases with different mass flow rates have been investigated. The selected injection mass flows are 1.6% and 3.2% of the annulus mass flow rate at the last stable point of the no-injection case, which are in the range reported by reference papers35.

Injection system.

Figure 4 shows a schematic of the five-hole pressure probe and its location in the test-stand. The five-hole-probe consists of a forward-facing hole at the center, two side-holes in the vertical direction and two side-holes in the horizontal direction. The probe was calibrated in a subsonic open-circuit wind-tunnel with a speed range of 2–100 m/s. By using the probe orientation mechanism, the pitch (α) and yaw(ψ) angles were changed from − 30 to + 30 degrees, with intervals of 5° and 10°, respectively. Data was recorded separately at each pitch and yaw angle. Finally, the calibration curves were derived using the following dimensionless parameters36:

The five-hole-probe. a) Location of the five-hole-probe in test-stand. b) Schematic of the five-hole-probe.

Where p1, p2, p3, p4, p5 are the pressures sensed by the probe (see Fig. 4), pm is the mean velocity of the side holes pressures, pT is the total pressure, v is the flow velocity, and ρ is the air density.

Measurements were done at five wind-tunnel speeds. Results showed almost the same calibration curves in all cases, which proved that the measurements were accurate enough.

It is important to note that the facing holes must be exactly in the vertical and horizontal alignment. Otherwise, the manufacturing error must be counted in the measurements. Therefore, the probe was once rotated 180° in the direction of the pitch angle and then 180° in the direction of the yaw angle (relative to the reference state), and the measurements were repeated. Results indicated a 0.6% manufacturing error which had been applied to the recorded data. The calibration curves are demonstrated in Figs. 5 and 6.

After the wind tunnel calibration tests, the five-hole probe was installed in the fan test stand, one axial chord downstream of the rear rotor. At different operating conditions containing design point, near stall point, stall and post stall conditions, the probe was traversed from hub to tip and pressures were recorded separately. Velocity components are driven using calibration curves and the following relationships:

.

Variation of calibration parameters, cα and cβ, with α and ψ.

Variation of calibration parameter, Qp, with α and ψ.

The whole test stand containing inlet duct, rotor blades, straightener, and outlet duct has been simulated in this work, depicted in Fig. 7. Due to the non-periodic nature of the stall phenomenon, single-passage computations are not valid under stall conditions. The full-annulus model is the most accurate approach for studying the stall process. However, to reduce computational costs, it is common practice to simulate one-third or one-half of the entire domain in symmetrical geometries, as carried out by other researchers (such as29). Consequently, in this study a half-annulus model has been implemented, which reduces simulation costs and provides a reasonable framework for stall studies. The half-annulus model effectively captures several stall characteristics, including the type of stall, its rotational speed, and the rotor responsible for stall inception (either the front or rear rotor).

In the simulation settings, rotor blocks have been simulated in a relative frame of reference, while other components have been simulated in a stationary frame.

Boundary conditions for the inlet are the specification of total pressure, total temperature and flow angle. The flow enters the rotor axially, because there are no inlet guide vanes and also there is no inlet distortion. Therefore, the flow is assumed to be normal to the boundary at the inlet surface. Furthermore, the average gage static pressure has been prescribed at the outlet. Solid walls have been set as adiabatic and smooth. At the injection boundary, mass flow rate and static temperature have been determined. Finally, the periodic boundary condition has been set along azimuthal direction. Transient rotor-stator model has been set between rotating and stationary blocks. (Inlet duct to the front rotor, the front rotor to the mid-duct, the mid duct to the rear rotor). For other blocks which are all stationary, the Frozen-rotor model has been set.

Simulation domain (Created by Ansys-CFX 19.2 https://www.ansys.com/products/fluids/ansys-cfx).

In the current study, the commercial computational fluid dynamic software Ansys-CFX 19.2 (https://www.ansys.com/products/fluids/ansys-cfx) was used to solve 3D RANS governing equations. The second order high-resolution scheme was employed for spatial discretization of equations. Also, an implicit second-order Euler method was applied for temporal discretization. The time step was set to be 3 × 10–5 seconds. The shear stress transport (SST) k – ω turbulence model was used for the simulations, which can accurately simulate the flow in a viscous sub-layer and is reliable for flows with adverse pressure gradients.

Figure 8 depicts the computational grid, containing a view of entire passages, injection slot, and the rotor blocks. On the rotor blocks, for the front rotor, 72, 100, and 56 elements, and for the rear rotor, 49, 100, and 64 elements have been distributed in the stream-wise, span-wise and pitch-wise directions, respectively. The tip gap consists of 30 nodes in a radial direction. The y + value has been kept at less than 3.

The injection slot consists of 15, 100, and 350 elements in the stream-wise, span-wise and pitch-wise directions, respectively. Finally, the half annulus domain including inlet duct, rotor blocks, the stationary block between the rotors, straightener, and outlet duct, consists of 18,450,102 elements.

Grid independence was studied by using five different grids, in order to ensure the accuracy of the numerical results. Figure 9 illustrates the results in terms of total pressure rise coefficient and adiabatic efficiency. The values reported in this figure are belongs to the peak efficiency point. As depicted, based on the grid independence study, the main grid consists of 2,087,962 elements for rotor blocks (1,062,958 and 1,025,004 elements for rotor-1 and rotor-2, respectively).

Computational grid (Created by Ansys-CFX 19.2 https://www.ansys.com/products/fluids/ansys-cfx). a) A view of the entire domain. b) Mesh treatment near the injection slot. c) Rotor blocks.

Mesh independence study.

To validate the numerical simulations, characteristic curves and hub-to-tip velocity distribution have been compared with the experimental measurements. Figure 10 illustrates the characteristic curves of adiabatic efficiency and total pressure rise coefficient, in the no-injection case. The near stall point is determined as the last stable operating point (by 1 Pa accuracy). As clear, good agreement has been achieved between numerical and experimental results. Figure 11 demonstrates the hub-to-tip axial velocity distributions (away from the stall point), for no-injection and injection cases. The axial velocities have been normalized using the mid-span blade linear velocity. As shown, CFD and experiment results have reasonable coverage. The small difference can be attributed to the radius of the probe, so that in the experimental measurements, the velocity at each location is related to an interval of the probe radius, while in the CFD, it belongs to exactly one point. Furthermore, comparing the axial velocity trend in each case demonstrates that in the 3.2% injection case (Fig. 11c), axial velocity at the tip region has been increased. This is due to the high injection momentum at the tip region which leads to a decrease in the tip leakage destructive effect. This will be analyzed more in the following sections.

Characteristic curves (no-injection case). a) Total pressure rise coefficient. b) Efficiency.

Normalized axial velocity distribution along the blade span. a) No-injection. b) 1.6% injection. c) 3.2% injection.

To evaluate the injection effects, experimental tests and numerical simulations have been conducted for two injection cases, with mass flow rate of 1.6% and 3.2% of the annulus mass flow rate.

Figure 12 shows the performance curves of the contra-rotating fan. As depicted, endwall flow control via tip air injection shifts the characteristic line to higher mass flow rates and total pressure rises. Based on Eq. 9, the peak total pressure rise is increased by 14% and 21% in 1.6% and 3.2% injection cases, respectively, as compared to the smooth casing. Figure 12 further shows 23% and 59% enhancement in the stable operating range for the low and high mass flow injection cases, respectively, based on Eq. 10.

.

To explore tip flow mechanism, stall inception, and injection efficiency, three conditions (as clarified in Fig. 12) have been defined and compared: Condition A refers to the design point of the no-injection case which is far enough from the stall point, condition B corresponds to the in-stall point of the no-injection case, and condition C indicates the operating point of the 3.2% injection case with mass flow rate equivalent to condition B.

Characteristic curves with and without injection.

In this section, Fast Fourier Transform (FFT) and wavelet transform have been performed to analyze the experimental data, gathered by the five-hole pressure probe. The probe is located downstream of the rear rotor (see Fig. 1), traversed from hub to tip regions (0.18–0.82 of the blade height). The wavelet transform gives good localization in the time and frequency domain. In this study, the Morlet wavelet transform was used.

Figure 13 depicts the FFT diagrams of the contra-rotating fan at design and in-stall points of the no-injection case (conditions A and B in Fig. 12), at three span-wise locations; near hub, mid-span, and tip.

Hub-to-tip FFT diagrams in conditions A and B. a) Near hub. b) Mid-span. c) Near tip.

The horizontal axis in this figure is the non-dimensional frequency, relative to the rotor frequency (RF), and the vertical axis is the Power Spectral Density (PSD), normalized with respect to its value at the blade passing frequency. As shown in Fig. 13, two peaks are clear in stall condition, which were not present at the design point. The first peak with greater PSD occurred at roughly 0.4 RF, and the second one occurred with lower PSD at twice the frequency of the previous (about 0.8 RF). The first peak can be attributed to the presence of a rotating disturbance at 40% of the rotor speed, while the second one indicates its harmonic. Comparing FFT diagrams at hub, mid and tip, demonstrates the PSD increment at the tip of the blade. This indicates that the rotor is tip stall.

Figures 14 and 15 depict the FFT and wavelet power spectrum diagrams, respectively, for conditions B and C. In these figures, the effect of air injection on the stall disturbances is evaluated. Considering the wavelet power spectrum in Fig. 15-a, a long length scale disturbance (LLSD) has occurred at 0.4 RF (which is also visible in Fig. 14 as a high PSD disturbance). Therefore, this is a rotating modal wave structure that rotates at 40% of the rotor speed (See Camp and Day). Applying tip air injection in this condition completely removes the endwall disturbances, as shown in Figs. 14 and 15b. It proves that the 3.2% injection has enough momentum to eliminate tip instabilities.

FFT diagram at tip region, (Conditions B and C).

Wavelet transform spectrum. a) Condition B. b) Condition C.

In this section, numerical results are analyzed and interpreted to identify the disturbances’ structure and growth pattern. Also, it is tried to find which rotor is responsible for stall initiation. Finally, the effect of tip air injection is being investigated.

Figure 16 demonstrates the time averaged normalized axial velocity contours for the smooth casing in the design and stall conditions (Points A and B shown in Fig. 12). Three planes are shown on each rotor, located at 10, 50, and 90% of the tip chord from the leading edge. Axial velocities have been normalized with respect to the blade mid-span speed.

The tip leakage flow causes some blockage at the tip of each rotor, which can be observed as low axial velocity zones (dark blue). As shown, in condition B, the tip leakage vortex has been strengthened, especially in the rear rotor, which is identified as the reverse flow region. Furthermore, some low velocity fluid has been collected near the pressure side, at the tip (and near the leading edge) of both rotors, mostly the rear rotor. The same phenomenon occurs in high speed rotors, as reported by some researchers, such as Greitzer et al.19.

Figure 17 shows the instantaneous normalized axial velocity contours in stall condition, on the same planes that were applied to Fig. 16. Circumferential non-uniform disturbances are clear in the endwall flow-field, which can be attributed to a modal wave as described by Camp and Day37. Results are in agreement with long-length scale disturbances observed in the wavelet power spectrum (see Fig. 15a). According to the presence of the above-mentioned modal waves at the tip of the rear rotor, it can be deduced that instability has been initiated from the rear rotor.

In order to study the effect of tip injection, the time averaged normalized axial velocity contours on the same planes used in Figs. 16 and 17, demonstrated in Fig. 18 (at operating point C). It is clear that air injection has completely removed the reverse flow from the tip region of the front rotor and also, it has significantly decreased the reverse flow at the tip of the rear rotor. This postpones the stall initiation and provides range extension, demonstrated in Fig. 12.

Time averaged normalized axial velocity contours for the smooth casing at the design and in-stall conditions (Conditions A and B, respectively). (Created by Ansys-CFX 19.2 https://www.ansys.com/products/fluids/ansys-cfx). a) Condition A. b) Condition B.

Instantaneous normalized axial velocity contours in condition B. (Created by Ansys-CFX 19.2 https://www.ansys.com/products/fluids/ansys-cfx)

Time averaged normalized axial velocity contours of 3.2% injection case, in condition C. (Created by Ansys-CFX 19.2 https://www.ansys.com/products/fluids/ansys-cfx)

To more characterize the tip flow, the instantaneous static pressure rise coefficient contours together with streamlines and velocity vectors (in the absolute coordinate) on a plane at 99.6% of the span, have been depicted in Fig. 19; in conditions A, B, and C (see Fig. 12). As clear, in condition A, the tip leakage vortex does not destroy the main flow pattern. However, at point B (which is an unstable point), large reverse flow regions live near the leading edge of the rear rotor. In this situation, applying tip air injection (condition C) almost eliminates the tip vortex, causing the streamlines to be purely aligned with the main flow path.

Static pressure rise coefficient contours at the tip of the blades. a) Condition A. b) Condition B. c) Condition C.

Figure 20, depicts the instantaneous relative velocity streamlines (normalized with the blade mid-span speed), starting from the tip of the blades for both rotors. Figure 20a refers to design condition (A), whereas Fig. 20b belongs to in-stall condition (B). It is clear that at the design point, the main flow overcomes the tip vortex flow and directs it through the passage. However, in stall condition, the tip leakage vortex has grown and moved upstream. Furthermore, flow spillage (which is known as the stall initiation criteria8) occurred at the leading edge of the rear rotor.

Instantaneous streamlines at the tip of blades, in conditions A and B. (Created by Ansys-CFX 19.2 https://www.ansys.com/products/fluids/ansys-cfx). a) Condition A. b) Condition B.

In this investigation, the effects of tip leakage flow control on endwall flow structure, stall evolution, pressure capacity, and efficiency of a one stage contra-rotating fan were studied through experimental tests and numerical simulations. The control method involved annular injection of high momentum fluid at 1.6% and 3.2% of the main mass flow rate. The following results were obtained from this work:

The injection of 3.2% of the main mass flow rate increased the total pressure rise coefficient and stable operating range by 21% and 59%, respectively.

FFT and wavelet transform analyses have been applied to the experimental data. A comparison of the results under design and stall conditions at the hub, mid-span, and tip regions revealed that the investigated fan is tip critical. Furthermore, a rotating long-length scale disturbance (a modal wave) at 40% of the rotor speed has been identified.

Half-annulus transient simulations have been conducted. The results indicated the presence of circumferential perturbations under stall conditions, confirming the modal wave structure of the disturbances.

Flow spillage was observed at the leading edge of rotor-2, which is recognized as a stall inception criterion. Consequently, rotor-2 is identified as the cause of stall initiation in the current study.

The data that support the findings of this study are available from the corresponding author, upon reasonable request.

Lesley, E. P. Experiments with a Counter-Propeller. NACA TN No. 453 (1933).

Lesley, E. Tandem Air Propellers, Washington (1939).

Mistry, C. & Pradeep, A. M. Effect of variation in axial spacing and rotor speed combinations on the performance of a high aspect ratio contra-rotating axial fan stage, Proc. Inst. Mech. Eng. A J. Power Energy 227, 138–146. https://doi.org/10.1177/0957650912467453 (2013).

Nouri, H., Ravelet, F., Bakir, F., Sarraf, C. & Rey, R. Design and experimental validation of a ducted counter-rotating axial-flow fans system. J. Fluids Eng. Trans. ASME. 134 (10), 1–12. https://doi.org/10.1115/1.4007591 (2012).

Article Google Scholar

Mistry, C. & Pradeep, A. M. Influence of circumferential inflow distortion on the performance of a low speed, high aspect ratio contra rotating axial fan. J. Turbomach. 136 (7), 1–11. https://doi.org/10.1115/1.4025953 (2014).

Article Google Scholar

Wang, Y., Chen, W., Wu, C. & Ren, S. Effects of tip clearance size on the performance and tip leakage vortex in dual-rows counter-rotating compressor. Proc. Inst. Mech. Eng. G J. Aerosp. Eng. 229, 1953–1965. https://doi.org/10.1177/0954410014562483 (2015).

McDougalli, N. M., Cumpsty, N. A. & Hynes, T. P. Stall inception in axial compressors. Proc. ASME Turbo Expo 1, 63. https://doi.org/10.1115/89-GT-63 (1989).

Duc Vo, H., Tan, C. S. & Greitzer, E. M. Criteria for Spike initiated rotating stall. J. Turbomach. 130 (1), 1–11. https://doi.org/10.1115/1.2750674 (2008).

Article Google Scholar

Zhang, H., Yang, C., Wang, W., Chen, J. & Qi, M. Investigation on the casing static pressure distribution and stall behaviors in a centrifugal compressor with volute. Int. J. Mech. Sci. 160, 318–331 (2019).

Article Google Scholar

Yang, C., Fu, L., Hu, C. & Shi, X. Modelling and dynamic mode analysis of compressor impeller spike-type stall with global stability approach. Int. J. Mech. Sci. 201, 1 (2021).

Article Google Scholar

Shahriyari, M. J., Firouzabadi, A., Khaleghi, H. & Esmailifar, S. M. A new model for compressor surge and stall control. Sci. Rep. 14 (1), 1–16. https://doi.org/10.1038/s41598-024-55816-w (2024).

Article CAS Google Scholar

Dejene, T. & Pradeep, A. M. Near-Casing Flow Behavior of a High Aspect Ratio, Low Speed, Contra Rotating Axial Flow Fan Stage, vol. 2015 (2015).

Wang, Z., Yuan, W., Li, Q. & Lu, Y. Experimental investigation on the stall inception of a counter-rotating compressor. Proc. ASME Turbo Expo 8, 2405–2415. https://doi.org/10.1115/GT2012-68235 (2012).

AM, P. & M. MP and Influence of distorted inflows on the performance of a contra-rotating fan. Aeronaut. J. 125, 702–719 (2021).

Article Google Scholar

Khaleghi, H., Shahriyari, M. J. & Heinrich, M. A theory for rotating stall in contra-rotating fans, Proc. Inst. Mech. Eng. C J. Mech. Eng. Sci. 235, 3764–3773. https://doi.org/10.1177/0954406220962538 (2021).

Shahriyari, M. J., Khaleghi, H. & Heinrich, M. A model for stall and surge in low-speed contra-rotating fans. J. Eng. Gas Turbines Power. 141 (8), 251. https://doi.org/10.1115/1.4043251 (2019).

Article Google Scholar

Moore, F. K. & Greitzer, E. M. A theory of post-stall transients in axial compression systems: part 1—development of equations. J. Eng. Gas Turbines Power. 108 (1), 68–76. https://doi.org/10.1115/1.3239887 (1986).

Article Google Scholar

Takata, H. & Tsukuda, Y. Stall margin improwement by casing treatment-its mechanism and effectiveness. J. Eng. Gas Turbines Power. 99 (1), 121–133. https://doi.org/10.1115/1.3446241 (1977).

Article Google Scholar

Greitzer, E. M., Nikkanen, J. P., Haddad, D. E., Mazzawy, R. S. & Joslyn, H. D. A fundamental criterion for the application of rotor casing treatment. J. Fluids Eng. Trans. ASME. 101 (2), 237–243. https://doi.org/10.1115/1.3448945 (1979).

Article Google Scholar

Crook, A. J. et al. Numerical simulation of compressor endwall and casing treatment flow phenomena. In ASME 1992 Int. Gas Turbine Aeroengine Congr. Expo. GT 1992, vol. 1, 501–512. https://doi.org/10.1115/92-GT-300 (1992).

Yan, S. & Chu, W. L. Influence of self-circulating casing treatment with double-bleed ports structure on compressor performance. Proc. Inst. Mech. Eng. G J. Aerosp. Eng. 234, 1743–1756. https://doi.org/10.1177/0954410020918405 (2020).

Nejad, P. G. & Zenouz, R. T. Axial compressor blades tip leakage flow control using natural aspiration. Sci. Rep. 13 (1), 1–11. https://doi.org/10.1038/s41598-023-49391-9 (2023).

Article CAS Google Scholar

Mao, X., Liu, B., Tang, T. & Zhao, H. The impact of casing groove location on the flow instability in a counter-rotating axial flow compressor. Aerosp. Sci. Technol. 76, 250–259. https://doi.org/10.1016/j.ast.2018.01.037 (2018).

Article Google Scholar

Mao, X., Liu, B. & Zhao, H. Numerical analysis of the circumferential grooves casing treatment in a counter-rotating axial flow compressor. Appl. Therm. Eng. 130, 29–39. https://doi.org/10.1016/j.applthermaleng.2017.11.044 (2018).

Article Google Scholar

Guo, Y., Mao, X., Gao, L., Yu, Y. & & Numerical study on the stability enhancement mechanism of self-recirculating casing treatment in a counter-rotating axial-flow compressor. Eng. Appl. Comput. Fluid Mech. 16 (1), 1111–1130 (2022).

Google Scholar

Dong, B., Zhao, Y., Tang, T. & He, F. Flow effects of microperforated-panel casing treatments in a contra-rotating fan. Int. J. Mech. Sci. 239, 1 (2023).

Article Google Scholar

Day, J. Active suppression of rotating stall and surge in axial compressors. Proc. ASME Turbo Expo 1, 87. https://doi.org/10.1115/91-GT-087 (1991).

Weigl, H. J. et al. 1997 best paper award—controls and diagnostics committee: active stabilization of rotating stall and surge in a transonic single-stage axial compressor. J. Turbomach. 120 (4), 625–636. https://doi.org/10.1115/1.2841772 (1998).

Article Google Scholar

Taghavi-Zenouz, R. & Ababaf Behbahani, M. H. Improvement of aerodynamic performance of a low speed axial compressor rotor blade row through air injection. Aerosp. Sci. Technol. 72, 409–417. https://doi.org/10.1016/j.ast.2017.11.028 (2018).

Article Google Scholar

Khaleghi, H., Boroomand, M., Teixeira, J. A. & Tousi, A. M. A numerical study of the effects of injection velocity on stability improvement in high-speed compressors. Proc. Inst. Mech. Eng. A J. Power Energy 222, 189–198. https://doi.org/10.1243/09576509JPE443 (2008).

Khaleghi, H., Teixeira, J. A., Tousi, A. M. & Boroomand, M. Parametric study of injection angle effects on stability of transonic axial compressors. J. Propuls. Power. 24 (5), 1100–1107. https://doi.org/10.2514/1.34817 (2008).

Article Google Scholar

Bolandi, H. R. Design and Construction of Air Injection System To Increase the Performance Range of an Axial Fan (Amirkabir University of Technology, 2018).

Raoufi, M. H. Design, Construction and Performance Testing of a Contra-Rotating Fan (Amirkabir University of Technology, 2020).

Shahriyari, M., Khaleghi, H., Sadoddin, M. & Benini, E. Impact of solidity and speed ratio on the performance of a contra-rotating fan. Proc. Inst. Mech. Eng. A J. Power Energy 238, 365–381 (2024).

Hiller, S. J., Matzgeller, R. & Horn, W. Stability enhancement of a multistage compressor by air injection. J. Turbomach. 133 (3), 1–7. https://doi.org/10.1115/1.4001228 (2011).

Article Google Scholar

Gündoǧdu, M. Y. & Göksel, Ö. T. Calibration and use of a five-tube pressure probe in the swirling flow field. Turk. J. Eng. Environ. Sci. 22 (3), 223–233 (1998).

Google Scholar

Camp, T. R. & Day, J. A study of spike and modal stall phenomena in a low-speed axial compressor. Proc. ASME Turbo Expo 1, 526. https://doi.org/10.1115/97-GT-526 (1997).

Download references

Department of Aerospace Engineering, Amirkabir University of Technology, Tehran, 15875-4413, Iran

Maryam Sadoddin & Hossein Khaleghi

Search author on:PubMed Google Scholar

Search author on:PubMed Google Scholar

Both authors had substantial contributions. H.K designed and supervised the project. M.S carried out the experiments and numerical simulations. Both authors analyzed and discussed the results and contributed to the final manuscript.

Correspondence to Hossein Khaleghi.

The authors declare no competing interests.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

Sadoddin, M., Khaleghi, H. Experimental and numerical investigation of endwall flow control through air injection in a contra rotating fan. Sci Rep 15, 21080 (2025). https://doi.org/10.1038/s41598-025-05295-4

Download citation

Received: 18 October 2024

Accepted: 02 June 2025

Published: 01 July 2025

DOI: https://doi.org/10.1038/s41598-025-05295-4

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative