Experimental Investigation of Jet Mixing of a Co-Flow Jet Airfoil

5th Flow Control Conference 28 June - 1 July 2010, Chicago, Illinois AIAA 2010-4421 Experimental Investigation of Jet Mixing of a Co-Flow Jet Airfoil B. P. E. Dano, D. Kirk and G.-C. Zha Dept. of Mechanical and Aerospace Engineering University of Miami, Coral Gables, FL 33124 Abstract The jet mixing of a co-flow jet (CFJ) airfoil is investigated to understand the mechanism of lift enhancement, drag reduction, and stall margin increase. Digital Particle Image Velocimetry, flow visualization and aerodynamic forces measurements are used to reveal the insight of the CFJ airfoil mixing process. At low AoA and low momentum coefficient, the mixing between the wall jet and mainflow is dominant with large structure coherent structures for the attached flows. When the momentum coefficient is increased, the large vortex structure disappears. At high AoA with flow separation, the CFJ creates a upstream flow strip between two counter rotating vertical shear layer, i.e., the outer shear layer and inner flow induced by CFJ. The UFS is characterized with large vortex free region. The co-flow wall jet is deflected normal to the airfoil surface characterized with a saddle point. With increased momentum coefficient of the CFJ, the saddle point moves downstream and eventually disappears when the flow is attached. Turbulence plays a key role in mixing the CFJ with mainflow to transport high kinetic energy from the jet to mainflow so that the mainflow can remain attached at high AoA to generate high lift. When the flow is separated, increased CFJ momentum coefficient also increases the turbulence intensity at jet injection mixing region. AoA Angle of Attack C Chord length CC Circulation control CFD Computational Fluid Dynamics CFJ Co-Flow Jet C L Lift Coefficient C D Drag Coefficient Cµ Jet Momentum Coefficient D Drag L Lift LE Leading edge M Mach number Nomenclature Copyright 2010 by all the authors of this paper. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.

p Static pressure Pt Total Pressure Re Reynolds number S Wing Span Area TE Trailing edge ZNMF Zero net mass flux m Jet mass flow rate u,v,w Velocity components in x-, y-, and z-direction V Velocity vector x,y,z chord, normal and spanwise directions, with respect to the airfoil Subscripts: Freestream J Jet Greek Letters: γ Ratio of Specific Heats ρ Density α Angle of Attack β Angle between slot surface and the line normal to chord I. Introduction Flow control is a promising aerodynamic concept that can break through the conventional aerodynamic constraints and achieve drastic performance enhancement [1, 2, 3, 4, 5, 6, 7]. Airfoils are fundamental elements of many fluid systems including airplane, gas turbine engine turbomachinery, wind turbine, propeller, pumps, etc. Airfoil flow control principles can be readily applied to various fluid systems working under adverse pressure gradient. Therefore, study of a flow control performance enhancement method for airfoils will have broader impacts and may bring revolutionary design concepts to numerous systems using fluids as working medium. The motivation behind active airfoil flow control is to achieve increased lift, stall margin and reduced drag with low energy expenditure by using air flow with a minimal solid structure device, which imposes weight penalty, system complication, and parasitic drag when not in use. Various flow control techniques have been pursued recently including zero-net mass flux (ZNMF) synthetic jets [8] using acoustic wave excitation and dielectric-barrier discharge plasma actuators [9, 10]. However, at present, both ZNMF synthetic jets and plasma actuators are generally lacking in terms of sufficient actuator authority for high speed flows. The promise of these various flow control techniques has not yet been achieved. Pulsed fluidic actuators are just beginning to deliver effective mixing and performance enhancement for realistic high Reynolds number flows [11, 12]. In contrast, a circulation control (CC) [13, 14] airfoil is one method of flow control driven by fluidic actuators that has been actively pursued for aircraft performance improvement over the past three decades. A typical CC airfoil relies on the Coanda effect and requires a large leading edge (LE) and trailing edge (TE). However, a large LE and TE may induce a large drag during

flight. A movable flap at the airfoil TE is suggested to overcome the blunt TE drawback [14] but subsequently imposes a weight penalty. To maintain sufficient stall margin, LE blowing is usually needed. A considerable penalty of the CC airfoil is the dumped blowing jet mass flow, which is imposed on the propulsion system. Furthermore, for a CC airfoil, the drag measured in the wind tunnel is not representative of the actual drag that occurs during flight. This is because, in a wind tunnel test, the penalty to draw the mass flow from the free stream as the supply for the jet injection acts as a side force and can not be included in the drag measurement. The actual drag, also called the equivalent drag, must therefore include this penalty [15, 16, 17], which is composed of the ram drag and captured area drag. Most recently, a new ZNMF jets flow control airfoil based on fluidic actuators (hereafter called co-flow jet airfoil or CFJ airfoil) has been developed by Zha et al. [18-20, 23], to avoid dumping the jet mass flow and achieve performance enhancement without relying on the Coanda effect. The co-flow jet airfoil is to open an injection slot near leading edge and a suction slot near trailing edge on the airfoil suction surface. A high energy jet is injected near leading edge tangentially and the same amount of mass flow is sucked in near trailing edge. The turbulent shear layer between the main flow and the jet causes strong turbulence diffusion and mixing under severe adverse pressure gradient, which enhances lateral transport of energy from the jet to main flow and allows the main flow to overcome severe adverse pressure gradient and remain attached at high angle of attack(aoa). The high energy jet induces high circulation and hence generates high lift. The energized main flow fills the wake and therefore reduce drag. The zero net jet mass flow of a CFJ airfoil can minimize the penalty to propulsion system. The purpose of this research is to investigate the turbulent jet mixing of a CFJ airfoil in order to understand the mechanism responsible for lift enhancement, drag reduction, and stall margin increase. Digital Particle Image Velocimetry, flow visualization and aerodynamic forces measurements are used to reveal the insight of the CFJ airfoil mixing process. II. Experimental setup A schematic of a CFJ airfoil concept is shown in Fig. 1. A high energy jet is injected near the LE in the direction tangential to the main flow and the same amount of mass flow is drawn into the airfoil near the TE. Pressurized air is injected in a spanwise long cavity near the LE, and then exits through the spanwise long rectangular slot. A Duocel high density aluminum foam was placed between the inlet and the exit slot to equilibrate the pressure and ensure a uniform exit velocity. Similarly, a spanwise long cavity placed near the trailing edge is used to let the air settle down before being sucked through the three suction ports. The injection and suction slot height are 0.65% and 1.42% of the chord, respectively, the airfoil chord is 12" and the span is 24". The injection slot and suction slots are located at 7.5% and 88.5% of the chord, respectively. All airflow and aerodynamic variables were acquired at the University of Miami 24 x24 wind tunnel facilities. Fig. 2 and Fig. 3 show an overview of the UM Aero-lab facilities and the CFJ airfoil. The injection and suction flow conditions were independently controlled. A compressor supplies the injection flow line and the flow rate

controlled using a Koso Hammel Dahl computer controlled valve. A vacuum pump generates the necessary low pressure for suction and is controlled with a manual needle valve. Both mass flow rates in the injection and suction lines are measured using Oripac orifice mass flowmeters equipped with high precision MKS pressure transducers. The aerodynamics variables are measured using an AMTI 6 components transducer. All wind tunnel freestream, CFJ airflow, and aerodynamic variables were recorded using a state-of-the-art Labview data acquisition system. LE trip was not implemented since we found that the results are insensitive to the LE trip. All the data (e.g., wind tunnel speed, aerodynamics forces, blowing and suction mass flow rates) were acquired at a rate of approximately 50 samples per second, allowing for limited airfoil flutter analysis around stall angle(s). The range of angles of attack (AoA) varied between 0 o and 30 o. The nominal free stream velocity was V =10m/s for all tests and the chord Reynolds number was about 1.89x10 5. Similarly to C L and C D, a jet momentum coefficient Cµ was defined as follows: mv j C (1) 2 1/ 2 V S With increasing AoA, the pressure over the airfoil is expected to decrease and facilitate the CFJ jet penetration. This results in a higher Cµ. Fig. 4 shows the value of Cµ and corresponding jet velocity for all tests and illustrates that compensation was relatively well achieved for lower Cµ. For this paper, three nominal values of Cµ at AoA=0 o are presented: 0.06, 0.15 and 0.25, corresponding to mass flow rates of 0.030 kg/s, 0.045 kg/s and 0.060 kg/s, respectively. The jet velocity is varied correspondingly at about 23m/s, 34m/s, and 45m/s. Various laser flow visualization techniques were used to monitor the circulation over the upper surface. A LaVision Digital Particle Image Velocimetry (DPIV) system with a Litron Nano Nd:YAG 200 mj/pulse was used to monitor and acquire the velocity field surrounding the airfoil. An adaptive 64x64 to 32x32 pixel cross-correlation analysis method was used with masking over the airfoil, resulting in 75x100 total vectors (including the airfoil). A series of 1,000 velocity fields were acquired for each AoA and Cµ. A customized flow seeder using the same particles for PIV tracers was used for flow visualization. The flow along the span direction was found to be uniform within 1% discrepancies between the root and the tip, while the suction showed a 7% skewness toward the root at the suction side. All the results presented were acquired along section B, on Fig. 3. III. Large Structure Vortex Detection For each series of test, the 1,000 instantaneous DPIV samples were averaged to yield the mean velocity and vorticity fields. A large structure vortex detection algorithm (VDA) based on Graftiaux et al.[21] was used to describe the vortical structure [2]. For all grid points P of the DPIV grid, the vortex detection function, Γ 1 was calculated over a sub-area centered on P. The sub-area size, A M, centered about each velocity grid point within the flow, ranged from a 3x3 to 11x11 square grid with a grid node spacing of 2.4 mm. In other words, the vortex structure size detected will be greater than 2.4x2.4 mm 2.

The vortex detection function is defined by: 1 ( P) A M ( PM U hp ( M )) zˆ 1 da M da PM U M A sin( ) ( ) M 1 (2) A hp A where M are the discrete grid points within the sub-area, A M, that correspond to velocity data points. The larger the value of Γ 1, the greater the swirl component of the flow caused by the fluctuating velocity. The criterion Γ 1 >0.90 is used to identify a vortex motion within the flow, where Γ 1 varies between -1 and 1. Based on the sign of Γ 1, the rotation is either clockwise (CW) or counterclockwise (CCW). Upon detection of a vortex, a circle of the size of the detected vortex is plotted and filled with a color corresponding to the vortex rotation direction: blue=cw and red=ccw. When applied to the 1,000 instantaneous flow fields, the VDA results are compiled into a single figure showing the distribution of the detected vortices. Aerodynamic Performances IV. Results and Discussion Wind tunnel results for lift coefficient (C L ) and drag coefficient (C D ) for the baseline and CFJ airfoils for different angle of attacks and different Cµ are shown in Fig.5a and 5b. Compared to baseline, the CFJ airfoil shows a dramatic gain in C L and is maintained with higher values of Cµ. Lift increase at AoA=0 o varies between 10% and 80% for the range of Cµ used. The max C L increase at stall AoA varies between 22% and 72% for the range of Cµ used. Fig. 6 regroups the performance increase for all Cµ at zero AoA and stalled AoA. These results can be explained by the combined effect of the co-flow jet and suction increasing the circulation and preventing flow separation. Drag coefficient results for the CFJ show a drastic decrease in drag compared to baseline. Moreover, zero to negative drag (thrust) can be observed for a large range of AoA values. This trend is further increased for higher Cµ. This can be explained by the reduced or reversed wake velocity deficit resulting in drag reduction and thrust generation [17,18]. Since the aerodynamic efficiency of an aircraft is measured by L/D, the CFJ airfoil hence could be an efficient device to simultaneously increase lift and reduce drag with certain expense of pumping power consumption. The baseline results show a stall occurring around AoA=22, whereas the CFJ airfoil results show a systematic increased stall AoA between AoA=25 and AoA=30 for all Cµ. Due to the large amount of data recorded, only the results pertaining to the mixing and flow mechanism of the CFJ performance increase are presented here. All the following results are for a freestream velocity of U =10m/s. Two samples of instantaneous smoke flow visualization for the baseline and CFJ airfoil at AoA=25 o are shown in Fig. 7. One can see that the baseline airfoil flow is largely separated. On the top surface, Kelvin Helmholtz (KH) vortices can be seen shedding from the separation point. The flow from the intrado past the TE generates a

large recirculation that feeds the separated wake. In contrast, the CFJ airfoil shows a smooth flow hugging the top surface with a thin downward inclined wake past the TE. At low AoA, a closer look at the jet flow in the vicinity of the jet exit without and with CFJ, shown in Fig. 8, reveals a street of small KH that dissipates rapidly. The coherent vortex structure rotates clockwise with no CFJ, and the rotating direction is reversed with CFJ due to the higher jet velocity than that of the freestream. The vortex street modulates in size with increasing Cµ and disappears for Cµ> 0.14. General Mixing Characteristics A comparison of the time averaged velocity field for baseline with a typical CFJ at AoA=25 o is shown in Fig. 9. For clarity, only 1 in 5 vectors are shown in all directions. The baseline airfoil shows a large separation originating from the LE, characteristic of a stalled airfoil. The flow is observed passing from the TE to the LE all the way to the separation point and is then entrained in the outer flow, creating a shear layer. Instantaneous PIV snapshots show that the shear layer region contains vortices, but in average, the air flow appears to merge nicely. The average velocity in the wake is observed to have a much slower magnitude than the freestream. In contrast, the CFJ airfoil shows a smooth flow along the entire surface of the airfoil with no sign of separation. The injection jet velocity is about V j /U =2.2, but a region of high velocity originating from the LE due to the strong LE suction effect is observed, with values up to V max /U =3.0. Comparisons of the VDA results, shown in Fig. 10, provide more information on the vortical structure. For baseline, the shear layer is revealed with the multiple large CW vortices originating from the flow separation point. In contrast, the VDA results for the CFJ airfoil show that no large vortical structure is present. The wakes of airfoils are always of great interest. A comparison of the average vorticity field for baseline and CFJ at Cµ=0.06 at AoA=25 o is shown in Fig. 11. The flow of the baseline airfoil from the intrado passes the TE and creates a large shear layer that turns rapidly horizontal. VDA results (not shown) confirm that this shear layer corresponds to numerous CCW vortices spatially averaged. Further downstream, the flow of the baseline airfoil turns back, creating a large vortical structure. The air in the separated region flows upstream along the extrado as observed in Fig. 9a. When the CFJ in turned ON, the wake consists in a thin layer with opposite vorticity that extends downward with an angle of about 25 o from the freestream direction. This shear layer is characteristic of a jet, which reduces the drag or generates thrust depending on the intensity of the CFJ[17,18]. The VDA results reveal a few large structure vortices that disappear completely for higher Cµ. Fig. 12 to 14 show the time averaged PIV CFJ airfoil velocity fields and the VDA results for AoA=30 o for all three Cµ. For the lower Cµ=0.06, one can see that the flow is largely separated since the CFJ is not strong enough to achieve attached flow at such high AoA. The outer flow shear layer is seen shedding numerous CW (blue) vortices. Between the outer flow and the jet flow, the flow can be observed traveling upstream to meet the separation point, which is reminiscent of the baseline stalled flow. The upstream flow strip (UFS) is sandwiched between the outer shear layer generating CCW vortices and the flow induced by the CFJ injection that produces CW

vortices. The UFS is a region of large vortex free. Below the UFS, the jet appears to decay rapidly and strongly interact with the return flow from the TE. A white triangular marker and a dash line indicate the location of the saddle point area. The jet is deflected normally to the airfoil surface feeding the UFS and the outer shear layer. The jet-to-return saddle point area is characterized by a rapid change from the CCW vortex of outer shear layer to the CW vortices induced by the CFJ. With the increased Cµ, the saddle point is pushed downstream along the airfoil surface. The UFS becomes thinner and the large vortices appear to increase in number and size. Finally, the results for the higher Cµ=0.25 are shown in Fig. 14. The UFS appears to be completely suppressed, the number of vortices is greatly decreased and the saddle point is, again, pushed further downstream. Inspection of instantaneous samples confirms that the above features are not artifact from the averaging process and show that the UFS is a combination of pairing of two counter-rotating vortices funneling flow from the wake to the LE separation point, as shown in Fig. 15a. The jet deflection normal to the surface also appears to be eased by outer shear layer CW rollups, as seen in Fig. 15b. The flow in the vicinity of the TE for Cµ=0.25 is shown in Fig. 16. While the jet action of delaying the separation benefits to the airfoil circulation, the suction appears to slightly accelerate the flow from the TE in the upstream direction. The wake is typical of a separated flow. CW vortices are observed on top of the shear layer with the delayed outer flow and CCW vortices are shedding off the TE and the shear layer flow from the intrado. Turbulent Mixing Combining the time averaged velocity and VDA results provides ample information on the mean and the vortical characteristics of the flow, but mixing should also be considered through non-vortical fluctuations. The intermittent characteristics of the flow can be observed using the turbulent kinetic energy field (TKE). Corresponding TKE fields for Fig. 12-16 are shown in Fig 17 to Fig. 19. For Cµ=0.06, most TKE originates in the jet area, indicative of the high velocity gradient caused by the injection jet penetration. The outer shear layer starting from LE and stretching to downstream also shows high value TKE, again mostly due to the high velocity gradient between the high speed free stream and the UFS. While the wake above the TE shows very low levels of fluctuation, the area below the TE shear layer shows very high TKE values, indicative of a high intensity shear layer between the flow from the upper suction surface and lower pressure surface, as shown in Fig. 7. The TE shear layer turbulence intensity is reduced when Cµ is increased. This is because the increased CFJ intensity decreases the velocity dissimilarity between the suction and pressure surface. For Cµ=0.25 shown in Fig. 19, the LE separation point and the CFJ injection area shows the highest value of TKE. The flow is about to be attached. Observations of instantaneous DPIV velocity fields shows that an intermittent flow pattern is established with either a separated flow or a re-attached flow. In contrast, past the TE, the level of turbulence in the wake is largely decreased due to reduced wake velocity deficit. The increased turbulence intensity at the CFJ injection area indicates the increased turbulence mixing intensity, which transports high kinetic energy from the jet to the main flow so that the main flow can overcome the severe adverse pressure gradient at high AoA and remain attached.

When the flow is attached, the turbulence kinetic energy is significantly reduced due to the disappearance of large structure vortices as shown from Fig. 20 to 22. IV. Conclusion The turbulence mixing mechanism of a CFJ airfoil is investigated in this paper. At low AoA and low momentum coefficient, the mixing between the wall jet and mainflow is dominant with large structure coherent structures for the attached flows. When the momentum coefficient is increased, the large vortex structure disappears. At high AoA with flow separation, the CFJ creates a upstream flow strip between two counter rotating vertical shear layer, i.e., the outer shear layer and inner flow induced by CFJ. The UFS is characterized with large vortex free region. The co-flow wall jet is deflected normal to the airfoil surface characterized with a saddle point. With increased momentum coefficient of the CFJ, the saddle point moves downstream and eventually disappears when the flow is attached. Turbulence plays a key role in mixing the CFJ with mainflow to transport high kinetic energy from the jet to mainflow so that the mainflow can remain attached at high AoA to generate high lift. When the flow is separated, increased CFJ momentum coefficient also increases the turbulence intensity at jet injection mixing region. When the flow is attached, the turbulence kinetic energy is significantly reduced due to disappearance of large structure vortices. Acknowledgement: This research is supported under ARO/AFOSR Grant 50827-RT-1SP. We appreciate the assistance from P. Ly, M. Castillo, A. Beaudry and K. Chin-Sim for operating the wind tunnel tests. REFERENCES [1] W. L. I. Sellers, B. A. Singer, and L. D. Leavitt, Aerodynamics for Revolutionary Air Vehicles. AIAA 2004-3785, June 2003. [2] M. Gad-el Hak, Flow Control: The Future, Journal of Aircraft, vol. 38, pp. 402 418, 2001. [3] M. Gad-el Hak, Flow Control, Passive, Active, and Reactive Flow Management. Cambridge University Press, 2000. [4] S. Anders, W. L. Sellers, and A. Washburn, Active Flow Control Activities at NASA Langley. AIAA 2004-2623, June 2004. [5] C. P. Tilmann, R. L. Kimmel, G. Addington, and J. H. Myatt, Flow Control Research and Application at the AFRL s Air Vehicles Directorate. AIAA 2004-2622, June 2004. [6] D. Miller,, and G. Addington, Aerodynamic Flowfield Control Technologies for Highly

Integrated Airframe Propulsion Flowpaths. AIAA 2004-2625, June 2004. [7] V. Kibens and W. W. Bower, An Overview of Active Flow Control Applications at The Boeing Company. AIAA 2004-2624, June 2004. [8] R. Holman, Y. Utturkar, R. Mittal, and L. Cattafesta, Formation Criterion for Synthetic Jets, AIAA Journal, vol. 43, No. 10, pp. 2110 2116, 2005. [9] T. C. Corke and M. L. Post, Overview of Plasma Flow Control: Concepts, Optimization, and Applications. AIAA Paper 2005-0563, Jan. 2005. [10] C. Enloe, T. E. McLaughlin, G. I. Font, and J. W. Baughn, Frequency Effects on the Efficiency of Aerodynamic Plasma Actuator. AIAA Paper 2006-0166, Jan. 2006. [11] Raman, G., Using controlled unsteady fluid mass addition to enhance jet mixing., AIAA Journal, vol. 35, No. 4, pp. 647 656, 1997. [12] Freund, J.B. and Moin, P., Jet mixing enhancement by high-amplitude fluidic actuation, AIAA Journal, vol. 38, No. 10, pp. 1863 1870, 2000. [13] N. Wood and J. Nielsen, Circulation Control Airfoils-Past, Present, Future. AIAA Paper 85-0204, 1985. [14] R. J. Englar, Circulation Control Pneumatic Aerodynamics: Blown Force and Moment Augmentation and Modifications; Past, Present and Future. AIAA 2000-2541, June 2000. [15] M. Wilson and C. von Kerczek, An Inventory of Some Force Procedure for Use in Marine Vehicle Control. DTNSRDC-791097, Nov, 1979. [16] G. S. Jones, Pneumatic Flap Performance for a 2D Circulation Control Airfoil, Steady & Pulsed. Applications of Circulation Control Technologies, Chapter 7, p. 191-244, Vol. 214, Progress in Astronautics and Aeronautics, AIAA Book Series, Editors: Joslin, R. D. and Jones, G. S., 2006. [17] G.-C. Zha, W. Gao, and C. Paxton, Jet Effects on Co-Flow Jet Airfoil Performance, AIAAJournal, No. 6,, vol. 45, pp. 1222 1231, 2007. [18] G.-C. Zha and D. C. Paxton, A Novel Flow Control Method for Airfoil Performance Enhancement Using Co-Flow Jet. Applications of Circulation Control Technologies, Chapter 10, p. 293-314, Vol. 214, Progress in Astronautics and Aeronautics, AIAA Book Series, Editors: Joslin, R. D. and Jones, G.S., 2006. [19] G.-C. Zha, C. Paxton, A. Conley, A. Wells, and B. Carroll, Effect of Injection Slot Size on High Performance Co-Flow Jet Airfoil, AIAA Journal of Aircraft, vol. 43, 2006.

[20] G.-C. Zha, B. Carroll, C. Paxton, A. Conley, and A. Wells, High Performance Airfoil with Co-Flow Jet Flow Control, AIAA Journal, vol. 45, 2007. [21] Graftieaux, L., Michard, M. and Grosjean, N., 2001. Combining PIV, POD and vortex identification algorithms for the study of unsteady turbulent swirling flows, Meas. Sci. Technol., 12, pp. 1422-1429. [22] Dano, P.E., and Liburdy, J.A., Vortical Structure of a 45 o Inclined Pulsed Jet in a Crossflow, AIAA-2006-3543,2006. [23] Zha, G.-C., Gao, W. and Paxton, C. D. "Jet Effects on Co-Flow Jet Airfoil Performance ", AIAA Journal, Vol.45, No.6, 2007, pp1222-1231

Fig. 1. Schematic and concept of CFJ airfoil Fig. 2 Wind tunnel facility C B A Fig. 3 CFJ airfoil and PIV plans locations Fig. 4 Variation of Cµ and jet exit velocity for various AoA

Lift Improvement (%) 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 20% at AoA=0deg at Stall AoA 35% Fig. 5 Lift and Drag coefficient results comparison for baseline and CFJ at various AoA and Cµ 46% 59% 0.06 0.14 0.25 Cµ 77% 82% Drag Reduction (%) 1200% 1000% 800% 600% 400% 200% 0% at AoA=0deg at Stall AoA 92% 53% 445% 58% 1009% 0.06 0.14 0.25 Cµ Fig. 6 Performance enhancement at zero AoA and stall AoA. 65% Fig. 7 Smoke flow visualization for Baseline and CFJ (Cµ=0.14) at AoA=25 o

Fig. 8 Close-up flow visualization for CFJ at Cµ=0 and Cµ=0.02 for AoA=5 o

Fig. 9 Time averaged velocity field at leading edge, for baseline and CFJ (Cµ=0.06) for AoA=25 o Fig. 10 VDA results at the leading edge for baseline and CFJ (Cµ=0.06) for AoA=25 o Fig. 11 Time averaged vorticity field at the trailing edge for Cµ=0 and Cµ=0.25 for AoA=25 o

Fig. 12 Leading edge results for mean velocity and VDA for Cµ=0.06 and AoA=30 o Fig. 13 Leading edge results for average velocity field and VDA for Cµ=0.14 and AoA=30 o Fig. 14 Leading edge results for average velocity field and VDA for Cµ=0.25 and AoA=30 o

Fig. 15 Instantaneous DPIV samples Cµ=0.06 and AoA=30 o

Fig. 16 Trailing edge results for average velocity field and VDA for Cµ=0.25 and AoA=30 o Fig. 17 TKE fields at Leading edge and trailing edge for Cµ=0.06 and AoA=30 o Fig. 18 TKE fields at Leading edge and trailing edge for Cµ=0.14 and AoA=30 o Fig. 19 TKE fields at Leading edge and trailing edge for Cµ=0.25 and AoA=30 o

Fig. 20 TKE fields at Leading edge and trailing edge for Cµ=0.06 and AoA=25 o Fig. 21 TKE fields at Leading edge and trailing edge for Cµ=0.14 and AoA=25 o Fig. 22 TKE fields at Leading edge and trailing edge for Cµ=0.25 and AoA=25 o