PIV Study on Steady and Periodic-Pulsed Dielectric-Barrier-Discharges

Transcription

1 49th AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition 4-7 January 211, Orlando, Florida AIAA th AIAA Aerospace Sciences Meeting Including the New Horizons Forum and Aerospace Exposition, 4-7 Jan 211, Orlando, Florida PIV Study on Steady and Periodic-Pulsed Dielectric-Barrier-Discharges Chong Ma, Huaxing Li Northwestern Polytechnical University, Xi an 7172, China Feng Liu, and Shijun Luo University of California, Irvine, CA The flows induced by different modes of plasma discharge over a 2 conical forefody are studied with a two-dimensional particle image velocimetry in a still air chamber with one atmosphere pressure. The effects of pulse duty ratio on the induced maximum crossflow velocity and maximum axial vorticity are compared with those of steady discharge. Effects of pulse frequency and ac voltage are discussed. Nomenclature F = frequency of a.c. voltage source f p = frequency of pulsed actuation q = velocity component in crossflow plane r, θ = polar coordinates at body center, θ measured from windward generator clockwise r c = body radius V p p = peak-to-peak voltage of a.c. voltage source v, w = horizontal, vertical component of q τ p = duty ratio of pulse over a pulse period ψ = phase angle over a pulse period ω x = axial vorticity component in x direction I. Introduction Proportional lateral control on slender forebodies at high angles of attack is highly needed in aerodynamic design of air vehicles. The fact that the separation vortices over pointed forebodies generate large airloads and are very sensitive to small perturbations near the body apex offers an exceptional opportunity for manipulating them with little energy input to achieve active lateral control of the vehicle in place of conventional control surfaces. It has been found experimentally that unsteady dynamic control techniques are needed to achieve this goal. 1 3 Recently, Liu et al. 4 reported wind-tunnel experiments that demonstrate nearly linear proportional control of lateral forces and moments over a slender conical forebody at high angles of attack by employing a novel design of a pair of single dielectric barrier discharge (SDBD) plasma actuators near the cone apex combined with a duty cycle technique. In the duty-cycled plasma actuation, both starboard and port plasma actuators discharge in the mode of periodic pulse. The periodic-pulsed discharges are the key factors in the action of the duty-cycled actuation. The information on the performance of periodic-pulsed discharge may Graduate Student, Department of Fluid Mechanics Professor, College of Aeronautical Engineering. Professor, Department of Mechanical and Aerospace Engineering. Associate Fellow AIAA. Researcher, Department of Mechanical and Aerospace Engineering. 1of16 Copyright 211 by the American Institute of Aeronautics and American Astronautics, Institute Inc. All of rights Aeronautics reserved. and Astronautics

2 provide insight into the mechanism of the duty-cycled SDBD plasma actuation. These considerations motived the current study. The action of asymmetric surface dielectric barrier discharge in air at atmospheric pressure has been tested along different kinds of surface profiles, e.g., flat plates, 7 circular cylinder, 8 9, 1 NACA airfoils. In this paper, the crossflow fields produced by a periodic-pulsed SDBD along a ray of 2 circular cone in air at atmospheric pressure are studied in detail using the PIV visualizations. The test model is mainly the same as that is used in Ref. 4. In the following sections, the experimental setup is described. The ensembleand phase-locked-averaged results for various modes of plasma actuations are investigated. Effects of pulse frequency and ac voltage are discussed. Finally conclusions are drawn. II. Experimental Setup Figure 1. The model The model is shown in Fig. 1 and mainly the same as used in Ref. 4. The circular-cone forebody consists two separate pieces. The frontal portion of the cone is made of plastic and has a length of 1 mm. Therest of the model is made of metal. The total length of the cone is mm with a base diameter of mm. Two long strips of SDBD plasma-actuators are installed symmetrically on the plastic frontal cone near the apex as shown in Fig. 2(a). The plasma actuator consists of two asymmetric copper electrodes each of.3 mm thickness. A thin Kapton dielectric film of thickness.4 mm wraps around the cone surface and separates the encapsulated electrode from the exposed electrode as shown in Fig. 2(b). The right edge of the exposed electrode shown in Fig. 2(b) is aligned with the cone at the azimuth angle θ = ±12 (Fig. 2of16

3 2(a)). The length of the electrodes is 2 mm along the cone meridian with the leading edge located at 9 mm from the cone apex. The width of the exposed and encapsulated electrode is 1 mm and 2 mm, respectively. The two electrodes are separated by a gap of 1. mm, where the plasma is created and emits a blue glow in darkness. (a) arrangement (b) SDBD Figure 2. Sketches of the plasma actuators. Four modes of operations of the starboard actuators are defined. The first model is steady discharge of the starboard actuator. The rest modes are periodic-pulse discharge of the starboard actuator with pulse duty ratio τ p =.1,. and.9. The pulse durations for τ p =.1,. and.9 are1, and 9 milliseconds, respectively, at pulse frequency f p =1Hz. Each of the two actuators on the cone model is separately driven by an a.c. voltage source (model CTP-2K by Nanjing Suman Co.). The waveform of the a.c. source is sine wave. The peak-to-peak voltage and frequency are set at V p p 14. kv and F 11.8 khz, respectively. The periodic pulses are achieved by modulating the carrier a.c. voltage sources by a digital pulse wave generator, PC-7 made by Nanjing Suman Co. Figure 3. PIV system and plasma generators. The plastic frontal cone is used for the present experiments. The wind-tunnel study using the entire model of Fig. 1 will be reported in another paper. Particle image velocimetry (PIV) is chosen for the flowfield measurement because the method is to a large degree noninstrusive. The cross-flow velocity field over a station of x =24.6 mm which is measured from the cone apex, is visualized by a two-dimensional PIV system as shown in Fig. 3. The radius of the body cross-section is 4.34 mm. The possible adverse effects of the electrostatic force of the plasma actuator on the PIV measurement are neglected. The study is conducted inside a space formed by a rectangular cover, which is of 6 mm in length, mm in width and mm in height. All the five sides of the cover are made from mm thick Plexiglas 3of16

4 to allow for optical viewing and access for the laser sheet. The cover is shown between the Suman plasma voltage sources on the right and a multi-channel plasma generator on the left in the photograph in Figure 3. The plastic frontal cone is set inside the cover. The air under the cover at one atmosphere pressure is shielded from air flow within the laboratory room. The seeds are smoke particles of approximately 1 μm in diameter commonly used in cinema industry. The seeds would stay suspended for many hours and were only replenished when needed. The PIV system manufactured by the Dantec Dynamic Company, is located in the front in the photograph of Figure 3. The Nd:YAG Laser, a product of the Beamtech Optronics Co., emits single pulse of energy 2 mj and produces double pulses with a time interval of 6 μs. The laser sheet has thickness 1 mm. The repeat rate of the laser double-pulse is limited to 1 Hz. Forf p =1Hz, the repeat rate of the laser double-pulse is set at 9 Hz and consecutive 1 seconds of sampling are performed for each case. To calculate a phase-locked average, every 9 PIV samples are used. For f p =1Hz the order of the PIV sampling coincides with the order of phase angle.in the present experiments, the plasma generator and the PIV velocimetry are operated independently, and, thus, the exact phase angle corresponding to the pulse initiation is unknown. In this paper, the first phase angle is chosen at the point where the phase-locked-averaged maximum crossflow velocity in the entire neighborhood of the cone is the largest among all the phase angles. A CCD camera of pixels is used to record the cross-flow image of 36.1 mm 32. mm. A software of DynamicStudio version is used to calculate the cross-flow velocity vector field from the double-pulse images. The camera is located in front of the laser plane as shown in Fig. 3. Table 1 gives the input power versus periodic-pulse duty ratio of the starboard plasma actuator generated by an ac voltage source: V p p =14. kv and F =11.8 khz, whereτ p = 1 denotes steady plasma actuation. The pulse frequency f p =1Hz. The input power is approximately proportional to the duty ratio. Table 1. Input power versus periodic-pulse duty ratio, V p p =14. kv, F =11.8 khz, f p =1Hz. Duty ratio Voltage (V ) Current (A) Input power (W ) III. Ensemble-Averaged Results To study the convergence of ensemble-averaged crossflow velocity versus sampling time, the mode of steady plasma actuation (τ = 1) is considered. On the crossflow plane, a fixed point of r/r c =1.81 and θ = 23.9, is chosen, where the ensemble-averaged maximum crossflow velocity obtained from the PIV sampling time of 1 s in the entire neighborhood of the cone surface is located. At this point, the ensembleaveraged crossflow velocities over sampling time of 1 s, s and 1 s are listed in Table 2. The sampling time of 1 s almost yields convergent result. Table 2. Convergence of ensemble-averaged crossflow velocity versus sampling time for steady plasma actuation. PIV sampling time (s) Crossflow velocity (m/s) Figure 4 presents the PIV images of ensemble-averaged crossflow-velocity contours and vectors induced by the starboard plasma actuation at pulse duty ratio τ p =.1,. and.9, pulse frequency f p =1Hz and at steady plasma actuations (i.e., τ = 1). The PIV image size of 28 mm 27 mm covers the entire neighborhood of the body cross-section disturbed by the starboard plasma actuation. The actuator produces at azimuthal location, θ = 12, a locally tangential and counterclockwise blowing that adheres to the surface of the cone via an apparent Coanda effect. The radius of the body cross-section, r c =4.34 mm. The body center is located at y =26.89 mm and z =7.1 mm. For PIV, the sampling time is 1 s, the sampling frequency is 9 Hz, and the total number of sampling images is 13. The axial-vorticity ω x over the crossflow plane is calculated from the measured crossflow velocity. Figure 4of16

5 q(m/s) q(m/s) (a) τ p = (b) τ p = q(m/s) q(m/s) (c) τ p = (d) τ p =1 Figure 4. Ensemble-averaged crossflow-velocity contours and vectors for periodic-pulsed (f p =1Hz) and steady plasma actuations. of16

6 presents the PIV images of ensemble-averaged axial-vorticity contours for periodic-pulsed (f p =1Hz) and steady plasma actuations. The crossflow-velocity vectors are plotted over the axial-vorticity contours ω x (1/sec) ω x (1/sec) (a) τ p = y.mm (b) τ p = ω x (1/sec) ω x(1/sec) (c) τ p = (d) τ p =1 Figure. Ensemble-averaged axial-vorticity contours and crossflow-velocity vectors for periodic-pulsed (f p = 1 Hz) and steady plasma actuations. The characteristics of the induced flow is shown by two parameters: ensemble-averaged maximum crossflow velocity, q max and maximum magnitude of axial vorticity, ω x max.table3presentsq max and ω x max in the neighborhood of the plasma discharge edge, θ =6 12. The horizontal and vertical components of q max are denoted by v and w, respectively, where v points port when positive and w points upward when positive. The direction of axial-vorticity is couter-clockwise when it is positive. In the neighborhood of the plasma discharge edge, q max =.92 m/s and ω x max = 131 /s for the steady plasma actuation (τ p =1) is the largest among the various τ p considered. Table 4 presents q max and ω x max intheentireneighborhoodofthebody,θ = 18. q max = m/s and ω x max = 23 /s for the steady plasma actuation τ p = 1 are the largest among the various τ p considered. In the entire neighborhood of the body, q max and ω x max are located close to θ = for all cases. 6of16

7 Table 3. Ensemble-averaged maximum crossflow velocity q max and maximum magnitude of axial vorticity ω x max induced by periodic-pulsed (f p =1Hz) and steady discharge in neighborhood of starboard actuator. Pulse duty ratio q max (m/s) v(m/s) w(m/s) r/r c θ ω x max (1/s) Direction r/r c θ Table 4. Ensemble-averaged maximum crossflow velocity q max and maximum magnitude of axial vorticity ω x max induced by periodic-pulsed (f p =1Hz) and steady discharge in entire neighborhood of body. Pulse duty ratio q max (m/s) v(m/s) w(m/s) r/r c θ ω x max (1/s) Direction r/r c θ of16

8 The distributions of the ensemble-averaged velocity component w and axial-vorticity ω x along a radial ray originated from the body center at the azimuthal angle of θ =9 are presented in Fig. 6 for the periodic-pulsed frequency f p =1Hz, and compared with the results of the steady plasma actuation at V p p =14. kv. From Fig. 6(a), w<. Thus, the direction of the resulting flow agrees with that of the plasma discharges. The maximum magnitude of velocity componnent w and axial vorticity ω x induced by the steady plasma actuation are the largest among the four mades considered, and the half-height width of thw w profile induced by the steady plasma actuation is also the largest among the four modes considered. It ia found that the peak values of w and ω x induced by the steady plasma actuation is sharply decreased when the ac voltage V p p is decreased slightly. Table compares the peak values of w and ω x induced by the steady plasma actuation under V p p =14.kV and 14. kv. The ac voltage of V p p =14. kv appears as a threshold above which the peak values of the steady actuation are increased about one order of magnitude. Table. Peak values of w and ω x induced by steady plasma actuation under V p p =14. kv and 14. kv Voltage (kv ) Peak values of w( m/s) (,.) (,.9) Peak values of ω( 1/s) ( 28, 6) ( 62, 132) Under V p p =14. kv and f p =Hz, the distributions of the ensemble-averaged velocity component w and axial-vorticity ω x along a radial ray originated from the body center at the azimuthal angle of θ =9 are presented in Fig. 7. In this case, the positive and negative maximum values of w and ω x induced by the periodic pulsed actuations are, mostly, larger than those of the steady plasma actuation, the half-height width of w profile induced by pulsed actuation is larger, and stronger discrete vortices are formed in the periodic-pulse cases. This observation agrees with the finding over a cylinder model in the absence of external flow measured by hot-wire anemometry in Ref. 8. IV. Phase-Locked-Averaged Results The temporal characteristics of the flow induced by the periodic-pulsed plasma actuations are investigated under V p p =14.kV, f p =1Hz and τ p =.1,. and.9. The PIV sampling frequency is 9 Hz. There are 9 phase angles evenly distributed in a period of the pulse cycle. The phase angles are ψ = n (4 ), where n =, 1, 2,..., 8. The corresponding phase-locked averaged values are calculated from every 9 PIV samples in the sampling order. n = is chosen at the point where the phase-locked-averaged maximum crossflow velocity in the entire neighborhood of the cone is the largest among all the phase angles. In this paper, the PIV sampling time is 1 s. There are altogether 13 samples. At a fixed phase angle there are 1 samples. The phase-locked averages are obtained from the 1 PIV samples. To study the convergence, the phase-locked-averaged maximum crossflow velocity q max in the entire neighborhood of the cone surface for pulse duty ratio τ p =.1, obtained from the 1 samples is considered. At the location of this phase-locked-averaged maximum crossflow velocity, the phase-locked-averaged crossflow velocity from samples of 1, and 1 are listed in Tables 6 for pulse duty ratio τ p =.1, for five phase angles. The convergence of the 1-sample results are quite good. Table 7 compares the phase-locked-averaged maximum crossflow velocity, q max and its location in the entire neighborhood of the cone obtained from the 1 samples for pulse duty ratio τ p =.1 and frequency f p =1Hz with those of the corresponding ensemble-averages. The phase-locked averages varies with phase angle remarkably and deviate from the ensemble average significantly. Figures 8 and 9 present phase-locked-averaged crossflow-velocity and axial-vorticity contours, respectively, for periodic pulse ratio, τ p =.1 at various phase angles. The crossflow velocity vectors are plotted over both contours. The distributions of the phase-locked-averaged velocity component w and axial-vorticity ω x along a radial ray originated from the body center at the azimuthal angle of θ =9 are presented in Figure 1 for the periodic-pulse ratio τ p =.1 and frequency f p =1Hz at various phase angles. The phase-locked-averages have larger variation range than the ensemble averages. Table 8 compares the variation ranges of phaselocked-averaged w and ω x for the five phase angles considered with those of the ensemble-averaged at pulse duty ratio τ p =.1 and pulse frequency f p =1Hz. 8of16

9 -.2 w(m/s) τ p 1% % 9% 1% r/r c (a) Velocity component w τ p 1% % 9% 1% ω x (1/sec) r/r c (b) Axial-vorticity ω x Figure 6. Distributions of ensemble-averaged velocity component w and axial-vorticity ω x along radial line θ =9 for periodic-pulsed discharges of frequency f p =1Hz and steady plasma actuation, V p p =14. kv. 9of16

10 -.1 w(m/s) τ p 1% % 9% 1% r/r c (a) Velocity component w 4 ω x (1/sec) 2 τ p 1% % 9% 1% r/r c (b) Axial-vorticity ω x Figure 7. Distributions of ensemble-averaged velocity component w and axial-vorticity ω x along radial line θ =9 for periodic-pulsed discharges of frequency f p =Hz and steady plasma actuation, V p p =14. kv. 1 of 16

11 q(m/s) q(m/s) (a) ψ = (b) ψ = q(m/s) y(mm) q(m/s) (c) ψ = x(mm) (d) ψ = 24 Figure 8. Phase-locked-averaged crossflow-velocity vectors and contours, τ p =.1, f p =1Hz. 11 of 16

12 ω x (1/sec) ω x (1/sec) (a) ψ = (b) ψ = ω x (1/sec) ω x (1/sec) (c) ψ = (d) ψ = 24 Figure 9. Phase-locked-averaged axial-vorticity contours and crossflow-velocity vectors, τ p =.1, f p =1Hz. 12 of 16

13 Table 6. Convergence of phase-locked-averaged maximum crossflow velocity q max in entire neighborhood of cone vs PIV sampling number for τ p =.1, f p =1Hz. ψ Crossflow velocity ( m/s) by1 sample Crossflow velocity ( m/s) by samples Crossflow velocity ( m/s) by1 samples Crossflow velocity ( m/s) by1 samples Table 7. Comparison of phase-locked-averaged maximum crossflow velocity q max and its location in entire neighborhood of cone vs phase angle ψ with ensemble averages, τ p =.1, f p =1Hz. ψ Ensemble average q max ( m/s) by1 samples r/r c θ Table 8. Variation ranges of phase-locked-averaged w and ω x along radial ray of θ =9 for the five phase angles considered compared with those of the ensemble-averaged, τ p =.1, f p =1Hz. w(m/s) ω x (1/s) Range of phase-locked averages (.6,.18) ( 6, 78) Range of ensemble averages (.3, ) ( 3, ) The profiles of the phase-locked-averaged velocity component w(r/r c ) and axial vorticity ω x (r/r c )havethe highest peak and the widest thickness at ψ = among the phase angles considered. This indicates that the phase angle ψ = may be located at the initiation instant of the pulsed discharge. In the present experiments, the plasma actuation and the PIV velocimetry are operated independently, and, thus, the exact phase angle corresponding to the pulse initiation is unknown. The transient response of the periodic pulse is captured using PIV. The phase-locked-averaged maximum crossflow velocity, q max and maximum magnitude of axial vorticity, ω x max in the entire neighborhood of the body versus phase angle ψ are shown in Fig. 11 for pulse duty ratio of τ p =.1,. and.9 and frequency f p =1Hz. The phase-locked-averaged maximum values are, for most phase angles, much larger than the corresponding ensemble-averaged. For example, at τ p =., the phase-locked-averaged q max varies between.2 m/s and 2.8 m/s and ω x max varies between /s and 36 /s over a pulse period, while the corresponding ensemble-averged q max =.934 m/s and ω x max = 196; /s. The results reveal the effect of flow hysteresis. This is observed by that q max and ω x max remain non-zero even after the actuator is turned off. The input power can be saved by using periodic pulsed actuation that is able to utilize the flow hysteresis. This flow hyteressis under unsteady plasma actuation was also observed using high-bandwidth pressure sensors over an airfoil model in wind tunnel test in Ref. 11. V. Conclusions The flow induced by a single millisecond periodic-pulsed plasma actuator mounted over a 2 cone tip along azimuthal angle of 12 is investigated in atmospheric air without external flow. A two-dimensional particle image velocimetry is used to visualize the flow in the neighborhood of the cone surfase. 1. There exists a threshold of the ac voltage (V p p =14. kv ), over which the ensemble-averaged tangential velocity and axial vorticity induced by steady actuation along a radial line perendicular to the body symmetry plane are increased sharply. 2. At the threshold voltage and appropriate pulse-repetition frequency, the emsemble-averaged maximum 13 of 16

14 w(m/s) ψ ( ) r/r c (a) Velocity component w 6 ω x (1/sec) 4 2 ψ ( ) r/r c (b) Axial-vorticity ω x Figure 1. Distributions of phase-locked-averaged velocity component w and axial-vorticity ω x along radial line θ =9, τ p =.1, f p =1Hz. 14 of 16

15 τ p 3 1% % 9% τ p 1% % 9% q max (m/s) 2 ω max ψ ( ) ψ ( ) (a) q max (b) ω x max Figure 11. Phase-locked-averaged q max and ω x max vs. ψ in the entire neighhood of the body, τ p =.1,. and.9, f p =1Hz. crossflow velocity and maximum magnitude of axial vorticity induced by the periodic pulsed actuations are, mostly, higher than those of the steady plasma actuation. 3. The phase-locked-averaged maximum crossflow velocity and maximum magnitude of axial vorticity induced by periodic-pulsed discharge are, in general, larger than those of the corresponding ensemble averages. 4. The phase-locked-averaged maximum crossflow velocity and maximum magnitude of axial vorticity induced by periodic-pulsed discharge remain non-zero even after the actuator is turned off. This flow hysteresis can be used to save the power input. Acknowledgments The present work is supported by the Specialized Research Fund for Doctoral Program of Higher Education, SPFDP , and the Foundation for Fundamental Research of the Northwestern Polytechnical University, NPU-FFR-W1812. References 1 Bernhardt, J. E. and Williams, D. R., Proportional control of asymmetric forebody vortices, AIAA Journal, Vol. 36, No. 11, Nov. 1998, pp Hanff, E., Lee, R., and Kind, R. J., Investigations on a dynamic forebodey flow control system, Proceedings of the 18th International Congress on Instrumentation in Aerospace Simulation Facilities, Inst. of Electrical and Electronics Engineers, Piscataway, NJ, 1999, pp. 28/1-28/9. 3 Ming, X. and Gu, Y., An innovative control technique for slender bodies at high angle of attack, AIAA Paper , June Liu, F., Luo, S.J.,Gao, C., Meng, X.S., Hao, J.N., Wang, J.L. and Zhao, Z.J., Flow control over a conical forebody using duty-cycled plasma actuators, AIAA Journal, Vol. 46, No. 11, Nov. 28, pp Enloe, C.L., McLaughlin, T.E., VanDyken, R.D., Kachner, K.D., Jumper, E.J. and Corke, T.C., Mechanisms and responses of a single dielectric barrier plasma actuator: plasma morphology, AIAA Journal, Vol. 42, No. 3, Mar. 24, pp Pons, J., Moreau, E. and Touchard, G., Asymmetric surface dielectric barrier discharge in air at atmospheric pressure: electrical properties and induced airflow characteristics, J. Phys. D: Appl. Phys., Vol. 38, 2, pp Abe, T., Takizawa, Y., Sato, S. and Kimura, S., Experimental study for momentum transfer in a dielectric barrier discharge plasma actuators, AIAA Journal, Vol. 4, No. 8, Sep. 28, pp of 16

16 8 Thomas, F.O., Kozlov, A. and Corke, T.C., Plasma actuators for cylinder flow control and noise reduction, AIAA Journal, Vol. 4, No. 8, Aug. 28, pp Roth, J.R., Aerodynamic flow acceleration using paraelectric and peristaltic electrohydrodynamic effect of a one atmosphere uniform glow discharge plasma, Phys. Plasma, Vol. 1, 23, pp Post, M.L. and Corke, T.C., Separation control using plasma actuators stationary and oscillating airfoil, AIAA Paper , Jan Patel, M.P., Ng, T.T., Vasudevan, S., Corke, T.C.. Post, M.L., McLaughlin, T.E. and Suchomel, C.F., Scaling effects of an aerodynamic plasma actuation, AIAA Paper 27-63, Jan of 16