AIAA-2005-565-450Acoustic Testing of the dielectric barrier discharge (DBD) plasma actuator

推进器的声波特性

43rd AIAA Aerospace Sciences Meeting and Exhibit 10 - 13 January 2005, Reno, NevadaAIAA 2005-565Acoustic testing of the dielectric barrier discharge (DBD) plasma actuatorCorrie Baird * University of Colorado at Colorado Springs, Colorado Springs, Colorado 80918 C. L. Enloe and Thomas E. McLaughlin U.S. Air Force Academy, Colorado Springs, Colorado 80840 and James W. Baughn§ University of California, Davis, California 95616 The dielectric barrier discharge (DBD) plasma actuator has been shown to be effective for flow control. Much remains unanswered, however, as to how the actuator couples momentum into air. A better understanding of the coupling mechanism is crucial to determining the performance limitation of the actuator and the breadth of applications to which it can be applied. The small physical volume and transient nature of the actuator plasma make it difficult to make direct measurements. In previous work we have investigated the plasma actuator’s optical emission signature extensively. In this work, we measure and analyze the acoustic emissions, both directional characteristics and waveform, from an actuator in an attempt to shed light on the coupling process. Two sets of measurements were made, each using a different apparatus. Both sets of tests reveal that the actuator adds a larger amount of momentum into the air during the negative-going half of the AC voltage cycle and a smaller amount on the other half of the cycle. It was observed that the acoustic pattern produced is a radiation pattern characteristic of a coherentlydriven system. The results suggest that compressibility effects may play a role in the momentum coupling.I.IntroductionThe dielectric barrier discharge (DBD) plasma actuator has a wide range of demonstrated aerodynamic uses. The DBD plasma actuator has already been shown to reduce and in some cases eliminate the separation bubbles on turbine blades1-2. It has also been shown to be effective in reattaching separated flows at high angles of attack on thin airfoils3-5, providing velocity increments near aerodynamic surfaces in low speed flows to delay separation6-10 and stall11, and to control the phase of vortex shedding from a circular cylinder12. Further investigations of the spatial and temporal aspects have led to an analytical model to describe plasma behavior by Enloe et. al.13-14 However, there is still much unknown about how the actuator couples momentum into air. A better understanding of this interaction will lead to its optimization. This paper attempts to gain insight into the coupling of momentum into air through the actuator’s directional acoustic signature and waveform.This material is declared a work of the U.S. Government and is not subject to copyright protection in the United States.* Graduate Research Assistant, Mechanical Engineering, Student Member AIAA. Professor, Department of Physics. Senior Member AIAA Director, Aeronautics Research Center, Department of Aeronautics, Associate Fellow AIAA § Professor, Department of Mechanical and Aeronautical Engineering, Member AIAA 1 American Institute of Aeronautics and AstronauticsThis material is declared a work of the U.S. Government and is not subject to copyright protection in the United States.

推进器的声波特性

The DBD plasma actuator consists of two thin electrodes placed asymmetrically on an aerodynamic surface. The upper electrode is exposed to the air while the lower electrode is encapsulated by a dielectric material and a substrate material as shown in Figure 1. A high AC voltage (5-10 kV amplitude, with frequency in the range of 1-10 kHz) is applied to the electrodes. The plasma then forms in the region above the insulated electrode. The appearance of the plasma is accompanied by a coupling of directed momentum into the surrounding air.Figure 1: Diagram of the plasma actuator configurationValue of acoustic testing The plasma when energized emits a purple glow that is accompanied by a very distinct tone. It is evident that the acoustic tone changes based on orientation, forcing voltage, and forcing frequency. This change in the emitted sound with various parameters led to the thought that the acoustic waves may provide an indication of the interaction between the actuator and the air. As Enloe et. al.13-14 points out, there are two major contenders for momentum coupling mechanisms for the plasma: either the air is given an amount of momentum on one half of the forcing cycle and a similar or smaller amount in the same direction on the other half; or there is a larger momentum increase followed by a smaller decrease of momentum over the course of a cycle. We refer to these as the PUSHpush or the PUSH-pull options. The former mechanism would be associated by a net force on the bulk plasma, while the latter would tend to be the result of the current carriers in the gap interacting with the air. The authors believed that the plasma actuator’s acoustic emissions had the potential of sorting out these two possibilities, and hence contributing to understanding the underlying mechanism.II.Experimental Set-UpA. Overview of set-up Two sets of acoustic experiments were completed at the US Air Force Academy each using a different apparatus. The first set provided directional characteristics and a fast Fourier transform of the average waveform. The second set was designed to address concerns about possible reflections during the first set, the frequency response of the microphone used, and to determine and examine the acoustic waveform in more detail. The first set of experiments was completed in a sound-absorbing structure, designed to suppress reflections that would interfere with the recording of the actuator’s acoustic emissions. Figure 2 shows the configuration of the booth used. The plasma actuator with its accompanying electrical equipment was placed on a stand on the bottom of the booth. The acoustic sensor was suspended from wires approximately one meter above the actuator. The sensor’s output was fed through an oscilloscope which is connected to a computer-based data acquisition system (DAQ). 2 American Institute of Aeronautics and Astronautics

推进器的声波特性

Front ViewSide View1.22 mAcoustic sensorAcoustic sensor1.83 m 1.22 m Non-reflective booth Plasma Actuator Plasma Actuator’s standAngle of rotation (θ)Plasma actuator StandFigure 2: Diagram of the arrangement used for the first set of measurementThe second set of acoustic experiments was conducted in a large room with sound absorbing material covering the top half of the walls and the ceiling. The actuator was placed on one end of a rod with the microphone placed at the other end of the same rod one meter apart. The rod (with actuator and microphone) was connected to a tripod and elevated one meter above the floor. This set up is shown in figure 3.Figure 3: Arrangement used for second set of measurementsB. DBD plasma actuator description For all of the measurements reported here the electrodes were made of copper foil tape, 25 cm long and 1 cm wide for the upper electrode, and 3 cm wide for the lower electrode. For most of the measurements ordinary pane glass (approximately 30x8 cm) was used for the dielectric material and either thin glass epoxy or Kapton tape for the substrate material (Figure 1). The actuator was placed on a stand which could be rotated about the long axis of the actuator from 0 to 180 degrees. Five and ten degree increments in the angle, θ, were used. Figure 4 shows the actuator placement with respect to angle (the microphone is at the outer edge of the polar plot).3 American Institute of Aeronautics and Astronautics

推进器的声波特性

C. Sensor and output The acoustic emissions in the first set of tests were measured with a generic dynamic microphone placed in an aluminum box to eliminate electrical noise. This microphone was connected to an oscilloscope which in turn was connected to a data acquisition system (DAQ) which recorded the waveform averaged over a number of cycles. The microphone output voltage (peak-to-peak) was determined from the recorded waveforms. The raw waveform and its calculated fast Fourier transform were recorded by Labview. For the second set of experiments an ElectroVoice ND468 supercardioid microphone, with a frequency response (3 DB) of 22 kHz, was used. The output of the microphone was fed to a pre- Figure 4: Polar plot showing the orientation of amplifier then to an HP 3052B oscilloscope connected to a the actuator with respect to angle and induced laptop that recorded the raw data in Excel. To eliminate velocity. electro-magnetic noise post-processing filtering of high frequency noise (MHz and higher) was used and then the peakto-peak voltage of the remaining waveform determined. D. Test parameters The parameters varied in the first set of tests were the forcing voltage, dielectric material size, and forcing voltage waveform. Forcing frequency was held constant at 5 kHz because it is near the optimum point for our stepup transformer, and the signal was ensemble averaged over 64 traces to clean up the signal. The tests using the second experimental set-up held the voltage constant at 5 kV while varying the frequency from 5 kHz to 8 kHz. In this case ten thousand sample points were used for a single sweep of two to three cycles. Both tests included taking measurements at different angles; five degree increments for the first set, and ten degree increments for the second set of tests.III.ResultsA free field, a region surrounding the source where the sound pattern emulates that of an open space15 with no walls, was needed to avoid reflection from any hard surfaces. Realizing the outdoors is the best free field available because there are no hard surfaces near by except for the ground, a reference data set was taken to compare to indoor experiments. Figure 5 shows the fast Fourier transform (FFT) plot from this data. The distinctive properties of this plot are that there is a first harmonic at the same frequency as the forcing frequency and a second harmonic close to the strength of the fundamental (or forcing frequency). The third and fourth harmonics are more than an order of magnitude lower than the second harmonic. Testing was then conducted indoors while taking precautions to ensure that the absorbing surfaces of the booth were indeed eliminating acoustic reflections from a near-by wall. This was done by completing two set of tests where the plasma actuator rests on the stand described previously, and response was measured at angles 0 to 180 degrees in 5 degree increments. The sound-deadening material was removed on the side nearest one wall and then replaced. The resulting FFT plots are shown in figure 6. With the material removed, the second harmonic is stronger than the fundamental, indicating some additional signal Figure 5: Frequency spectrum in a free field most likely coming from reflection. With the material replaced, (outdoors) for first set of acoustic measurements. the FFT plot is similar to the free field plot of figure 6, indicating some attenuation of reflected energy. 4 American Institute of Aeronautics and Astronautics

推进器的声波特性

1.00E+00 1.00E-01 1.00E-02 A m p litu d e 1.00E-03 1.00E-04 1.00E-05 1.00E-06 1.00E-07 1.00E-08 1.00E-09 0 5000 10000 15000 20000 25000 30000 Frequency (Hz)1.00E+00 1.00E-01 1.00E-02 A m p litu d e 1.00E-03 1.00E-04 1.00E-05 1.00E-06 1.00E-07 1.00E-08 1.00E-09 0 5000 10000 15000 20000 25000 30000 Frequency (Hz)(a) (b) Figure 6: Frequency Spectrum in a sound booth (a) with sound absorbing material removed, (b) with material replaced (sinusoidal forcing voltage of 15 kV, at 5 kHz 45 degrees) Since the plot shown in figure 6b is quite similar to that of the free field, the booth appears to be successful in minimizing reflections as it was designed to do. Therefore, the results from tests taken in the booth are considered an accurate measurement for the direct acoustic emissions from the actuator. The next set of measurements with this first apparatus compared results using different forcing voltages. A sinusoidal forcing voltage at a frequency of 5 kHz was used to drive the 30x8 cm actuator with the glass dielectric material. Data was taken for three forcing voltages, Vppf, of 5, 7, and 10 kV. The measured microphone peak-topeak voltage, Vppm, is plotted versus the angle on a polar plot (see Figure 4) to determine the directional characteristics of the acoustic emissions from the actuator with regard to orientation along its longitudinal axis. Figure 7 shows a polar plot of the peak-to-peak microphone voltage as a function of the orientation angle. The angle at which the maximum peak-to-peak voltage occurs did not change as the voltage was increased. The amplitude difference between the three forcing voltages is large and apparently nonlinear. However, normalized plots (not shown) indicate that the pattern is fairly consistent with a large frontal lobe at approximately 40 degrees above horizontal in the direction of the flow. At the lowest voltage there appears to be a small back lobe.Figure 7: Polar plot of Vmpp, comparing acoustic emissions for different forcing voltages (Vfpp) at 5 kHz In another set of measurements with this first apparatus the size of the actuator surface was changed while maintaining the sinusoidal forcing voltage at 7 kV with a frequency of 5 kHz. Three actuator sizes were used; the original 30x8 cm, and two additional actuators (28x15 cm and 30x4 cm). The polar plot of these results, figure 8, shows the difference in the measured amplitudes of the acoustic emissions with angle. Although the directional pattern of the acoustic emissions does not appear to change much with the actuator size, the magnitude does. Interestingly enough, the smallest actuator (30x4 cm) falls between the other two actuators. 5 American Institute of Aeronautics and Astronautics

推进器的声波特性

Figure 8: Polar plot of Vmpp comparing acoustic emissions for different actuator sizes at Vfpp of 7 kV and 5 kHzNext the forcing voltage waveform was changed while using the 30x8 cm actuator. A forcing voltage of 7 kV was used at a frequency of 5 kHz. The waveforms used included square, negative ramp, positive sawtooth, triangle, and the sinusoidal waveform. These results showed that the square waveform produced the highest amplitude and the negative sawtooth the lowest amplitude. Previous force measurements16 showed an opposite relationship for the measured forces. The square wave produced the least amount of tangential thrust while the positive sawtooth produced the most. The directional characteristics of the acoustic emissions for all five forcing voltage waveforms were fairly consistent with a frontal lobe and a smaller back lobe similar to the patterns shown above. The next set of measurements used the second experimental apparatus described above (Figure 3). For these measurements the actuator forcing voltage was sinusoidal at 5 kV and an actuator size of 30x8 cm was used. The measured directional characteristics are shown in figure 9. A similar bimodal pattern was observed.Figure 9: Polar plot of Vmpp comparing acoustic emissions for different forcing frequencies at a Vfpp of 5 kV.Since this similar bimodal pattern is seen in the polar plots for both set of measurements it is concluded that the effects of acoustic reflections, if any, are minimal. The results in figure 9 were obtained in a different environment, with a different apparatus, microphone and set-up. In the measurements with this second apparatus the waveform for a single cycle was recorded. The waveform was recorded at two frequencies and at angles of ten degree increments from zero to 180 degrees. Figure 10 shows the effect of the frequency change from 5 to 8 kHz on the measured waveform at an angle near the peak of the frontal lobe (40 degrees). 6 American Institute of Aeronautics and Astronautics

推进器的声波特性

Figure 10: Filtered data showing the microphone waveform (40 degrees, 5 kHz and 8 kHz 5 kV) At both frequencies (5 KHz and 8 KHz) there are two positive peaks in the acoustic waveform, one being larger than the other, and two negative dips or valleys. Although similar, the difference in the acoustic waveform for the two different frequencies can be seen by examining the valleys of the waveform. For the 5 kHz case the first negative valley is nearly as large as the second but for the 8 kHz case the first negative valley is almost nonexistent and the second valley is much larger. Figure 11 and Figure 12 show the effect of the directional angle on the waveform for the two frequencies of 5 and 8 kHz, respectively. For the 5 kHz case the change in waveform with direction seems minimal. In the 8 kHz case the waveform changes with frequency with the first valley almost nonexistent at 40 degrees but rather large at 140 degrees. A possible cause of this is discussed in the next section.Figure 11: Microphone output waveform for different angles for 5 kHz at 5 kV.7 American Institute of Aeronautics and Astronautics

推进器的声波特性

Figure 12: Microphone output waveform for different angles for 8 kHz at 5 kV.IV.Analysis and DiscussionThese acoustical emission spectral measurements (FFT’s), polar plots and waveforms from a plasma actuator as presented in the previous section can provide insight into how the plasma actuator is coupling momentum into the surrounding air. The FFT’s showed two roughly equal peaks, one at the driving frequency and one at twice the driving frequency. Some basic modeling gives insight into the interpretation of this Fourier decomposition of the acoustic emissions. Two assumptions made in this modeling are: 1) the Fourier components of the air acceleration are preserved in the Fourier components of the sound, and 2) that the net force used to obtain the time varying response can be roughly averaged over time. Based on these assumptions, we assumed various inputs to the model of such a system and compared the spectral content of the response in this model to that observed experimentally. Applying a square wave-like input function for force vs. time is a reasonable choice based on our previous observations13,14. If the air is being given a large push and a small push, the second harmonic would be expected to have a strong presence in the acoustic emissions. Assuming the negativegoing half of the cycle is not as effective in producing the force (and therefore acoustic signal) as the positive-going half (a reasonable assumption based on previous results13,14) ; the time history and resulting FFT plot shown in figure 13 is obtained. This matches our measured FFT (Figure 5) supporting the idea that the plasma actuator is in fact interacting with the air by giving it a big “push” and then a second but smaller “push”. 8 American Institute of Aeronautics and Astronautics

推进器的声波特性

As a comparison, if instead the force were a large push and a small pull, there would be a very small contribution from the second harmonic, as shown in figure 14. Since this is contrary to what we observe, we believe that the acoustic data suggest that a “PUSH-push” mechanism is operating in the plasma actuator. The positive-going stroke is considerably less effective at moving air than the negative-going stroke. This is consistent with previous data showing that the discharge is reasonably uniform on the positive going cycle while on the negative going cycle the discharge is more “patchy”13. The polar plots also support a model with a “push” in the direction of momentum coupling (higher acoustic emissions in the direction of the flow). The waveform data and its change with frequency and direction further support this concept of a PUSH-push model with a strong initial PUSH and then a smaller secondary push for each cycle. The data in Figure 10 for 5 KHz suggests a strong positive compression, followed by a rarefaction, then a second smaller compression followed by a stronger rarefaction. At 5 KHz we see little effect of direction on this waveform (see Figure 11). However, at 8 KHz, not only is the waveform changed from the 5 KHz case (Figure 12), but there is also a directional effect in the waveform as well. The Figure 14: Force versus time for a PUSH-pull situation, with a relatively waveform viewed from the direction in small contribution by the “back stroke.” Again, the frequency spectrum is which the flow accelerates now shows also shown. a much smaller second compression followed by larger rarefaction. This does not appear to occur in the acoustic emissions observed upstream of the actuator for the 8 KHz case. This suggests the possibility that at the higher frequency the first compression does not have time to move significantly out of the way before the second compression occurs. If so, the compressible nature of the air is a factor in the momentum coupling.V.ConclusionThe directional and spectral acoustic emissions from a plasma actuator have been measured using two different experimental setups, microphones and procedures. The results from these measurements provide insight into the nature of the momentum coupling of the actuator and the air. It appears from these results that the dielectric barrier discharge plasma actuator is interacting with the air by adding a larger push (compression and acceleration) followed by a second but smaller push (“PUSH-push”) during each cycle. It also appears that compressibility effects of the air during the acceleration may be an important part of the momentum coupling process.AcknowledgmentsThis work was sponsored by the Air Force Office of Scientific Research. Dr. John Schmisseur, program manager, under project order plasma theme funding. This support is gratefully acknowledged. The authors also gratefully acknowledge the assistance of Cadets First Class Timothy Sutphen and Joshua Olson who conducted the measurements with the second apparatus and performed data reduction as part of their AE471 laboratory project under the direction of the authors.9 American Institute of Aeronautics and Astronautics

推进器的声波特性

References1. List, J., Byerley, A.R., McLaughlin, T.E., and VanDyken, R., “Using Plasma Actuators Flaps to Control Laminar Separation on Turbine Blades in a Linear Cascade”, AIAA 2003-1026, 41st Aerospace Sciences Meeting and Exhibit, 6-9 January 2003, Reno NV. 2. J. Huang, T. Corke, and F. Thomas, “Plasma Actuators for Separation Control of Low Pressure Turbine Blades” 41st Aerospace Sciences Meeting & Exhibit, Reno, NV, 2003, AIAA 2003-1027. 3. Corke, T.C. and Matlis, E., “Phased Plasma Arrays for Unsteady Flow Control”, AIAA 2000-2323, Fluids 2000 Conference, 19-22 June 2000, Denver CO. 4. Post, M.L., and Corke, T.C., “Separation Control of High Angle of Attack Airfoil Using Plasma Actuators”, AIAA 20031024, 41st Aerospace Sciences Meeting and Exhibit, 6-9 January 2003, Reno NV. 5. Corke, T.C., Jumper, E.J., Post, M.L., Orlov, D., and McLaughlin, T.E., “Application of Weakly-Ionized Plasmas as Wing Flow Control Devices”, AIAA 2002-0350, 40th Aerospace Sciences Meeting and Exhibit, 7-10 January 2002. 6. Roth, J.R., Sherman, D.M., Wilkinson, S.P., Boundary Layer Flow Control with a One Atmosphere Uniform Glow Discharge Surface Plasma”, AIAA 98-0328, 36th Aerospace Sciences Meeting and Exhibit, January 12-15, 1998, Reno NV. 7. Roth, J.R., Sherman, D.M., Wilkinson, S.P., “Electrohydrodynamic Flow Control with a Glow-Discharge Surface Plasma”, AIAA J. 38 No. 7, July 2000, p1166-1172. 8. Roth, J.R., Sin, H., Chandra, R., Madhan, M., “Flow Re-Attachment and Acceleration by Paraelectric and Peristaltic Electrohydrodynamic (EHD) Effects”, AIAA 2003-0531, 41st Aerospace Sciences Meeting and Exhibit, 6-9 January 2003, Reno NV. 9. S. Wilkinson, “Investigation of an Oscillating Surface Plasma for Turbulent Drag Reduction,” 41st Aerospace Sciences Meeting & Exhibit, Reno, NV, 2003, AIAA 2003-1023. 10. D. Ashpis and L. Hultgren, “Demonstration of Separation Delay with Glow Discharge Plasma Actuators” 41st Aerospace Sciences Meeting & Exhibit, Reno, NV, 2003, AIAA 2003-1025. 11. Van Dyken, R., McLaughlin, T. E., Enloe, C. L., “Parametric Investigations of a single dielectric barrier plasma actuator”, AIAA 2004-846, 42nd Aerospace Sciences Meeting and Exhibit, 5-8 January 2004, Reno NV. 12. Asghar, A., and Jumper, E.J., “Phase Synchronization of Vortex Shedding from Multiple Cylinders Using Plasma Actuators”, AIAA 2003-1028, 41st Aerospace Sciences Meeting and Exhibit, 6-9 January 2003, Reno NV. 13. Enloe, C.L., McLaughlin, T.E., Van Dyken, R.D., Kachner, K.D., Jumper, E.J., Corke, T.C, “Mechanisms and Responses of a Single Dielectric Barrier Plasma Actuator: Plasma Morphology”, AIAA J. 42 No. 3, March 2004, p589-594. 14. Enloe, C.L., McLaughlin, T.E., Van Dyken, R.D., Kachner, K.D., Jumper, E.J., Corke, T.C., Post, M., Haddad, O., “Mechanisms and Responses of a Single Dielectric Barrier Plasma Actuator: Geometric effects”, AIAA J. 42 No. 3, March 2004, p595-604 15. Raichel, Daniel R. The Science and Applications of Acoustics. Springer-Verlag New York, Inc. New York, NY. 2000. 16. Enloe, C.L, McLaughlin T.E, Van Dyken, R.D., Kachner, K.D., Jumper, E.J, Corke, T.C. “Mechanisms and response of a single dielectric barrier plasma”, AIAA 2003-1021, 41st Aerospace Sciences Meeting and Exhibit, 6-9 January 2003, Reno NV.10 American Institute of Aeronautics and Astronautics

本文来源：https://www.bwwdw.com/article/c7u1.html