Recently, much attention has been devoted to the nanosecond pulsed dielectric barrier discharge (ns-DBD) plasma actuators as they demonstrated superior control feature at higher speed regimes as compared to the alternating current (AC) DBD actuators. This type of actuator usually has similar configurations as the AC-DBD device, but is driven by high-voltage repetitive pulses with voltage of up to 50kVs and rise time from a few to tens of nanoseconds. It has been generally agreed that the main control mechanism of the ns-DBD plasma actuator is through Joule heating, which is different from the AC-DBD by momentum adding from the induced ionic wind. Joule heating causes steep gradients in pressure and temperature inside the heated gas volume. As a result, moving shock waves are generated and interact with the external flow. The very recent experimental and numerical studies in quiescent air by Zhao el al. (2014) and Zheng el al. (2014) further revealed that the shock wave is basically a kind of micro blast wave. It is observed that the blast wave can be fairly strong in terms of the induced fluid velocity (of up to hundreds m/s) and the overpressure (of up to tens of kPas) shortly after its initiation but decays very fast during a few microseconds. The shock induced perturbations (overpressure and induced velocity) are restricted to a narrow region behind the shock front, lasting for a few microseconds only, showing extremely localized effects in space and transient effects in time.
While the previous studies have focused on the effectiveness of the actuators on specific applications, how the generated shock waves influence and interact with the external flow has not been thoroughly investigated. Therefore, this paper is to address this issue through studying the suppression of the separated flow over a ramp with nanosecond pulsed plasma actuators in a wind tunnel using smoke-wire visualization, Schlieren imaging technique and particle imaging velocimetry (PIV).
Experiments were performed in a small scale blow-down low speed wind tunnel located at Temasek Laboratories of the National University of Singapore. This wind tunnel has a contraction ratio of 9.8 and a test section of 0.16m × 0.16m with turbulence intensity level less than 0.25% and maximum speed of 30m/s. The test section has a length of 0.75 m and was made from Perspex to facilitate flow visualizations and PIV measurements. Figure 1 shows the ramp dimensions used in experiments. The ns-DBD plasma actuator comprises two electrodes (copper foil with a thickness of 66µm) mounted on both sides of a dielectric layer. This dielectric layer is made of three layers of Kapton films (Kapton 500FN131, 175µm thick) and cover the whole plateau and downstream ramp. The widths of the exposed and the encapsulated electrodes are 8mm and 15mm, respectively. The connection point between the two electrodes is located at the turning point of the ramp with the gap of about 0 to 0.1mm.
The nanosecond pulse to drive the ns-DBD plasma actuator was achieved by means of a nanosecond pulse generator (NPG-18/3500). This generator produces high-voltage pulses with the peak voltage from 12kV to 20kV at matched 75 Ohm load, the pulse rise time about 4 ns, the repetition rate of up to 3.5 kHz and the energy of up to 30mJ/pulse. Note that the peak voltage, current and the associated energy are strongly load-dependent. The applied voltage and current were measured using a high voltage probe (Tektronix P6015A) and a current shunt probe (Megaimpulse CS-10/500), respectively. In order to apply this probe to measure a pulse with nanosecond rise times, a calibration was conducted using a square wave (30Vp-p) with rise time ≤ 10 ns and with duration of 100 ns (generated from Digital delay generator Model DG645) following the manual of Tektronix P6015A.
Typical smoke-wire visualization results are shown in Fig. 2 with freestream velocity of 3m/s. The unsteady flow separation is clearly observed at the downstream of the ramp at plasma off condition, while the separated flow is suppressed with plasma on. For higher speed (say 30m/s), no clear images were obtained as the smoke diffuses very fast. Detail Schlieren imaging and PIV measurements are undergoing at freestream of 30m/s. Here, the Schlieren imaging technique is used to study how the micro ballast waves generated by the nanosecond pulsed plasma evolve, and PIV is used to study how the external flow is affected by the pulsed plasma.
Fig. 1 Sketch of the ramp model with DBD actuators.
Fig. 2 Smoke-wire visualizations with (a) plasma off and (b) plasma on at U = 3m/s.
References
Zhao, Z.J., Li, J., Zheng, J.G., Cui, Y.D., and Khoo, B.C., “Study of shock and induced flow dynamics by pulsed nanosecond DBD plasma actuators,” AIAA SciTech 2014, National Harbor, Maryland, US, 13-17 January, 2014. AIAA- 0402, DOI: 10.2514/6.2014-0402.
Zheng, J.G., Zhao, Z.J., Li, J., Cui, Y.D., and Khoo, B.C., “Numerical simulation of nanosecond pulsed dielectric barrier discharge actuator in a quiescent flow,” Physics of Fluids, Vol. 26, 036102, 2014, DOI:10.1063/1.4867708.
