Ionic wind produced by a DC needle-to-plate corona discharge with a gap of 15 mm

Abstract

The goal of the present study is to better understand the electrohydrodynamic (EHD) phenomena occurring in 15 mm gap point-to-plate corona discharges supplied by positive and negative DC voltages, and to link them to the discharge regime. For that, discharge current measurements have been conducted and a 20 kHz Particle Image Velocimetry system has been used to characterize the ionic wind produced by the discharges. The main results are as follows: (a) the Townsend’s law can always correctly interpolate the experimental I–V characteristics of a negative corona, even in presence of Trichel pulses, (b) in the case of a positive streamer discharge, the current does not follow the Townsend’s law as it evolves in Vk with k = 4 in the present study, (c) Indeed, the discharge current becomes higher than the Townsend’s law when the glow-to-streamer regime transition occurs, (d) the ionic wind is unsteady, more especially in the case of a positive corona for which it seems that its velocity is pulsed at the same frequency than the streamer one, (e) when the positive high voltage is switched on, a strong streamer occurs at the end of the voltage rising (t = 25 µs for +6 kV), resulting in an over-velocity region upstream the head of the ionic wind jet that progresses in quiescent air toward the plate, (f) the topology of the time-averaged ionic wind is fully different compared to the one observed in a previous study where the gap was equal to 25 mm, (g) in the case of a negative corona, the ionic wind velocity is nearly constant in the electrode gap, (h) in the case of the positive corona, the results are even more surprising since the velocity is minimum at the tip and increases when one approaches the grounded plate, indicating that there is a significant space charge remained by the streamers in the second half of the inter-electrode region, (i) generally speaking, this study highlights that the spatio-temporal characteristics of the EHD force and the resulting ionic wind depend on the distribution of the space charge between both electrodes, the latter being linked to the voltage polarity, the discharge regime and the electrode gap.

Effect of the high voltage waveform on the ionic wind produced by a needle-to-plate dielectric barrier discharge

“…It can be observed that it increases firstly logarithmically with frequency (up to about 100 Hz, 300 Hz and 500 Hz for the square, sinus and triangle waveforms, respectively), and then, the slope suddenly increases. The last remark is general: the power consumption is comparable in amplitude to the power consumed by DC coronas 27 , as it is weak and limited to about 1 W.

Figure 2 Electrical power consumption versus voltage amplitude V AC for f AC = 200 Hz ( a ) and electrical power consumption versus frequency f AC for V AC = 12 kV ( b ).

…”

“…This is highlighted by high current pulses with magnitudes reaching about 6 mA here and a frequency varying between 5 and 7 kHz (in the case of the sine high voltage). On the other hand, one can observe that the negative corona discharge results in more numerous current pulses but weaker in magnitude, each of them corresponding to a Trichel pulse, as in negative DC coronas 27 . When focusing on these Trichel pulses, one can see that their amplitude is limited to about 0.2 mA with a frequency between 100 and 400 kHz, depending on the v AC (t) value.…”

“…The needle is in a horizontal position and the plate electrode is placed perpendicularly to the needle. The needle is made of tungsten and its tip has a curvature radius of about 100 µm, when the grounded electrode is a stainless steel disk of 80 mm in diameter, as in Defoort et al 27 . However, here, the plate electrode is covered with a 5-mm thick dielectric plate made of Polymethyl Methacrylate (PMMA, blue shape in Fig.…”

“…To characterize the ionic wind produced by the discharge, the setup is exactly the same as in Defoort et al 27 . A LaVision time-resolved particle imaging velocimetry (PIV) system is used.…”

“…It has been demonstrated several times that these oil droplets are not significantly impacted by the electrostatic phenomenon. Therefore, it is assumed that they follow well the ionic wind produced by the discharge 25 , 27 . The light used to illuminate the seeding particles is produced by a 532 nm Nd-YAG laser generator (Continuum Mesa).…”