|Ph.D Student||Ifergan Oshri|
|Subject||Numerical and Experimental Investigation of Arc|
Plasma Wind Tunnels
|Department||Department of Mechanical Engineering||Supervisor||Professor David Greenblatt|
|Full Thesis text|
Hot plasma wind tunnels incorporate a tube which includes a constrictor that is characterized by a voltage gradient between its inlet and outlet created by an anode and a cathode. The tube contains high-temperature air (>7000 K) that is electrically conductive due to the ionization processes that begin at these temperatures, resulting in an electric arc.
This research is divided into two distinct, but interrelated, parts: one computational and one experimental. The computational objective was to develop, verify and validate a Simplified Numerical Model for the purpose of providing rapid and accurate solutions of the flow in a constrictor in the presence of an electric arc. The model strikes the middle ground between over-simplified empirically-based correlations and advanced computationally expensive simulations. By assuming a fully developed flowfield, the energy equation is solved in conjunction with a plasma local-equilibrium thermodynamic model and Ohm’s law. Despite its relative simplicity, the model attained excellent correspondence with much more sophisticated and elaborate CFD-based models and captured the essential details of the electro-aerodynamic coupling and the influence of the controlling parameters. Furthermore, voltage, enthalpy and efficiency calculations showed excellent correspondence with published experimental data as well as in-house data generated from the Technion Arc Plasma Tunnel (TAPT).
The experimental objective of this thesis was to simultaneously quantify voltage and heat losses along the length of the TAPT constrictor, for air and argon gas. Thus, by acquiring time-locked data along the constrictor, transient processes associated with the arc during ignition, extinction and nominally steady-state operation were documented for the first time. The voltage data consistently showed a significant, sharp jump between the cathode and the first constrictor segment caused by electrons moving from cathode surface and ions arriving at arc column. A smaller rise was seen in the near-anode region due to electrons moving from the arc to the anode. Downstream of the first segment, the voltage increased in a far less dramatic fashion and in some cases, in particular for argon gas, it was almost linear. The high spatial resolution voltage measurements allowed us to identify a wave-like structure in the upstream part of the constrictor. Although no clear explanation was found, it was postulated that electrode erosion produced an unstable arc geometry that was further distorted by the imposed magnetic fields. Alternatively, the wave-like structure could be related to vortex precession commonly observed in swirl-stabilized flows. Measurements of heat transfer to the wall, based on differences in cooling water temperature across each segment, revealed a mainly uniform heat-transfer distribution along the constrictor, with elevated heat transfer near the cathode and anode due to the strong electric field strengths in the near-cathode and near-anode regions. By carefully varying either mass flowrate or current, these measurements also allowed us, for the first time, to quantify differences in radiation and convection heat transfer.
A comparison of both voltage and heat loss data along the length of the constrictor with the Simplified Model results, revealed some model deficiencies, even though the integral values’ comparisons were excellent.