Advanced Techniques for Hall Thruster Research and Development

Abstract

Hall thrusters are the most common form of electric propulsion on spacecraft currently in Earth orbit, they offer excellent efficiency, high thrust to power, and have relatively simple power processing units. They have seen significant development since their first commercial flight in 1972, however, methods for characterizing their operation for both development and flight have remained relatively consistent. Therefore, advanced techniques have been developed to more effectively, efficiently, and expeditiously characterize thruster operation. These methods have been advanced through four primary approaches. The first approach was the development of a magnetically shielded Hall thruster with a movable inner pole to investigate the influence of magnetic field shape on thruster operation. The use of magnetically shielded geometries has been demonstrated to drastically reduce the rate of erosion on Hall thruster channel walls. However, the switch to a magnetically shielded configuration can increase plume divergence contributing to a decrease in efficiency and thrust. Varying the magnetic field shape near the exit of the thruster has been demonstrated to decrease plume divergence, however, the impact on thruster behavior is unknown. Three test campaigns on iterations of the ACME thruster, operating on both xenon and krypton, investigate the plume divergence, performance and efficiency changes, and map the plume of the thruster across a range of pole positions. The divergence, voltage, and current utilization efficiencies have strong dependencies on the relative pole position across all operating points. The dependence of anode efficiency on pole position can be mapped to the other efficiencies within the uncertainty of the experiment. The change in inner pole position of -1 mm on xenon and - 4 mm on krypton has been shown to significantly increase the anode efficiency of the thruster, primarily through divergence and voltage utilization efficiency gains. The second approach assesses the standard methods for characterizing thruster efficiency using plume diagnostics. The standard practice of measuring a centerline Ion Energy Distribution Function (IEDF) is insufficient for electric propulsion systems with unknown or highly divergent plume structures. This method is compared to spatially resolved IEDF measurements using a swept retarding potential analyzer to generate an angularly resolved IEDF. Using an adjustable Hall thruster, plumes ranging from highly divergent to over-focused were fully characterized and compared to the anode efficiency calculated using thrust stand measurements. Both centerline and spatially resolved measurements of the IEDF were sufficient to accurately measure the voltage utilization efficiency on well focused plumes. As the plume diverged, more complex plume structures were observed, and only the spatially resolved measurement maintained agreement with the thrust stand based efficiency. . The third approach to expedite Hall thruster development created a method to rapidly and autonomously optimize thruster performance. This method combines rapid thrust measurement, thruster-in-the-loop control, and derivative-free optimization schemes to automate thruster optimization and rapidly find operational areas of interest, thus reducing the reliance on current-voltage-magnetic field (IVB) maps. Rapid thrust measurements are achieved by comparing test points to a known operating point, allowing fast and accurate thrust measurement while minimizing the effects of long-term thermal drift. Fast mapping of a Hall thruster's operation was demonstrated at 72 test points per hour, with an average thrust measurement error of < 1% compared to conventional thrust measurements. Two-dimensional (thruster discharge voltage and magnetic field strength) Nelder-Mead and Powell optimization schemes are shown to converge rapidly to maxima in total efficiency or specific impulse in fewer than 15 test points. The Powell optimization scheme remained effective in five dimensions, further increasing the peak thruster efficiency while adjusting three additional thruster dimensions (keeper current, cathode flow fraction, and magnetic field skew). The fourth approach developed a method for autonomous optimization and modeling of an electric thruster that applies Bayesian optimization on a Gaussian Process Regression model generated in real time from experimental telemetry. The method can be combined with a prescribed objective function and optimization scheme to optimize the thruster for different mission objectives. A notional extrasolar probe mission powered by a Hall effect thruster is considered as an example where the goal of the optimization is to find the propellant gas mixture (argon:krypton:xenon) that minimizes overall mission cost. The results show that a thruster running on a mixture ratio of 11:87:2 benefits from a 7% reduction in total cost compared to the same thruster running on pure xenon. Analysis of the model reveals how the optimal propellant mixture depends strongly on propellant storage technologies, fluctuations in propellant price, and launch costs. Results from this analysis match trends seen in the commercial market with the move to cheaper propellants.