Nested Channel Hall-effect Thrusters


High power Hall-effect thrusters (HETs) are an enabling technology for longer distance and heavier payload space missions. Historically, HETs consist of a single annular channel, however scaling to higher powers poses several challenges which may be overcome by nesting multiple channels into a single device [1]. This permits independent or simulataneous operation of the individual channels. The approach was first published in the open literature by Raymond Liang [2], as he described the two channel 10kW class X2 thruster developed at PEPL.


While operating the thruster at constant mass flow rate, it was observed that the dual channel mode produced higher thrust than the superposition of thrust values from independent inner and outer channel modes. This suggests a coupling effect between the two channels, which should be studied further.

The next development at PEPL was to build a 100kW class Nested Channel HET (NHT): the X3 described by Roland Florenz in his thesis [3]. An indication of channel coupling was also observed between the three channels of the X3, while operating at constant current: nested channel operation required less fuel than independent operation.

Simulation Approach

Since the proof of concept X2 has been well characterized, the modeling and simulation work will be focused on this thruster.

The axisymmetric hybrid-PIC code HPHall was chosen to simulate NHTs due to its ability to capture oscillatory behavior. The framework was initially developed by Fife at MIT [4], and the code models the heavy species using a particle-in-cell technique, while using a quasi-1D fluid for the electrons. It is impossible to simulate a nested channel device with the current version of HPHall and an improved 2D axial-radial electron model is required to handle the complex magnetic field topology.

Before a full plasma simulation of and NHT is possible, first a single channel plasma simulation was performed, followed by a dual channel neutral simulation.


Inner Channel Plasma Simulation

The first step was to simulate the X2 inner channel. Figure 1 shows the simulation domain, and the computational grid.

Figure 1: Inner channel simulation domain and mesh.

Figure 2 below shows results from an HPHall simulation of the X2 thruster's inner channel. The anode flow rate is fixed at 7 mg/s and the discharge voltage is 200 V. The neutral cathode flow is also included in the simulation, at a rate of 10% that of the anode flow. Liang[2] measured a facility background pressure of 1.5*10-5 Torr and this value is also used in the simulation. The laboratory measured thrust was 92.0± 3.00 mN, and the simulation produced a value of 81 mN. The measured discharge current was between 5.56 and 5.79 A, while the time averaged value from the simulation was 6.08 A.



Figure 2: Number densities for xenon particles, and streamlines (a) neutrals (b) singly charged ions.

Figure 3 shows a comparison with experimentally determined plasma properties along the centerline of the inner channel. The electron temperature is under-predicted by the simulation, but the plasma potential is matched closely.



Figure 3: Properties along the inner channel centerline (a) electron temperature (b) plasma potential.

Dual Channel Neutral Simulation

It is hypothesized that the local pressure distribution of neutrals has been affecting the improved performance observed dual-channel operation. The direct simulation Monte Carlo (DSMC) framework Monaco was used to simulate neutral gas flow through the NHT. The benefit of adding simulation insight to the experimental pressure measurements is that individual species of xenon (inner channel, outer channel, etc) can be tracked easily, while for an experimental setup it would be necessary to use different gases in the two channels. Figure 4 shows the computational mesh which was refined based on xenon mean-free path.

Figure 4: DSMC simulation domain and mesh.

Figure 5 shows a comparison of experimental (a) and simulation (b) results. For the experimental results, the thruster was operating with only one channel, and the higher backpressure from the dual channel mode was matched by downstream neutral injection. However, for the simulation, both channels were introducing neutrals at the same time, and there was no background pressure. In both cases, the data is extracted radially at a fixed axial location and for the experimental results arbitrary units are used, but the peaks of the pressure correspond to the channel centerline, which is similar in the simulation results. The pressure values compare favorably between experiment and simulation at the points of maximum pressure, but the differences are larger at different radial locations. The simulation does not include facility effects (background pressure) and particles are free to leave the domain through the outflow boundary, so this may be the cause for the sharp decrease in particles that we observe when moving from the centerline of the channel of interest.



Figure 5: Pressure for xenon particles (a) experiment (b) simulation.

Future Work


Horatiu Dragnea


This work has been supported by a NASA Space Technology Research Fellowship.


  1. Hall, S.J., Jorns, B., Gallimore, A. D., and Hofer, R. R. Expanded Thruster Mass Model Incorporating Nested Hall Thrusters, 53rd AIAA/SAE/ASEE Joint Propulsion Conference, AIAA 2017-4729, Atlanta, GA, July 9-12, 2017.

  2. Liang, R., The Combination of Two Concentric Discharge Channels into a Nested Hall-Effect Thruster, Ph.D. Dissertation, Aerospace Engineering Dept., University of Michigan, Ann Arbor, MI, 2013

  3. Florenz, R.E.,The X3 100-kW Class Nested Channel Hall Thruster: Motivation, Implementation and Initial Performance, Ph.D. Dissertation, Aerospace Engineering Dept., University of Michigan, Ann Arbor, MI, 2014.

  4. Fife, J. M., Hybrid-PIC Modeling and Electrostatic Probe Survey of Hall Thrusters, Ph.D. Dissertation, Dept. of Aeronautics and Astronautics, Massachusetts Institute of Technology, Cambridge, MA, 1998.

Recent Publications