Electrical Facility Effects on Hall Current Thrusters: Electron Termination Pathway Manipulation

Electrical Facility Effects on Hall Current Thrusters:

Electron Termination Pathway Manipulation

Jonathan A. Walker,

∗

Samuel J. Langendorf,

†

and Mitchell L. R. Walker

‡

Georgia Institute of Technology, Atlanta, Georgia 30332

Vadim Khayms

Lockheed Martin Space Systems Company, Sunnyvale, California 94089

and

David King

and Peter Pertso n

Aerojet Rocketdyne, Inc., Redmond, Washington 98052

DOI: 10.2514/1.B35904

A nonnegligible fraction of the charged particles in the Hall current thruster plume completes the electrical circuit

through the conductive wall of a ground-based vacuum test facility. The resultant electrical circuit is different from

the electrical circuit completed by the Hall current thruster in the onorbit environment. To understand the electrical

circuit created in ground-based testing, this work examines the effect of an electrically biased metal plate, placed in the

far-field plume of a Hall current thruster, on the plasma plume characteristics, the Hall current thruster thrust, and

the electron termination pathways. An Aerojet Rocketdyne T-140 Hall current thruster is operated at 300 V and 10.3

A on xenon propellant. The operational neutral background pressure is 7.3 × 10

−6

torr, corrected for xenon. Two

aluminum plates, one representative of the radial wall of the vacuum chamber and one representative of the axial wall

of the vacuum chamber, are placed 2.3 m radially outward from the thruster centerline and 4.3 m axially downstream

from the discharge channel exit plane, respectively. At each axial bias plate voltage, measurements of thrust, electrical

characteristics of the Hall current thruster, thruster body electrical waveform, and radial–axial plate waveforms are

recorded. A Langmuir probe, a Faraday probe, and an emissive probe are placed 1 m downstream of the Hall current

thruster exit plane. The cathode-to-ground voltage and plasma potential behavior closely follow the trends observed

from in-flight measurements of the Small Missions for Advanced Research in Technology-1 PPS-1350 Hall current

thruster. This investigation experimentally quantifies the impact of the varying in-flight plasma plume conditions on

Hall current thruster operation in a ground-based vacuum facility.

I. Introduction

LECTRIC propulsion (EP) flight qualification and lifetime

testing typically occur in electrically conductive ground-based

vacuum chamber test facilities. Electrically conductive test facilities

present a unique challenge to EP devices because the electrically

grounded wall boundary imposed by the vacuum facilities is not

representative of the flight environment. Traditionally, operational

background neutral pressure corrections are used as a means to

translate ground-based Hall current thruster (HCT) behavior to

expected flight operation [1–11]. Recent work on the HCT suggests

that pressure considerations may not entirely capture the effect of the

vacuum chamber environment on HCT operation [12–17].

Flight data from the Russian Express satellites’ north–south and

east–west station-keeping SPT-100 HCTs revealed an expanded in-

space plume that was not measured in ground testing [15]. The

expanded plume of the SPT-100 had high-energy ions in the offaxis

plume of the HCT and created anomalous disturbance torques due to

impingement of these high-energy ions on satellite surfaces [15].

This behavior has been attributed, but not yet demonstrated, to be the

result of facility enhanced charge–exchange collisions removing

energetic offaxis ions from the population [15,18]. These high-

energy ions were not measured in ground testing due to the increased

presence of charge–exchange ions in the vacuum facility environ-

ment. Anomalous thruster behavior was also experienced during the

European Space Agency’s Small Missions for Advanced Research in

Technology-1 (SMART-1) mission using the PPS-1350 HCT. The

PPS-1350 HCT showed periods of positive cathode-to-satellite-bus

common voltage during times of elevated in-space plasma potential

measurements [12,13,19]. The causes of these elevated in-space

plasma potential measurements were attributed to exposure of low-

voltage solar panel contacts to the HCT plume [12]. To the best of the

authors’ knowledge, this behavior has not been replicated in a

ground-based testing environment.

Previous work by Frieman et al. showed that HCT cathode-to-

ground voltage was susceptible to the influence of a downstream bias

plate [20] and concluded that the conductive vacuum chamber was

indeed part of the HCT electrical circuit through the mediation of the

electron-ion loss rate of the plume. However, plume and performance

measurements were not taken. Further investigation by Frieman et al.

[17] identified three electron termination circuits: downstream axial

surfaces, downstream radial surfaces, and the thruster body. By

increasing the cathode radial position away from HCT centerline, the

magnitude of the electron current collected on the radial chamber

plate surfaces increased, indicating increased electron termination on

the radial walls of the vacuum facility. Figure 1 shows a graphical

representation of the electron termination pathways. It is important to

note that the resistors shown in Fig. 1 are done to help illustrate the

electron termination pathways in the facility testing, and the plasma

Received 2 June 2015; revision received 29 February 2016; accepted for

by Jonathan A. Walker, Samuel J. Langendorf, and Mitchell L. R. Walker.

Published by the American Institute of Aeronautics and Astronautics, Inc.,

with permission. Copies of this paper may be made for personal and internal

use, on condition that the copier pay the per-copy fee to the Copyright

Clearance Center (CCC). All requests for copying and permission to reprint

should be submitted to CCC at www.copyright.com; employ the ISSN 0748-

4658 (print) or 1533-3876 (online) to initiate your request.

*Graduate Research Assistant, Aerospace Engineering, High-Power

Electric Propulsion Laboratory; [email protected]. Student Member

AIAA.

†

Graduate Research Assistant, Aerospace Engineering, High-Power

Electric Propulsion Laboratory; [email protected]. Student

Member AIAA.

‡

Associate Professor, Aerospace Engineering, High-Power Electric

Propulsion Laboratory; [email protected]. Associate Fellow

AIAA.

Electric Propulsion Architect; [email protected]. Member

AIAA.

Technical Fellow, Engineering; [email protected]. Member AIAA.

**Project Engineering Specialist, Engineering; peter.peterson@rocket.

com. Senior Member AIAA.

1365

JOURNAL OF PROPULSION AND POWER

Vol. 32, No. 6, November-December 2016

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

of the HCT makes these pathways much more dynamic than a simple

resistive path. Plume measurements show an increase in plasma

potential with increased electron termination on radial walls. Their

findings indicated that electron magnetization near the cathode

orifice played a significant role in what the available electron

termination pathways are, but they were not able to measure a time-

averaged influence of the electron termination pathways on the thrust

of the HCT.

From the prior work, the exact impact of the artificial boundary

condition imposed by the facility walls on the HCT electrical circuit is

unclear. In the flight environment, the HCT plume does not typically

encounter these artificial termination sites [18]. Work done by

Frieman et al. showed that the HCT floating circuit voltage relative to

ground changed synchronously with changes in the downstream

potential of the axial chamber plate [20]. The work did not measure

plume characteristics or thrust. However, they concluded that the

plasma potential changed with the bias voltage of the downstream

plate, and the thrust generated by the HCT would not be affected. The

changing plasma potential would also affect electron current

collection at the radial walls of the vacuum chamber. The changing

plasma potential might have an impact on the discharge current

oscillations of the HCT, as work from Walker et al. [21] showed that

the discharge current oscillation frequency and full-width/half-

maximum of the discharge current oscillation frequency changed

along with changes in the electron flux to the radial chamber plate.

Time-resolved measurements of the discharge current by Walker

et al. [21] revealed that the discharge current oscillation frequency

and full-width/half-maximum depended on cathode position.

Their results also indicated that the electron Hall parameter near

the cathode orifice played an important role in determining the

frequency, peak to peak, and full-width/half maximum of the

discharge current oscillations. These changes in the discharge current

were also measured on the time-resolved current waveforms of the

chamber witness plates, and they demonstrated a strong statistical

correlation between the HCT discharge current and the chamber

plates that were independent of the magnitude of the electron Hall

parameter at the cathode orifice.

With previous cathode positioning work done by Frieman et al.

[17] and Walker et al. [21], it is difficult to separate the effect of the

cathode position and the electron termination pathways on the HCT.

The methodology of Frieman et al. [20] by electrically biasing the

downstream chamber plate presented a possible way to precisely

control the influence of the electron termination pathways on the

HCT electrical circuit. The biased downstream chamber plate had

already shown that it could influence the floating voltage of the HCT,

which was similar to the behavior seen during the SMART-1 mission.

The lack of precise knowledge of plume plasma potential prevented

direct comparison between the prior work of Frieman et al. [20]

and the results obtained from the SMART-1 mission [13]. To better

understand the influence of the electron termination pathways on

the HCT electrical circuit and its connection to the in-flight

environment, it is the goal of this investigation to better understand

the impact of electrically biasing the downstream axial plate on the

HCT performance, plume characteristics, and electron termination

pathways.

II. Experimental Apparatus

A. Vacuum Facility

All experiments were performed in Vacuum T est Facility 2 (VTF-2)

at the Georgia Institute of Technology’s High-Power Electric

Propulsion Laboratory. Figure 2 shows a schematic of this facility.

VTF-2 is a stainless-steel chamber measuring 9.2 m in length and 4.9 m

in diameter. VTF-2 was ev acuated to a rough vacuum using one 495

cubic feet per min ute (CFM) rotary-v ane pump and one 3800 CFM

blower. High vacuum was achieved using 10 liquid-nitroge n-cooled

CVI TM1200i reentrant cryopumps . The cryopump shroud s were fed

using the Stirling Cryogenics SPC-8 RL special closed-loop nitrogen

liquefaction system, detailed by Kieckhafe r and Walker [22]. The

facility had a combined nominal pumping speed 350;000 l∕s on xenon

and could achiev e a base pressure of 1.9 × 10

−9

torr.PressureinVTF-2

was monitored using two Agilent Bayard Alpert (B A) 571 hot filament

ionizationgaugescontrolledbyanAgilent XGS-600 gauge controller .

The pressure measurement uncertainty of the Agilent BA 571 was

expected to be 20%,whichwas−10% of the indicated pressure [23].

One gauge was mounted to a flange on the exterior of the chamber,

whereas the other was mounted 0.6 m radially outward from the thruster

centered on the exit plane. To prevent plume ions from having a direct

line of sight to the ionization gauge filament of the interior ion gauge,

and potentially affect the pressure measurement, a gridded neutralizer

assembly identical to the one used by Walker and Gallimore [1] was

attached to the gauge orifice. The nominal operating pressure for this

work as measured by the interior and exterior ion gauges was 1.3 ×

−5

torr and 7.3 × 10

−6

torr, corrected for xenon, respectiv ely. As

specified by the manufacturer, the corrected pressure P

was found by

relating the indicated pressure P

and the vacuum chamber base

pressure P

to a gas-specific constant using the follo wing equation [24]:



− P

2.87

 P

(1)

B. T-140 Hall Current Thruster

All experiments detailed in this work were performed using the

Aerojet Rocketdyne T-140 HCT originally developed by Space

Power, Inc., in collaboration with the Keldysh Research Center and

Matra Marconi Space [25]. The T-140 HCT is a laboratory-model

HCT that has a discharge channel made of M26-grade boron nitride

with an outer channel diameter of 143 mm. The performance of

the T-140 has been extensively mapped by prior investigations [25].

The thruster body was isolated from the facility ground such that the

thruster body could be electrically configured as either floating or

grounded. The resistance to ground when the thruster body was

electrically grounded was measured to be less than 0.2Ω.

High-purity (99.9995%) xenon propellant was supplied to the

thruster and cathode using stainless-steel lines metered with MKS

Instruments 1179A mass flow controllers. The controllers were

Fig. 1 Notional diagram of discharge circuit of Hall current thruster

and electron termination pathways in a ground-testing environment.

Fig. 2 Overhead view of the vacuum chamber test facility, HCT, and

chamber plates.

1366

WALKER ET AL.

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

calibrated before each test by measuring gas pressure and

temperature as a function of time in a known control volume. After

calibration, the mass flow controllers had an uncertainty of

0.03 mg∕s (5.1% referenced at the tested flow rate) for the cathode

flow and 0.12 mg∕s (2% reference at the tested flow rate) for the

anode flow [26].

An Electric Propulsion Laboratory hollow cathode plasma

electron emitter 500-series cathode was located at the nine o’clock

position of the HCT. The cathode flow rate was set to a constant

1.16 mg∕s for all HCT operating conditions. The orifice location of

the cathode was located approximately 2.5 cm downstream of the

thruster exit plane at a fixed declination of 55 deg with respect to the

thruster centerline. The nominal radial position of the cathode was

18.1 cm outward from HCT centerline. Time-resolved measurements

of the discharge current, radial chamber plate current and voltage, and

axial chamber plate current and voltage were taken at each axial bias

plate voltage.

The T-140 HCT discharge was controlled using a Magna-Power

TSA800-54 power supply. The thruster inner and outer magnet coils

were powered with TDK-Lambda GEN60-25 power supplies. A

TDK-Lambda Genesys 150 V∕10 A and a TDK-Lambda Genesys

40 V∕38 A power supply were used to power the cathode keeper and

heater, respectively. The thruster discharge supply was connected to a

discharge filter consisting of a 95 μF capacitor and 1.3Ω resistor in

order to prevent oscillations over 1.4 kHz in the discharge current

from reaching the discharge supply. Diagnostic and power

connections entered VTF-2 through separate feedthroughs to

eliminate potential crosstalk between the thruster discharge power

lines and diagnostic lines. Figure 3 shows the circuit used to operate

the T-140 HCT and current and voltage measurement points in

this work.

The discharge current oscillations, measured on the thruster side of

the discharge filter, of the T-140 HCT were recorded using a Teledyne

LeCroy CP150 current probe connected to a Teledyne LeCroy

HDO6104 oscilloscope. The uncertainty and bandwidth of the

current probe were 1% and 10 MHz; for the oscilloscope, they were

0.5% full scale and 1 GHz. In the floating thruster body

configuration, the thruster body floating voltage was measured

differentially using Teledyne LeCroy PP018 passive probes with a

bandwidth of 500 MHz and an accuracy of 0.5% connected to the

Teledyne LeCroy oscilloscope. When the thruster body was

grounded, the current conducted through the thruster body to ground

was measured using a Teledyne LeCroy CP030 current probe

connected to the Teledyne LeCroy oscilloscope. The CP030 had a

bandwidth of 50 MHz and an accuracy of 1%. A filter sensitivity

analysis of the discharge current filter operating in tandem with the

discharge supply and hall thruster, as described by Spektor et al. [27],

was not performed.

The mean discharge voltage of the T-140 HCT was measured

differentially using a pair of Teledyne LeCroy PPE2KV 100∶1 high-

voltage probes connected to a Teledyne LeCroy HDO6104

Oscilloscope. The bandwidth of the voltage probes was 400 MHz;

the oscilloscope had an uncertainty and bandwidth of 2% and

300 MHz, respectively. This was done to ensure that the HCTelectrical

circuit remained floating relative to the ground. Figure 3 shows the

location of each telemetry measurement in the T-140 HCT circuit.

C. Thrust Stand

Thrust was measured using the null-type inverted pendulum thrust

stand of the NASA John H. Glenn Research Center design, detailed in

the work of Xu and Walker [28]. The thrust stand consisted of a pair of

parallel plates connected by a series of four flexures that supported

the top plate and permitted it to deflect in response to an applied force.

The position of the upper plate was measured using a linear voltage

differential transformer (LVDT) and was controlled using two

electromagnetic actuators. During operation, the current through

each actuator was controlled using a pair of proportional-integral-

derivative control loops that used the LVDT signal as the input and

then modulated the current through the actuators in order to remove

any vibrational noise (damper coil) and hold the upper plate

stationary (null coil). The thrust was correlated to the resulting

current through the null coil that was required to keep the upper plate

stationary. To maintain thermal equilibrium during thruster firings,

the thrust stand was actively cooled using three parallel loops: one

each through the structure, the null coil, and the outer radiation

shroud. Cooling water was supplied by a 1100 W VWR International

1173-P refrigerated recirculation chiller and did not vary by more

than 5°C as compared to the thruster-off condition [28]. The thrust

stand was calibrated by loading and offloading a set of known

weights that spanned the full range of expected thrust values. A linear

fit was then created in order to correlate the null coil current to the

force applied to the thrust stand. To minimize the thermal drift of the

zero position, the T-140 HCT was initially fired for 3 h at the 3.1 kW

nominal operating point to permit initial heating of the system to

near-thermal equilibrium [25] before the first calibration and was

then shut down every 40–60 min so that a recalibration could be

performed. The thrust stand uncertainty for this work was 3mN

(1.7% full scale). All data were collected with the T-140 HCT

operating at a discharge voltage of 300 V, a discharge power of

3.16 kW, an anode xenon flow rate of 11.6 mg∕s, and a cathode

xenon flow rate of 1.16 mg∕s. The thruster discharge voltage, inner

and outer magnet currents, anode mass flow rate, and cathode mass

flow rate were held constant for all test configurations.

D. Configuration of Plates

To assess the impact of the conductive walls of the vacuum chamber

facility on HCT operation, two 0.91 m × 0.91 m × 0.16 − cm-thick

square aluminum plates served as representative chamber surfaces.

Each plate was mounted adjacent to, but electrically isolated from, the

walls of the vacuum test facility. One plate was placed 4.3 m

downstream from the exit plane of the thruster, and it is referred to as

the “axial chamber plate” or “axial plate.” The other plate was located

2.3 m radially outward from the thruster centerline and centered on the

exit plane of the T-140 HCT, and it is referred to as the “radial chamber

plate” or “radial plate.” These two locations were chosen because each

location was representative of unique plasma environments: inside the

HCT beam and outside the HCT beam 95% half-angle [29]. The axial

plate was in a quasi-neutral plasma environment composed primarily

of accelerated HCT ions and electrons, and the radial plate was in a

quasi-neutral plasma environment primarily composed of charge–

exchange ions and electrons. Figure 2 shows the physical location of

the plates with respect to the T-140 HCT. Identical plates have been

used in previous studies of electrical facility effects [17,20,21]. The

radial chamber plate was electrically grounded using RG-58 coaxial

cable with a grounded shield that passed through a Bayonet Neill-

Concelman feedthrough into the chamber. The resistance between the

Fig. 3 Electrical diagram of current and voltage measurements of the

HCT discharge circuit.

WALKER ET AL. 1367

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

radial chamber plate and the chamber walls was measured to be 1.1Ω.

Based on current measurements made by Frieman et al. [17,20], the

current capacity of the inner conductor of the RG-58 was sufficient for

radial plate current collection and would not pose any thermal issues

during thruster testing. For radial plate current measurements, ground

loops were not a concern because the current probes were active clamp

current monitors. The axial chamber plate was biased relative to the

ground using a TDK Gen 60 V and 12.5 A power supply. To avoid

thermal issues with maintaining the axial plate bias voltage, the axial

chamber plate was connected to the power supply via 6-AWG copper

wire that connected to a 150 A power vacuum feedthrough. During this

investigation, the axial plate current did not exceed 10 A for testing

conditions. The resistance between the axial chamber plate and the

chamberwallswasmeasuredtobelessthan0.2Ω.

The current and voltage waveforms of the axial chamber plate were

measured using a LeCroy CP030 current clamp and a PP005A 10∶1

voltage divider connected to a Teledyne LeCroy HDO6104

oscilloscope. The radial chamber plate was connected to the chamber

ground with the current conducted to the ground measured with a

Teledyne LeCroy CP030 current sensor connected to a Teledyne

LeCroy HDO6104 oscilloscope; the plate currents and thruster

telemetry waveforms were measured simultaneously at a sampling

frequency of 125 MS∕s (Mega-samples per second) for a 20 ms

window to ensure that multiple fundamental discharge current mode

periods were captured.

E. Plume Diagnostics

Plume diagnostics were taken along a 1 m 0.01 m radius centered

at the thruster centerline and discharge plane. Ion current density

measurements occurred throughout a full range of 180 deg, whereas

emissive probes and Langmuir probes sweeps were taken at select

angular positions based on the ion current density profile of the HCT

plume. A schematic overvie w of the plume diagnostics relative to the

HCT is shown in Fig. 4. The probe diagnostics were mounted to a

Parker Daedel RT-series 8 in. rotary motion table. All three of the

plume diagnostics (a Langmuir probe, a Jet Propulsion Laboratory

(JPL) nude-style Faraday probe, and an emissive probe) were attached

in an array on a radial probe arm. The arms of the array were angled

such that the probe-to-probe centerline linear distance was at 0.17 

0.01 m apart and remained at a 1 m radial distance throughout the

probe arm sweep. Figure 4 shows the relative position of the

diagnostics arm, and Figs. 5 and 6 show a photograph and notional

diagram, respectively, of the probe arrangement on the diagnostics arm.

1. Emissive Probe

The probe tip used for this work was constructed from a loop of

0.13 mm thoriated-tungsten wire housed in a 4.8 mm double-bore

alumina tube. Emissive probe sweeps were performed at select

thruster-to-centerline angles at a radius of 1 m from the thruster

centerline and discharge exit plane. The inflection point method was

used for data collection. In this method, the probe was heated and then

the emission current was monitored as the probe bias voltagewasswept

in a manner similar to that used with Langmuir probes. The changing

characteristic of the emission current trace as a function of applied bias

voltage was then used to determine the plasma potential [30]. During

each measurement, the heating current to the emissive probe fila-

ment was held at five different heating current values to change the

electron emission of the probes. These heating current values varied

throughout the probe lifetime but were within a range between 1.2 and

1.8 A. One bias voltage sweep was taken per emissive probe filament

heating current. During each bias voltage sweep, the probe voltage was

varied over a range of 0 to 100 Vin 1 Vincrements, with a 300 ms dwell

time. The inflection point was then found in each of the Current-

Voltage (I-V) traces for each of the different heating current levels, and

the plasma potential was found by linearly extrapolating these values to

zero emission [30]. The uncertainty associated with this method was

approximately 0.5 V [30]. The heating current was controlled using

a Xantrex XPD 60-9 power supply. The probe bias was controlled by a

Keithley 2410 1100 V source meter, and the emission current was

measured using a Keithley 6487 picoammeter. The source meter and

picoammeter were simultaneously controlled using a LabView virtual

instrument to ensure synchronous recording of the probe bias voltage

and emitted current.

2. Langmuir Probe

The ion and electron number densities were measured using a

cylindrical Langmuir probe. Langmuir probe sweeps were performed

at select thruster-to-centerline angles at a radius of 1 m from the

thruster centerline and discharge exit plane. The probe used in this

work was constructed using a 0.13-mm-diam, 18.3-mm-long tungsten

tip housed inside an alumina tube. The probe was bent at a right angle,

such the probe tip was not pointed at the HCT and was pointed 90 deg

out of the plane of the probe arm sweep. The bend in the probe was

done to minimize the effect of the probe tip on the I-V trace of the

Fig. 4 Overhead view of the vacuum chamber test facility, HCT, and

plume diagnostics.

Fig. 5 Probe arm with plume diagnostics attached Langmuir probe (A), Faraday probe (B), and emissive probe (C).

1368

WALKER ET AL.

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

Langmuir probe. A Keithley 2410 1100 V source meter was used to

control the probe tip bias and measure the collected current. During

each current-voltage sweep, the tip voltage was varied over a range of

−50 to 100 V in 0.2 V increments, with a 300 ms dwell time. For low

axial plate bias voltage, ion saturation occurred near 12 V; and at higher

axial plate bias voltage, ion saturation occurred near the potential of the

downstream axial chamber plate. Two sweeps were taken per measure-

ment and were averaged together before processing. The results were

interpreted using an orbital-motion-limited theory, with an expected

uncertainty in ion and electron density measurements of

40% [31,32].

3. Faraday Probe

The ion current density was measured using a nude-type JPL Faraday

probe [33]. Faraday probe measurements occurred continuously along a

180 deg, 1m 0.01 m arc. Angular resolution during the sweep was

limited by the multiplexor hardware-required settling time given a

measurement voltage range. Angular resolution ranged from 0.34 deg in

the “wings” of the plume to 0.2 deg in the center of the HCT plume. The

diameter of the collector was 2.31 cm. The probe had a guard-ring

diameter of 2.54 cm with a 0.036 cm gap between the collector and the

guard ring. The guard ring and collector were biased to −30 V belo w

ground for all axial plate bias voltages. T o measure the current flowing to

the collector , the voltage drop across a 100Ω  12% precision resistor

was measu red using an Agilent 34980A Mainframe with an Agilent

34922 A armature multiplexor . Rotary table encoder information and

voltage drop measurements were taken using a LabView virtual instru-

ment to ensure the synchronous recording of the angular position and

voltage drop across the shunt resistor. Data reduction and correc tion

factors used to calculate the ion current density from the Faraday probe

data were performed according to the work of Bro wn [34] and Brown

and Gallimore [35].

III. Results

This section introduces the overall trends of each of the diagnostics

before discussing the impact on biasing of the axial plate on the

electrical facility effects. The following section covers the impact of

the axial chamber plate on the HCT cathode-to-ground voltage (HCT

circuit floating potential), the HCT plume, and the radial plate current

collection.

A. Axial Chamber Plate Behavior

As shown in Fig. 7, the axial chamber plate current collection

behavior exhibited three distinct regions. Because the plate was

biased with a positive voltage relative to the ground, the current

collected on bias plate transitioned from a net flux of ions to a net flux

of electrons. With the axial chamber plate bias at 0 V, the beam ions

generated by the HCT composed the majority of the net charge flux to

the plate. As the bias voltage increased, more electrons were gathered

to the axial chamber plate. Between 0 and 5 V, sufficient electrons

were collected by the plate such that there was no net charge flux to

the plate. This voltage was also known as the floating voltage of the

axial plate and was not precisely measured during this study. Based

on the measurements of Frieman et al. [17], the floating voltage of the

axial plate was expected to be approximately 4 V. At the axial

chamber plate bias beyond the floating voltage, there was a net

electron current collection reaching the axial chamber plate. This

increase in electron current continued monotonically until the net

electron current collection approached the beam current of the HCT.

Based on the acceleration voltage and the thrust measured by Frieman

et al. [17], the beam current of the HCT can be estimated as follows

[29]:



HCT



2MV

(2)

where I

is the beam current, T

HCT

is the thrust produced by the HCT,

M is the mass of a xenon atom, and V

is the acceleration potential.

Fig. 6 Overhead view of notional layout of the radial diagnostics probe arm.

Fig. 7 Axial chamber plate current collected as a function of axial plate

bias voltage. Error bars are encompassed by plot markers.

WALKER ET AL. 1369

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

Using this model, the beam current is estimated to be near 7 A. This

relationship neglects the presence of multiply charged ions and does

not take into account the beam divergence. Based on E × B measure-

ments by Ekholm and Hargus [36] running a BHT−200 HCT at

250 V, a lower range to the beam current estimate can be made, and it

is expected that the T-140 HCT operated at 300 V will have a doubly

charged xenon population that is approximately 12% of the total ion

population. This reduces the estimated beam current to be on the

order of 6 A. A better estimate of the beam current is not possible

using the Faraday probe because testing occurs at one operating

background pressure. Without current density profile measurements

at multiple background pressures, it is not possible to extrapolate the

vacuum current density profile, thus making the beam current

measurement via an integrated current profile artificially inflated.

For an axial plate bias greater than 20 V, the slope of the electron

current collection vs axial plate bias voltage decreases by approxi-

mately 85%. Data collection above the axial chamber plate bias of

50 V above ground is not possible due to arcing events on grounded

surfaces within the vacuum chamber. After raising the potential of the

axial chamber plate past the floating voltage of the chamber plate, the

axial chamber plate begins collecting a net flux of beam electrons

from the surrounding plasma. As the plasma sheath begins to expand

to collect more electrons, the plasma potential begins to increase.

Once the axial chamber plate begins to collect a net electron current

equal to the expected beam current, the electron current collection as

a function of the axial bias plate voltage above ground begins to level

off, forming a knee in the curve.

The floating thruster body and the grounded thruster body

configurations have similar overall current collection behaviors. At

thruster body biases between 0 and 15 V, however, the floating

thruster body configuration has an axial chamber plate current

collection 49% and 15% smaller current than comparable grounded

thruster body configurations. At greater than 15 V of the axial plate

bias voltage above ground, the floating thruster body configuration

has an axial chamber plate current collection 1–2% greater than

comparable grounded thruster body configurations. The reason for

the discrepancy between the thruster body configurations is not

yet clear.

B. Radial Plate Facility Interaction

As seen in prior work, the grounded radial chamber plate collects a

net flux of electron current [17]. As the bias voltage of the

downstream axial chamber plate increases, the electron current

collected on the radial plate decreases. Figure 8 shows the radial

chamber plate current collection as a function of axial chamber plate

bias voltage. The negative current collected corresponds to the net

electron current, and the positive current collected indicates the net

ion current.

Between 10 and 15 Vof the axial chamber plate bias potential, the

current collection on the radial chamber plate transitions from a net

flux of electron current to a net flux of ion current. At axial bias

chamber plate biases greater than 15 V, the plasma potential near the

radial chamber plate rises sufficiently relative to the chamber walls

that the potential difference between the plasma and the grounded

radial chamber plate repels electrons. The current collection behavior

of the radial chamber plate as shown in Fig. 9 indicates that, for axial

bias plate voltages greater than 25 V, electrons are driven away from

the radial chamber plate, and the radial chamber only collects a net

ion current.

C. Influence of the Axial Chamber Plate Bias on the HCT

The effect of the downstream bias voltage of the axial chamber

plate on the HCT was measured in two ways: thrust and characteris-

tics of the HCT electrical circuit. Measurements of the thrust showed

no statistically significant change in the thrust production of the HCT

circuit. Based on the conclusions drawn by Frieman et al. [20], this

was expected. In Fig. 10, the cathode-to-ground voltage and center-

line plasma potential measurements as a function of axial plate bias

voltage above ground are shown. As the axial plate bias voltage

increased, the cathode-to-ground voltage began to move syn-

chronously with the plasma potential. The cathode-to-ground voltage

relative to ground changed sign between 20 and 25 V of the axial

chamber plate bias. This axial plate bias voltage range corresponded

to the collected electron current on the axial chamber plate above

the HCTestimated beam current. From the axial chamber plate bias of

5 to 50 V, the difference between the centerline plasma potential and

the cathode-to-ground voltage remained a constant 32 V  2V.

Because the difference between the cathode-to-ground voltage and

the plasma potential remained nearly constant, there was no expected

measurable change in thrust of the HCT with the axial chamber plate

bias voltage. This was confirmed with direct thrust measurements.

Thrust stand measurements showed that the time-averaged thrust

of the HCT remained 177 mN  3mN for all axial bias plate

conditions.

D. Plume Plasma Properties

Figure 11 shows the HCT plume profile. The HCT beam and

exponential decline region of the current density profile show little to

no dependence on the axial chamber bias plate bias. At angular

positions greater than 50 deg off the thruster centerline, elevated

current densities at a high (greater than 20 V) axial plate bias relative

to the ground are measured. In this region, colloquially referred to as

Fig. 8 Radial chamber plate current collection as a function of axial

chamber plate bias voltage relative to ground. Error bars are

encompassed by plot markers.

Fig. 9 Plasma potential measured at 25.4 cm radial distance away from

the radial plate. Measurement is centered on radial plate centerline.

1370

WALKER ET AL.

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

the wings, the ions present are primarily composed of charge–

exchange ions [29,34]. This region is highlighted in Fig. 11. The

current density profiles of the grounded thruster body and floating

thruster body show no significant differences, and so Faraday probe

sweeps for the floating thruster body are not shown.

At least two Langmuir probe and emissive probe measurements

were sweeps taken in each ion current density profile region: thruster

centerline, exponential decline, and the wings. Assuming the plasma

properties to be axisymmetric, Langmuir and emissive probe

measurements were taken throughout an arc of 90 deg relative to the

thruster centerline. As shown in Fig. 12, the plasma potential

measurements showed a global increase in potential with respect to

axial plate bias voltage. The plasma density profile showed no change

outside the uncertainty of the measurement with respect to the axial

plate bias voltage, as shown in Fig. 13. For both the plasma potential

and ion number density, the floating thruster body configuration HCT

plume profile did not a show a measureable difference as compared to

the grounded thruster body, and is not presented.

IV. Discussion

With the information presented in the Results section (Sec. III), it is

important to begin to understand how the axial plate is able to drive

global changes in the HCT testing environment. The discussion

begins with examining the measured influence of the axial chamber

plate bias voltage on the plasma potential and whether this impact

agrees with first-order plume models. Once the impact of the axial

plate bias is established, the discussion then proceeds to address the

impact of the axial chamber plate on the behavior of the electron

termination pathways. The discussion concludes with how the

observed thruster behavior is reflective of measurements taken

onorbit with the PPS-1350 HCT on the SMART-1 mission.

A. Plasma Potential and Plate Current

To better understand the interaction between the axial plate and the

HCTelectrical circuit, it is first important to understand the interaction

between the axial plate and the thruster plume environment. Between

Fig. 10 Cathode-to-ground voltage and centerline plasma potential as a function of axial plate bias. Error bars for cathode-to-ground voltage are

encompassed by plot markers (GND, electrically grounded; 2GND, -to ground; FLT, electrically floating).

Fig. 11 Grounded thruster body HCT ion current density profile at differing axial plate bias voltages: a) full plume profile, and b) enlarged section of the

plume profile.

WALKER ET AL. 1371

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

the axial plate and the plume environment, a plasma sheath mediates

the current collection. It is then critical to understand how this sheath

responds to changes in the axial plate bias voltage.

Examining the data presented in Fig. 7, the current collected by the

axial plate rises with bias voltage until it reaches the thruster beam

current that occurs at 20 V abov e ground. At this potential, the plate

collects an electron current equiv alent to the neutralization current

supplied by the cathode. Above 20 V, increases in the axial plate bias

voltage result in an equal increase of the plasma potential and the floating

potential of the thruster anode and cathode (Fig. 10). At these voltages, it

is possible to increase the cathode potential above ground. The collected

current increases slowly, as all the cathode neutralization electrons are

already being collected and additional electrons must be sourced from

grounded chamber surfaces. The physical connection between the two

phenomena (knee in current collection and start of plasma potential rise)

is the plasma charge balance. If the plate is biased positive, the plasma

will electrostatically respond and the plasma potential will adjust to

equalize charge loss rates and keep the plasma electrically neutral.

To illustrate the effect of the plasma charge balance and to

determine if there are other sources for the additional electron current

collected on the axial plate outside of the HCT beam, we model the

current collection by the plate and chamber boundary to compare to

the experiment. To know the current collection, we need the local

plasma parameters at the boundary. To this end, we apply the self-

similar plume model of Korsun and Tverdokhlebova [37] as reported

by Azziz [38]. This model neglects collisional effects in the chamber,

and it assumes a two-component plasma, and adiabatic expansion of

the HCT plume. The model gives the following relations to calculate

the ion flux j

, electron density n

, electron temperature T

, and

plasma potential ϕ at any location in the plume:



2πR

tan

1∕2

(3)



cos

θ1 tan

θ∕tan

1∕2



1γ ∕2

(4)



(5)

Fig. 12 Plasma potential profile as a function of axial plate bias voltage for the grounded thruster body.

Fig. 13 Ion number density profile as a function of axial plate bias voltage.

1372

WALKER ET AL.

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904



1 tan

θ∕tan

1∕2



(6)





γ−1

(7)

ϕ  ϕ

−

γ − 1



1 −





γ−1



(8)

In Eqs. (3–8), R and θ are polar coordinates with the origin at the

center of the thruster exit plane, θ

1∕2

is the thruster beam divergence, γ

is the plasma polytropic index that we set to 1.3, and subscript c refers

to the centerline or reference value. We assume that the model form of

the plasma potential is always true, no matter the bias voltage of the

axial plate, and thus the plasma potential in the chamber has a fixed

spatial distribution. This can be considered true to the first order

because the expanding plume structure described by the model is set

up by the operation of the thruster, and the measured plasma potential

profile (Fig. 12) has approximately the same plasma potential spatial

distribution relative to other positions.

The chamber wall and axial plate are paneled as a series of rings of

0.1 m width. At each boundary panel, the ion flux, electron density,

electron temperature, and plasma potential are calculated from

Eqs. (3–8). We then calculate the ion current to the panel from the ion

flux assuming singly charged ions [Eq. (9)] and the electron current to

the panel [Eq. (10)]:

 j

eA (9)





πm

exp



eϕ

−ϕ



< ϕ



πm

≥ ϕ

(10)

We then solve numerically for the value of plasma potential that

equalizes the total ion and electron currents lost from the plasma to all

panels. Figure 14 shows the result for the plasma potential 1 m from

the thruster compared to the experimental data. Figure 15 shows the

collected current on the axial plate compared to the experimen-

tal data.

Figure 14 shows that the plasma potential behavior in the model

agrees well with experimental data and shows a 3–4 Voffset between

the model and the experimental data. The remaining offset between the

model and the experiment may result from collisional effects changing

the plasma scaling in the far-field region, where the charge–exchange

background plasma becomes significant with respect to the expanding-

plume plasma. It may also be due to the simplified assumed geometry,

which neglects ion and electron fluxes to detailed chamber features

such as the central I beam and personnel support grating. Figure 15

shows that the collected current agrees qualitatively, but all of the

cathode electrons are theoretically collected at a much lower bias

voltage than is observed experimentally. This is most likely because the

model does not take into account collisional effects and the charge–

exchange background plasma. In the model, the plasma density is very

low at chamber wall surfaces thatare not directly impinged by the beam

where, in reality, the plasma is denser at the walls due to charge–

exchange collisions and the associated diffusion of the plasma. This in

turn means that the plate does not collect all the electrons until a higher

bias voltage.

It is important to note that secondary electron emission effects on

the aluminum chamber plate collected current are neglected in the

model for the following reasons: Electron energy distribution

measurements of the secondary electron emission (SEE) of

aluminum (from Baglin et al. [39], Pillon et al. [40], and Yamauchi

and Shimizu [41]) show that the energy distribution of secondary

electrons is to the first-order invariant of incoming energy of electrons

or ions, and the maximum energy of these electrons is on the order of

15 eV with a most probable energy on the order of 3–4 eV. This means

that electrons produced via SEE from the aluminum plate do not have

enough energy to overcome the potential difference between the

biased axial chamber plate and the surrounding plasma, and they are

recollected by the axial chamber plate. The overall first-order net

effect is that electrons produced by the SEE from the aluminum

chamber plate do not influence the current collection measured. The

collected current in the model does not increase above the cathode

electron current because additional electrons gained from other

chamber sources are not included in the model. At the axial chamber

plate biases greater than 20 V, there is an experimentally measured

current collection that is well outside what is known to be generated

by the HCT beam; therefore, it is concluded that an electrical circuit is

formed between the axial chamber plate and the grounded vacuum

chamber surfaces through the plasma.

B. Electrical Facility Interaction

The bias voltage of the axial chamber plate is able to control the

electron termination pathways of the HCT plume. When the axial

plate is grounded, electrons sourced from the cathode are driven

Fig. 14 Plasma potential as a function of axial plate bias voltage: model

versus experiment.

Fig. 15 Axial plate current collected as a function of applied bias

voltage: model versus experiment.

WALKER ET AL. 1373

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

electrostatically into the plume and are collected onto grounded

chamber surfaces. This includes the electron current collected onto a

grounded thruster body. As the axial chamber plate electron current

collection surpasses the available beam current of the HCT, the

cathode potential (relative to ground) floats above ground due to the

increase in the global plasma potential, as shown in Figs. 10, 12, and

14. Due to the adverse potential gradient between the grounded

chamber walls and the cathode-to-ground voltage, electron

termination on grounded chamber surfaces diminishes. Figure 16

shows that the decrease in collected electron current on grounded

surfaces is also seen in the collected electron current on the grounded

thruster body. The floating potential of the electrically floating

thruster body also begins to shift positive to attract additional electron

flux to maintain a zero net current condition.

At axial plate bias voltages above 20 V, 100% of the HCT beam

current is collected on the axial chamber plate, but electron current

collection continues to increase with increases in axial chamber plate

bias voltage. As demonstrated by the first-order analysis of the

current collection on the axial plate (shown in Fig. 15), grounded

chamber surfaces are possible sources for these electrons due to field

emission or secondary electron emission from chamber surfaces to

the plume plasma, because increasing the axial chamber plate bias

voltage increases the potential gradient between the plasma and the

grounded chamber. As the potential gradient between the chamber

wall and the plasma increases, at bias voltages above 40 V, arcing

events are witnessed on grounded chamber surfaces. These arcing

events indicate a momentary discharge between grounded surfaces

and the ambient plasma. At axial plate bias voltages greater than 40 V,

the potential gradient between the plasma potential and grounded

chamber surfaces drives all electrons away from grounded surfaces.

This potential difference removes the vacuum chamber as an effective

electron termination pathway and allows grounded chamber surfaces

to become a source of electrons. Figure 17 is a graphical

representation of the electron termination pathways for the

aforementioned three axial chamber plate bias voltages.

C. Impact of the Axial Chamber Plate Electrical Power

As seen from radial chamber plate current measurements (Fig. 8)

and plasma potential measurements near the radial plate (Fig. 9),

electrons are driven away from grounded surfaces due to the

increased potential difference between the facility walls. The

decrease in electron current to grounded surfaces is more indicative of

a spacelike environment [18,42]. According to Korsun et al. [18],

testing in a ground-facility environment produces a secondary

plasma that interacts with the facility walls, and currents “leak” out of

the HCT plume into this secondary plasma. These currents represent

a loss of energy from the HCT plume into the vacuum chamber walls.

With the axial chamber plate, the forced collection of electrons

provides additional energy into the plume. By multiplying the current

and the voltage of the axial chamber plate, the power being

introduced by the axial chamber plate is calculated and shown in

Fig. 18. Based on the data presented in this investigation, it is within

Fig. 16 Thruster body current to ground and thruster body floating

voltage as a function of axial plate bias.

Fig. 17 Notional diagram of electron pathways: a) no axial plate bias voltage or nominal condition, b) low axial plate bias voltage, and c) high axial plate

bias voltage.

Fig. 18 Power sourced by the axial chamber plate for both thruster

electrical configurations.

1374

WALKER ET AL.

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

reason to conclude that the power introduced by the axial chamber

plate, in manipulation of the electron termination pathways, helps

offset the energy normally loss to conductive grounded surfaces. The

compensation of power loss to the conductive walls of the vacuum

chamber helps make the HCT plume more representative of the

onorbit environment.

D. Enhanced Current Density in the Offaxis Plume

As shown in Fig. 11, the offaxis region of the plume has a current

density that is influenced by the axial chamber plate bias voltage. As

stated earlier, the Faraday probe has a fixed electron repulsion voltage

throughout the testing; therefore, it is not immediately clear if the

measured increase in the current density profile is due to changes in

the plume or as a result of the Faraday probe fixed electron repulsion

voltage. Without further modeling, the increase in current collected

on the Faraday probe due to the increase in the potential gradient

between the probe and the plasma cannot be estimated. In Fig. 19,

data are presented that offer an alternative means of assessing the

validity of the measured current density in the offaxis region of the

plume. Faraday probe sweeps are taken at four different cathode

radial locations relative to thruster centerline: 18.1 (nominal

position), 21.9, 27.0, and 43.4 cm. At the cathode nominal position

and an axial chamber plate bias of 50 V, the current density is

approximately 25% higher as compared to values measured for the

grounded axial chamber plate condition. At other cathode positions,

the increase in current density is on the order of 45–50% relative to

current densities measured for the grounded axial chamber plate

conditions. This variation of behavior in the current density measured

is enough to suggest that the increase in current density measured for

high axial chamber plate voltages is due in part to actual changes in

the HCT plume.

E. Comparison to the Small Missions for Advanced Research in

Technology-1 Mission

From the perspective of the HCT electrical circuit, the bias voltage

of the axial chamber plate acts to enforce a pseudo-far-field plasma

potential boundary condition. The axial plate is able to drive the

plasma potential by mediating the electron-ion loss rate to the facility

walls. The resulting increase in plasma potential and cathode-to-

ground potential is similar to behavior observed during the SMART-1

mission [12,13]. The plasma potential, during the SMART-1 mission,

was measured using the electric propulsion diagnostic package

(EPDP) and placed downstream from the thruster exit plane and in a

“low” ion-energy region of the PPS-1350 plume [43]. The measured

difference between the cathode-to-ground voltage and the plasma

potential remained approximately 19 V [13] throughout the mission.

Fig. 19 Current density profiles in the offaxis region of the HCT plume for varying cathode positions relative to thruster centerline: a) 18.1 cm, b) 21.9 cm,

c) 27.0 cm, and d) 43.4 cm.

WALKER ET AL. 1375

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

As shown in Fig. 10, there is a similar fixed voltage difference

between the HCT floating voltage and the centerline plasma

potential. The influence on the axial chamber plate on the HCT

floating circuit voltage and plasma potential is only evident once the

axial chamber plate is able to collect a net electron current. Due to the

high mobility of the electrons versus xenon ions, this occurs at a low

(greater than 5 V) voltage above ground. Once the axial plate bias

voltage is able to established net electron current collection, the axial

chamber plate begins to induce global changes in the HCT plume and

HCT electrical circuit. Overall, the potential difference behavior

between the plasma potential and the cathode-to-ground potential for

the T-140 HCT tested is similar to the behavior experienced by the

PPS-1350 in-flight operation.

A closer examination of the data collected in this experiment

reveals that the difference between the plasma potential and the

cathode-to-ground voltage has a small dependence on the axial

chamber plate bias. The difference between the plasma potential at an

angular position of −45 deg relative to thruster centerline and

cathode-to-ground potential is shown in Fig. 20. Given the placement

of the EPDP, this angular position is in a similar region of the HCT

beam. At lower axial chamber plate biases (0 –15 V), where the beam

current is not fully collected by the plate, there is an increase in this

voltage difference by approximately 1–2 V as compared to the

nominal case. At higher axial chamber plate biases (20–50 V), the

difference between the cathode-to-ground voltage and the plasma

potential increases by approximately 2–4 V relative to the nominal

condition. This voltage difference behavior is consistent with trends

observed at other angular positions. The increase in the potential

difference will result in a change in thrust of the T-140 that will be

smaller than the resolution of the calibration of the thrust stand used

in this investigation and is consistent with expectations based on the

work of Frieman et al. [20]. Though the change in thrust is not

measurable with the thrust stand used in this investigation, the change

in the potential difference between the cathode and the ambient

plasma potential is indicative of a change in the efficiency of the HCT

electrical circuit in extracting electrons from the cathode [29]. In

connection to the in-flight environment, the HCT plasma potential is

heavily influenced by the interaction with any charged surfaces, such

as the unshielded low-voltage solar panel contacts as seen in the

SMART-1 mission [12,13]. As seen in this investigation, a variation

in the plasma potential boundary condition relative to the HCT will

result in changes to cathode coupling efficiency.

V. Conclusions

This investigation reveals the impact of biasing a downstream

electrode on a Hall current thruster (HCT) in a vacuum chamber

environment. The bias of a downstream electrode is able to induce

global changes to the HCT plume and impacts the electron

termination pathways. At an axial chamber plate electron current

collection below the HCT beam current, the resultant change in HCT

behavior is consistent with observed in-flight behavior. The axial

chamber plate is able to drive global changes in the plasma potential

that affect the floating voltage of the HCT circuit and the available

electron termination pathways. As shown through first-order

analysis, there is electron current collected on the axial chamber plate

that is not sourced from the HCT, but it is relatively small compared to

the incident beam current. At an axial chamber plate electron current

collection well beyond above the HCT beam current, grounded

chamber surfaces become a source for electrons due to the

electrostatic potential gradient between the grounded facility walls

and the local plasma potential. Such a source of electrons is not

representative of the space environment and augments the

electrostatic acceleration potential by a few volts. The overall results

of this work show that, by biasing an electrode in the downstream

plume of the HCTabove the chamber ground, the net electron current

collection on the grounded chamber surfaces and thruster surfaces

can be eliminated and global changes in the plasma potential can have

an effect on cathode coupling efficiencies.

Acknowledgments

Jonathan Walker is supported by National Science Foundation

Graduate Research Fellowships under grant no. DGE-1148903. The

authors would like to thank Natalie Schloeder and Aaron Schinder for

their assistance in the collection of data for and in preparation of this

paper. Jonathan Walker, Samuel Langendorf, and Mitchell Walker

would like to acknowledge the Lockheed Martin Space Systems

Company for support of this investigation.

References

[1] Walker, M. L. R., and Gallimore, A. D., “Neutral Density Map of Hall

Thruster Plume Expansion in a Vacuum Chamber,” Review of Scientific

Instruments, Vol. 76, No. 5, 2005, Paper 053509.

doi:10.1063/1.1915011

[2] Kamhawi, H., Huang, W., Haag, T., and Spektor, R., “Investigation of

the Effects of Facility Background Pressure on the Performance and

Operation of the High Voltage Hall Accelerator,” 50th AIAA/ASME/

SAE/ASEE Joint Propulsion Conference, AIAA Paper 2014-3707,

July 2014.

[3] Huang, W., Kamhawi, H., Lobbia, R. B., and Brown, D. L., “Effect of

Background Pressure on the Plasma Oscillation Characteristics of the

HiVHAc Hall Thruster,” 50th AIAA/ASME/SAE/ASEE Joint Propulsion

Conference, AIAA Paper 2014-3708, July 2014.

[4] Diamant, K., Spektor, R., Beiting, E., Young, J., and Curtiss, T., “The

Effects of Background Pressure on Hall Thruster Operation,” 48th

AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit,

AIAA Paper 2012-3735, July 2012.

[5] Hofer, R. R., and Anderson, J. R., “Finite Pressure Effects in

Magnetically Shielded Hall Thrusters,” 50th AIAA/ASME/SAE/ASEE

Joint Propulsion Conference, AIAA Paper 2014-3709, 2014.

[6] Diamant, K. D., Liang, R., and Corey, R. L., “The Effect of Background

Pressure on SPT-100 Hall Thruster Performance,” 50th AIAA/ASME/

SAE/ASEE Joint Propulsion Conference, AIAA Paper 2014-3710,

2014.

[7] Nakles, M. R., and Hargus, W. A., Jr., “ Background Pressure Effects on

Ion Velocity Distribution Within a Medium-Power Hall Thruster,”

Journal of Propulsion and Power, Vol. 27, No. 4, 2011, pp. 737–743.

doi:10.2514/1.48027

[8] Brown, D. L., and Gallimore, A. D., “Evaluation of Facility Effects on

Ion Migration in a Hall Thruster Plume,” Journal of Propulsion and

Power, Vol. 27, No. 3, 2011, pp. 573–585.

doi:10.2514/1.B34068

[9] Walker, M. L. R., Victor, A. L., Hofer, R. R., and Gallimore, A. D.,

“Effect of Backpressure on Ion Current Density Measurements in Hall

Thruster Plumes,” Journal of Propulsion and Power, Vol. 21, No. 3,

2005, pp. 408–415.

doi:10.2514/1.7713

[10] Walker, M. L. R., “Effects of Facility Backpressure on the Performance

and Plume of a Hall Thruster,” Ph.D. Dissertation, Aerospace

Engineering, Univ. of Michigan, Ann Arbor, MI, 2005.

[11] Hofer, R. R., Peterson, P. Y., and Gallimore, A., “Characterizing

Vacuum Facility Backpressure Effects on the Performance of a Hall

Fig. 20 Potential difference between plasma potential and cathode-to-

ground potential as a function of axial plate bias voltage.

1376

WALKER ET AL.

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904

Thruster,” 27th International Electric Propulsion Conference, Electric

Rocket Propulsion Soc. IEPC Paper 01-045, Fairview Park, OH, 2001.

[12] Passaro, A., Vicini, A., Nania, F., and Biagioni, L., “Numerical

Rebuilding of SMART-1 Hall Effect Thruster Plasma Plume,” Journal

of Propulsion and Power, Vol. 26, No. 1, 2010, pp. 149–158.

doi:10.2514/1.36821

[13] Koppel, C., Marchandise, F., Estublier, D., and Jolivet, L., “ The Smart-1

Electric Propulsion Subsystem In Flight Experience,” 40th AIAA/

ASME/SAE/ASEE Joint Propulsion Conference and Exhibit, AIAA

Paper 2004-3435, 2004.

[14] Boyd, I. D., and Dressler, R. A., “Far Field Modeling of the Plasma

Plume of a Hall Thruster,” Journal of Applied Physics, Vol. 92, No. 4,

2002, pp. 1764–1774.

doi:10.1063/1.1492014

[15] Manzella, D. H., Jankovsky, R., Elliot, F., Mikellides, I., Jongeward, G.,

and Allen, D., “Hall Thruster Plume Measurements On-board the

Russian Express Satellites,” 27th International Electric Propulsion

Conference, Electric Rocket Propulsion Soc. IEPC Paper 2001-044,

Fairview Park, OH, 2001.

[16] Walker, J. A., Frieman, J. D., Walker, M. L. R., Khayms, V., King, D.,

and Peterson, P. Y., “Electrical Facility Effects on Hall-Effect-Thruster

Cathode Coupling: Discharge Oscillations and Facility Coupling,”

Journal of Propulsion and Power, Vol. 32, No. 4, 2016, pp. 844–855.

doi:10.2514/1.B35835

[17] Frieman, J. D., Walker, J. A., Walker, M. L. R., Khayms, V., and King,

D., “Electrical Facility Effects on Hall Effect Thruster Cathode

Coupling: Performance and Plume Properties,” Journal of Propulsion

and Power, Vol. 32, No. 1, 2016, pp. 251–264.

doi:10.2514/1.B35683

[18] Korsun, A. G., Tverdokhlebova, E. M., and Gabdullin, F. F., “The

Distinction Between the EP Plume Expansion in Space and in Vacuum

Chamber,” 29th International Electric Propulsion Conference, Electric

Rocket Propulsion Soc. IEPC Paper 2005-073, Fairview Park, OH,

2005.

[19] Tremolizzo, E., Meier, H., and Estublier, D., “In-Flight Disturbance

Torque Evaluation of the Smart-1 Plasma Thruster,” Proceedings of the

18th International Symposium on Space Flight Dynamics, Vol. 548,

2004, pp. 303–306.

[20] Frieman, J. D., King, S. T., Walker, M. L. R., Khayms, V., and King, D.,

“Role of a Conducting Vacuum Chamber in the Hall Effect Thruster

Electrical Circuit,” Journal of Propulsion and Power, Vol. 30, No. 6,

2014, pp. 1471–1479.

doi:10.2514/1.B35308

[21] Walker, J. A., Frieman, J. D., Walker, M. L., Khayms, V., King, D., and

Peterson, P., “Electrical Facility Effects on Hall Effect Thruster Cathode

Coupling: Discharge Oscillations and Facility Coupling,” 50th AIAA/

ASME/SAE/ASEE Joint Propulsion Conference, AIAA Paper 2014-

3711, 2014.

[22] Kieckhafer, A., and Walker, M. L. R., “Recirculating Liquid Nitrogen

System for Operation of Cryogenic Pumps,” 32nd International Electric

Propulsion Conference, IEPC Paper 2011-217, 2011.

[23] Tilford, C. R., “Sensitivity of Hot Cathode Ionization Gages,” Journal of

Vacuum Science and Technology, A, Vol. 3, No. 3, 1985, pp. 546–550.

doi:10.1116/1.572991

[24] “571 Ionization Gauge Tube Instruction Manual,” Manual

No. 6999905571, Rev. C, Agilent Technologies Vacuum Products

Division, Lexington, MA, 2002.

[25] McLean, C., McVey, J., and Schappell, T., “Testing of a U.S.-Built HET

System for Orbit Transfer Applications,” 35th Joint Propulsion

Conference and Exhibit, AIAA Paper 1999-2574, 1999.

[26] Snyder, J. S., Baldwin, J., Frieman, J. D., Walker, M. L. R., Hicks, N. S.,

Polzin, K. A., and Singleton, J. T., “Flow Control and Measurement in

Electric Propulsion Systems: Towards an AIAA Reference Standard,”

33rd International Electric Propulsion Conference, Electric Rocket

Propulsion Soc. IEPC Paper 2013-425, Fairview Park, OH, 2013.

[27] Spektor, R., Willholf, M. A., Beiting, E., and Diamant, K.,

“Characterization of the High Power Propulsion System Subscale,”

60th JANNAF Propulsion Meeting, Johns Hopkins Univ. JANNAF

Paper 2013-3142, Baltimore, MD, 2013.

[28] Xu, K. G., and Walker, M. L. R., “High-Power, Null-Type, Inverted

Pendulum Thrust Stand,” Review of Scientific Instruments, Vol. 80,

No. 5, 2009, Paper 055103.

doi:10.1063/1.3125626

[29] Goebel, D. M., and Katz, I., Fundamentals of Electric Propulsion: Ion

and Hall Thrusters, Wiley, Hoboken, NJ, 2008, pp. 24, 124–133, 342,

408–411.

[30] Sheehan, J. P., and Hershkowitz, N., “Emissive Probes,” Plasma

Sources Science and Technology, Vol. 20, No. 6, 2011, pp. 1–22.

doi:10.1088/0963-0252/20/6/063001

[31] Demidov, V. I., Ratynskaia, S. V., and Rypdal, K., “Electric Probes for

Plasmas: The Link Between Theory and Instrument,” Review of

Scientific Instruments, Vol. 73, No. 10, 2002, pp. 3409–3439.

doi:10.1063/1.1505099

[32] Hutchinson, I. H., Principles of Plasma Diagnostics, Cambridge Univ.

Press, New York, 2002.

[33] Hofer, R. R., Walker, M. L. R., and Gallimore, A., “

A Comparison of

Nude and Collimated Faraday Probes for Use with Hall Thrusters,” 27th

International Electric Propulsion Conference, Electric Rocket

Propulsion Soc. IEPC Paper 2001-020, Fairview Park, OH, 2001.

[34] Brown, D. L., “Investigation of Low Discharge Voltage Hall Thruster

Characteristics and Evaluation of Loss Mechanisms,” Ph.D.

Dissertation, Aerospace Engineering, Univ. of Michigan, Ann Arbor,

MI, 2009.

[35] Brown, D. L., and Gallimore, A. D., “Evaluation of Ion Collection Area

in Faraday Probes,” Review of Scientific Instruments, Vol. 81, No. 6,

2010, Paper 063504.

doi:10.1063/1.3449541

[36] Ekholm, J., and Hargus, W., “E × B Measurements of a 200 W Xenon

Hall Thruster,” 41st AIAA/ASME/SAE/ASEE Joint Propulsion

Conference and Exhibit, AIAA Paper 2005-4405, 2005.

[37] Korsun, A., and Tverdokhlebova, E., “The Characteristics of the EP

Exhaust Plume in Space,” 33rd Joint Propulsion Conference and

Exhibit, AIAA Paper 1997-3065, 1997.

[38] Azziz, Y., “Instrument Development and Plasma Measurements on a

200-Watt Hall Thruster Plume,” M.S. Thesis, Aeronautics and

Astronautics, Massachusetts Inst. of Technology, Cambridge, MA,

2001.

[39] Baglin, V., Bojko, J., Grobner, O., Henrist, B., Hilleret, N., Scheurlein,

C., and Taborelli, M., “The Secondary Electron Yield of Technical

Materials and its Variation with Surface Treatments,” 7th European

Particle Accelerator Conference, Large Hadron Collider Project Rept.

433, European Organization for Nuclear Research, Vienna, 2000.

[40] Pillon, J., Roptin, D., and Cailler, M., “Secondary Electron Emission

from Aluminium,” Surface Science, Vol. 57, No. 2, 1976, pp. 741–748.

doi:10.1016/0039-6028(76)90359-9

[41] Yamauchi, Y., and Shimizu, R., “Secondary Electron Emission from

Aluminum by Argon and Oxygen Ion Bombardment Below 3 keV,”

Japanese Journal of Applied Physics, Vol. 22, No. 4A, 1983, Paper

L227.

[42] Anatoly, K., Ekaterina, T., Flur, G., and Sylvia, B., “The Electric

Currents and Potentials Generated by Plasma Plume over SC Structure

Elements,” 42nd AIAA Aerospace Sciences Meeting and Exhibit, AIAA

Paper 2004-0985, 2004.

[43] Matticari, G., Noci, G., Estublier, D., del Amo, J. G., Marini, A., and

Tajmar, M., “The SMART-1 Electric Propulsion Diagnostic Package,”

Third International Conference on Spacecraft Propulsion, Vol. 465,

Aeronautique Astronautique de France, Paris, France, 2000, pp. 661–

668.

J. Blandino

Associate Editor

WALKER ET AL. 1377

Downloaded by GEORGIA INST OF TECHNOLOGY on July 1, 2017 | http://arc.aiaa.org | DOI: 10.2514/1.B35904