Microsoft Word - ALFA2 NEP Miss

41st AIAA/ASME/SAE/ASEE Joint Propulsion Conference and Exhibit

AIAA 2005-3894

Tucson, Arizona, 10-13 July 2005.

* Senior Staff, Advanced Propulsion Technology Group, 4800 Oak Grove Drive, M/S 125-109, Member AIAA.
** Ph.D. Candidate, 1200 E. California Blvd., Member AIAA.
Copyright © 2005 by the American Institute of Aeronautics and Astronautics, Inc. The U.S. Government has a
royalty-free license to exercise all rights under the copyright claimed herein for Governmental purposes. All other
rights are reserved by the copyright owner.

Identification of Mission Sensitivities for High-Power

Electric Propulsion Systems

Robert H. Frisbee*

Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, 91109

Robert C. Moeller**

California Institute of Technology, Pasadena, CA, 91125

The Advanced Lithium-Fed, Applied-field Lorentz Force Accelerator (ALFA

) project is

developing a high-power magnetoplasmadynamic thruster that could demonstrate attractive
performance for a variety of high-power electric propulsion missions under consideration by
NASA. Both robotic and Human Cargo interplanetary missions may benefit from ALFA

technology, as well as LEO-to-GEO transfers and escape trajectories. The thruster uses
lithium (Li) propellant to achieve an efficiency of about 60% at a nominal specific impulse
(I

) of 6,200 lb

-s/lb

, with other specific impulses in the range of 4,000 to 7,000 lb

-s/lb

under consideration. This paper presents the results of mission analyses that expose various
mission performance sensitivities and system advantages of the ALFA

technology for a

small but representative subset of nuclear electric propulsion (NEP) missions considered
under NASA’s Project Prometheus. Multiple technology parameters (e.g., thruster
efficiency, throughput, and specific mass) are examined to determine the overall mission
sensitivity to these performance metrics. The analyses include comparison studies of the
ALFA

technology relative to state-of-the-art Ion propulsion systems and quantify the

significant benefits of the technology such as high power-per-thruster and high I

in a

simplified, steady-state, low mass, and small volume propulsion system.

I. INTRODUCTION

In 2004, NASA released a competitive NASA Research Announcement (NRA) to develop Advanced

Electric Propulsion (AEP) technologies.

The ultimate goal of this activity is defined by the NRA Scope of Program

description:

“The goal of this AEP Technologies program is not to develop flight-qualified hardware, but to promote and advance
the development of very high power, AEP thruster technologies that result in reduced AEP system mass and
complexity and that may enable future missions that might otherwise not be considered credible and to deliver
conceptual AEP system designs. As spacecraft power levels become very high, building high powered gridded ion or
Hall thrusters (>100 kWe) or clustering large numbers of moderately powered (~25 kWe) thrusters becomes massive,
voluminous, and complicated. The proposed AEP thruster system must offer advantages at a system level over an
equivalently performing gridded ion or Hall thruster systems, as well as improvements in component and system
lifetimes and performance over the current state-of-the-art (SOA) of AEP systems.”

(Emphasis added)

As one of the awards granted by the NRA, the Advanced Lithium-Fed, Applied-field Lorentz Force

Accelerator (ALFA

) project combines a Princeton University-led team in collaboration with the Jet Propulsion

Laboratory (JPL), NASA, industry, and academia to develop a next-generation lithium-fed, applied-field
magnetoplasmadynamic thruster (AF-MPDT). Such a device, also called a “Lorentz Force Accelerator” (LFA), will
be designed to optimize and demonstrate its performance and life at 245-250 kW

and efficiency around 60-63%. A

specific impulse (I

) of 6,200 lb

-s/lb

is the nominal design point, although a range of 4,000 to 7,000 lb

-s/lb

under consideration. ALFA

leverages research conducted over the past two decades at the Moscow Aviation

Institute (MAI), Princeton, and JPL. The ultimate goal of the ALFA

project is to develop a robust and compact

steady-state thruster that could benefit various high-power missions considered by Project Prometheus.

This leap forward in high-power electric propulsion thruster technology will give NASA new robotic

exploration capabilities and is also a major step toward MW

-class systems for supporting human exploration. In

this paper we present mission analyses for several such candidate nuclear electric propulsion (NEP) missions, and
quantify the mission and system benefits of employing ALFA

technology as the primary onboard propulsion.

Parametric variation of key technology performance metrics (e.g., thruster efficiency, throughput, and specific mass)
is also used to determine the mission sensitivities of ALFA

in NEP systems. Comparisons are made to relative to

American Institute of Aeronautics and Astronautics

state-of-the-art Ion propulsion systems (the Herakles Prometheus design) as a baseline. The results quantify the key
benefits of the ALFA

technology in reducing system complexity due to the high power-per-thruster and high Isp in

a simplified, steady-state, low mass, and small volume propulsion system.

II. DESCRIPTION OF THE ALFA

PROPULSION SYSTEM

Under NASA Project Prometheus NRA funding, the ALFA

project is currently in the first phase of its

program to develop an applied-field magnetoplasmadynamic thruster (AF-MPDT) operating with lithium (Li)
propellant relevant high-power electric propulsion mission applications. The current base period of study will
produce a conceptual design and development plan. A simplified cutaway view of the thruster can be seen in Figure
1. Subsequent phases will result in a laboratory demonstration of an AF-MPDT meeting the NRA performance
objectives and validated models of performance and life-limiting phenomena.

Figure 1. Cutaway View of ALFA

Thruster.

MPDTs utilize the JxB electromagnetic Lorentz force to accelerate plasma. By operating in steady-state

with high currents within the thruster plasma discharge between an inner cathode and concentric anode, the plasma
discharge produces the necessary current (J) and self-induced azimuthal magnetic field (B) to produce the required
acceleration mechanism from the Lorentz force for thrust. As shown in Figure 2, an AF-MPDT like ALFA

adds

additional performance enhancements by introducing an externally-applied poloidal magnetic field (with radial and
axial components) as an additional design parameter.

2,3

The ALFA

design is presently optimized for an efficiency of 60 to 63% and a nominal I

of 6,200 lb

-s/lb

at 245 kW

total thruster power, which includes power to the thruster, the solenoid for generating the applied

magnetic field, and the lithium vaporizer. The project will demonstrate electrode geometries, materials and thermal
management schemes that mitigate erosion processes, yielding life ≥3 years. ALFA

will significantly advance the

state-of-the-art (SOA) in lithium-fed AF-MPDTs, and such a thruster would greatly reduce the mass, volume, and
complexity of a potential high-power electric propulsion (EP) system. Table 1 compares the proposed ALFA

engine with the state-of-the-art (SOA) resulting from past research. To achieve these goals, the ALFA

project will

optimize the design using the detailed analysis models of the project team and the natural benefits of Li as a

American Institute of Aeronautics and Astronautics

propellant to achieve these goals. The project is also investigating two technologies (magnetic field optimization and
anode propellant injection) with strong potential for further performance gains beyond the nominal design point.

Figure 2. Simplified View of Plasma Acceleration in ALFA

Thruster.

Table 1. ALFA

Provides Significant Improvement of State-of-the-Art Lithium AF-MPDT Technology.

Parameter

NRA Requirement

MAI MPD-200 SOA

ALFA

% Gain Over SOA

Power/Thruster (kW

)

100-250

192.7

245

Efficiency (%)

≥ 60

60-63

Specific Impulse (lb

-s/lb

)

≥ 4000

4250

6200

Lifetime (years)

1-3

(*)

≥3

*Previous programs focused primarily on demonstrating performance.

Improvements in performance and life in such a high-power, compact, steady-state thruster will yield major

benefits for NASA high-power electric propulsion missions, from vehicles with many 100s of kW

for robotic deep

space missions and up to MW

-class systems for Lunar and Mars Cargo missions. The ongoing effort will advance

the thruster technology from technology readiness level (TRL) 4 to 5, and associated technologies from TRL 3 to 5.
Examples of the potential benefits of this technology include:

Flight System Benefits:

•

Significantly reduced volume for configuration and packaging of thrusters due to ability to process high power
in small volume.

•

Fewer thrusters.

•

Reduced propulsion system complexity and parts count (PPUs, feed system components, etc.).

•

Lower propulsion system mass.

•

Steady-state operation greatly simplifies propellant feed and power systems, enhances robustness and reliability.

•

Feed system with no moving parts (Li fed using small electromagnetic pumps) improves reliability.

•

Li propellant efficiently stored as compact solid in tanks without need for cryogenics and low pressure liquid
during operation.

•

Passive cooling of the thruster (via anode radiator coupling) avoids need for less reliable active cooling systems.

•

Availability of Li relative to xenon propellant (presently about 12,000 metric tons [MT] per yr Li production
compared to 35 MT/yr Xe production).

•

Potential benefits of lithium propellant itself to provide significant radiation shielding in NEP applications,
particularly with lithium as a good neutron moderator, ay reduce the mass of reactor shielding required.

Demonstrated Ability to Process 100s of kW

•

Lithium MPDTs have the unique and demonstrated ability to efficiently (>50%) process very high power (up to
500 kW

demonstrated steady-state), in a single compact thruster, to produce steady state thrust-to-power

exceeding 20 N/MW

, specific impulses exceeding 4,000 lb

-s/lb

and thrust densities above 200 N/m

2,3,4

American Institute of Aeronautics and Astronautics

Advantages of Lithium:

•

Li has uniquely low frozen flow losses as a propellant because the ionization energy is so low (5.39 eV) yet the
first excited state and second ionization potential energies are high. Thus, little power is consumed in ionization
and maintaining the discharge.

•

Also, as a significant benefit to high-power ground testing facilities, lithium condenses on inexpensive, water-
cooled vacuum chamber surfaces and does not need to be pumped out of the chamber. In contrast to
noncondensable gases, which require very high pump speeds. This reduces facility pumping needs by orders of
magnitude. For future long lifetime system tests on the ground, the Li propellant can easily be recycled with
closed loop Li purification cycles, similar to those already done with other existing alkali metal (e.g., sodium)
closed loop test facilities for power conversion.

III. MISSION ANALYSIS RESULTS FOR ROBOTIC OUTER SOLAR SYSTEM MISSIONS

A. Introduction

A number of outer Solar System Nuclear Electric Propulsion (NEP) robotic exploration missions have been

under consideration as follow-on missions beyond the Jupiter Icy Moon Orbiter (JIMO) mission. Typically, because
of the need for short trip times to these distant destinations, the anticipated mission

∆

Vs and NEP total or “bus”

power levels are significantly higher than those for the more near-term JIMO mission. However, Ion thrusters, such
as the Herakles ion thruster proposed for use in the JIMO mission, are inherently low power density electric
propulsion devices; at higher vehicle power levels (multi-100 kWs to MWs), the low power-per-thruster of an Ion
thruster can result in the need for many thrusters with a corresponding increase in system mass and complexity.
Thus, we see the potential advantage of the ALFA

thruster with its almost order-of-magnitude increase in power-

per-thruster over the Ion thruster.

For these mission analyses, we selected two extreme cases of possible JIMO follow-on missions: the first is

a Saturn Orbiter with Moon Tour with a somewhat higher mission

∆

V than JIMO (e.g., for departure from Earth

escape [C

=0], the total mission

∆

V is ca. 33 km/s for the Saturn mission vs only ca. 31 km/s for the JIMO

mission),

and an Interstellar Precursor mission, with

∆

Vs ranging from about 28 to 53 km/s (for C

=0) depending

on the final Solar System escape velocity (V

inf

B. Nuclear Electric Propulsion Robotic Outer Solar System Exploration Vehicle Assumptions

The NEP vehicle consists of a nuclear-electric power system, a main boom that is used both to support the

power system’s radiators and to separate the spacecraft systems and payload from the reactor’s radiation, a power
management and distribution (PMAD) system (high-power cabling between the power system and the spacecraft
bus), and the spacecraft bus and payload. The spacecraft bus contains the reaction control system (RCS), various
miscellaneous spacecraft systems (e.g. telecommunications, etc.), and the electric propulsion system.

1. NEP Vehicle Configuration

A conceptual schematic of the NEP vehicle is shown in Figure 3. Note that some electric propulsion

options may require the use of a plume shield to protect sensitive spacecraft surfaces from the thrusters’ exhaust
plume. For example, spacecraft surfaces can be subjected to thermal contamination or physical erosion from the
high-energy plume. Also, there is the risk of material contamination by condensable propellants, such as the lithium
used in the ALFA

thruster. Material contamination can be a serious concern for optically-sensitive systems like

camera lenses, and especially for radiators where the contaminant material can change the radiator’s emissivity.
Also, placing the electric thrusters at the far end of the vehicle, with a 180

field of view to space, ensures that no

spacecraft surfaces will experience thermal contamination or erosion impact from the thruster plumes. Fortunately,
plume contamination with ALFA

can be mitigated by use of a plume shield. Thus, we included a 10-m diameter

plume shield in our analyses of Electric Propulsion (EP) vehicles using ALFA

thrusters.

2. Nuclear Electric Power System

Mass scaling estimates for the nuclear-electric power system were provided by Lee Mason (GRC).

The

power system consists of a reactor, shield, heat exchanger, dynamic (Brayton) thermal-to-electric power conversion
system, and waste heat rejection system (radiators, pumps, fluid loops, etc.). The overall system specific mass
(kg/kW

) is shown in Figure 4 as a function of total or “bus” power level. Also shown are the specific masses of the

main boom and PMAD systems discussed below. As is commonly observed in space-based NEP power systems,

American Institute of Aeronautics and Astronautics

there is a significant economy-of-scale at higher powers. This often tends to drive the power requirement to high
power levels in order to achieve short trip times, because the overall vehicle specific mass is inversely proportional
to overall vehicle acceleration for a given I

. Also, higher powers reduce the effective specific mass of fixed-mass

vehicle elements (e.g., payload), again favoring increased power for short trip times.

Figure 3. Conceptual Schematic of an NEP Vehicle.

Figure 4. Assumed NEP Power System, Main Boom, and Power Management and Distribution (PMAD) System

Specific Mass as a Function of Total “Bus” Power.

3. Main Boom

In an NEP vehicle, the main boom runs from the power system to the main spacecraft bus, as illustrated in

Figure 3. The boom serves to both support the power system’s radiators, as well as provide standoff distance to
reduce the radiation load on the spacecraft systems and payload. Mass scaling estimates for the main boom were
provided by Muriel Noca (JPL).

The mass of the main boom includes the forward (power-system end) equipment

structure that attaches the boom system to the power system, the boom deployment canister (including drive motors,
actuators, etc.), the boom structure itself, its mounting structure, cabling (PMAD) attachment hardware, thermal
control, and micrometeoroid protection. As shown in Figure 4, there is only a slight economy-of scale due to the
need to strengthen the boom structure to prevent buckling as the boom length increases with increased power.

For calculation purposes, the mass per unit length of the main boom was calculated based on the radiator

length, which in turn is a function of the radiation shield’s shadow half-angle (assumed to be 10

An additional

10-m length of boom was then added to represent separation distance between the hot radiator and the spacecraft
bus.

American Institute of Aeronautics and Astronautics

4. Power Management and Distribution (PMAD)

The power management and distribution (PMAD) system consists of high-power cabling between the

power system and the spacecraft bus. Mass scaling estimates for the NEP PMAD system were provided by Lee
Mason (GRC).

As with the main boom, the mass per unit length of the PMAD cabling was estimated based on the

boom length, and an additional 10 m of PMAD was added to correspond to the extra 10 m of boom length.

As shown in Figure 4, there can be a substantial economy of scale for the PMAD system. Also note that in

an NEP vehicle, the electric power produced by the turboalternators in the dynamic power system is in the form of
high-voltage AC. By contrast, the electric power from the photovoltaic arrays in a Solar Electric Propulsion (SEP)
vehicle is in the form of relatively low-voltage DC. Thus, for a given power level, the PMAD for an SEP “bus” will
be somewhat heavier than for an NEP system, although the mass penalty for a low-voltage SEP system is somewhat
offset by the additional electrical insulation and isolation required for a high-voltage NEP system. In fact, the
greatest PMAD impact is in the electric propulsion system power processing units (PPUs) that convert the “bus”
voltage to the form required by the electric thruster. For example, an Ion thruster requires very high-voltage
(typically many kV) DC. By contrast, an MPDT/LFA engine requires only modest voltage (typically ca. 100 V) DC.
Thus, the PPU for an NEP system requires only a transformer (to convert the “bus” voltage level to that required by
the thruster) and rectifiers (to convert the “bus” AC to DC). However, a PPU for an SEP vehicle requires an
additional initial DC-to-AC invertor, along with a transformer and rectifier as in the NEP PPU, so an SEP-PPU is
generally heavier than an NEP-PPU.

The topology of these various options is illustrated in Figure 5.

Figure 5. NEP and SEP Power System, PMAD, PPU, and Thruster Topology.

5. General Structural Mass Overhead, and Electric Power and Dry Mass Contingencies

The power system, main boom, and PMAD masses include their own structural overhead. For the other

various spacecraft systems described next, we have assumed a general structural overhead of 26% of the dry mass of
the various components based on JPL system design study practices.

Additionally, we have assumed a general

structural overhead of 4% of the mass of propellants.

These structural overheads represent miscellaneous structure

required to tie the various components together and to the NEP vehicle. Also, for those systems requiring electric
power, we have assumed a power contingency of 10%. The total power is subtracted from the total “bus” power in
determining the power available to the electric propulsion system.

Finally, all dry masses (other than the payload) have an additional 30% dry mass contingency.

This

contingency, although large, is representative of current baselines for estimating spacecraft mass growth. Note that
this is a relatively recent addition to mission analyses; many previous studies omitted this quantity.

6. Chemical Hydrazine (N

) Reaction Control System (RCS)

A sequentially-recharged blowdown chemical hydrazine (N

) reaction control system (RCS) is used to

maintain attitude control during all phases of the mission. The thrusters have an I

of 220 lb

-s/lb

; the mass of

hydrazine is determined based on a 50-m/s

∆

V for the fully loaded “wet” vehicle (with payload). The RCS

components have a fixed mass of 18 kg and a tankage factor of 8.92% (including 0.73% for residuals and holdup).
The RCS electric power requirement is 51 W

(primarily to power the hydrazine thruster catalyst bed heaters); a

10% margin is added to this for a total of 56 W

. Finally, as described above, an overall structural factor of 26% is

applied to the dry mass of the RCS (including tankage), and an additional 4% structure is applied to the mass of
hydrazine propellant.

7. Miscellaneous Spacecraft Systems

A number of “generic” robotic spacecraft systems

are included in the NEP vehicle’s mass. These systems

include attitude control systems (ACS), command and data handling (C&DH) systems, telecommunications
(Telecom), vehicle startup/emergency power (an allocation of 230 kg for some combination of batteries,

American Institute of Aeronautics and Astronautics

radioisotope thermoelectric generators [RTGs] for deep-space missions, or solar arrays for inner Solar System
missions), thermal control, and component-specific structure (108 kg). The total mass of the miscellaneous systems
is 387 kg (without component-specific structure); 26% of this mass is added to represent miscellaneous structure,
resulting in a total dry mass of 596 kg. The total power is 328 W

; a 10% margin is added to this for a total of 361

8. Science Payload

For these mission analyses, we have assumed a net science payload of 2,500 kg with a power requirement

of 1 kW

during NEP operation and cruise. This represents a significant mass of payload that could include landers

and probes for the Saturn Mission, or an independent, self-contained spacecraft for the Interstellar Precursor Mission
that could be jettisoned after the NEP vehicle’s propulsive burn to continue on its own out of the Solar System and
beyond. Also, a structural adaptor of 2.5% of the payload’s mass

is added to tie the payload to the NEP vehicle.

Additionally, the 30% dry mass contingency is added to the structural adaptor, but not the payload. Finally, a 10%
margin is added to the payload’s electric power requirement, such that the total power requirement is 1.1 kW

9. Ion (Herakles) Propulsion System

The Ion thruster used as a baseline for comparison with the ALFA

system is based on the Herakles thruster

proposed for the JIMO mission. We developed a series of scaling equations to calculate mass, power, efficiency, etc.
for this engine based on a theoretical model developed by Thomas Randolph (JPL), Doug Fiehler (QSS Group), and
Kurt Hack (GRC).

The thruster characteristics assumed for our analyses are illustrated in Figure 6. The thruster

mass is 58 kg (including an 8-kg cable between the thruster and the PPU) with a beam diameter of about 69 cm (the
overall outside diameter is roughly 94 cm). (The flattening out of thruster power above an I

of about 7,000 lb

-s/lb

is due to limiting of the beam current density.) Finally, the Herakles thruster is assumed to have a propellant
throughput of 5,500 kg of Xe propellant.

Figure 6. Ion (Herakles) Thruster and PPU Characteristics.

A mass list for a Xenon propellant storage and feed system was developed based on a design by Gani

Ganapathi (JPL).

For the Xe-propellant Ion thruster, we can store the propellant as high-pressure, room-

temperature supercritical gas and feed the high-pressure Xe to the thrusters. Also, as shown in the feed system
schematic below, a Xenon Recovery System (XRS) using a sorption compressor is used to scavenge residual Xe in
the main tank as it nears depletion. This has the effect of dramatically reducing the amount of residual Xe that would
otherwise be unavailable as pressure drops in the main tank. A schematic of the Xe-Ion thruster propellant storage
and feed system is shown in Figure 7. A summary of the various components is given in Table 2. Note that some
electric heater power is required to maintain the system at room temperature to compensate for heat lost due to
radiation in deep space far from the sun,

as well as additional electric power corresponding to the heat of

desorption of xenon in the Xenon Recovery System (XRS). Although this desorption power would only be needed
near the end of the mission when the XRS is operating, we have assumed that it would be required at all times; thus,
the various heater powers and XRS desorption power (equal to the XRS heat of desorption [440 J/g] multiplied by
the mass flow rate [g/s] of propellant into the Ion thrusters) are subtracted from the total “bus” power. Also, note
that there are both fixed-mass (or power) terms in the mass and power scaling relationships, as well as terms
dependant on propellant mass (M

) and on surface area (M

2/3

). Finally, a large number of ion thrusters are required

American Institute of Aeronautics and Astronautics

on a MW

-class vehicle. To facilitate vehicle integration and packaging in the launch vehicle, the Ion thrusters are

collected into two clusters or “pods,” with each pod having a gimbal and associated flex lines. The Ion thrusters are
mounted directly on the pod structure (i.e., not individually gimbaled).

Figure 7. Xe-Propellant Ion (Herakles) Thruster Storage and Feed System Schematic.

Table 2. Summary of Propellant Storage and Feed System Mass and Power Scaling for the Xe-Ion (Herakles)

Thruster System. (All masses in kg and powers in W

Component

Fixed Mass, Power, or Parts Term

% of M

Term

% of (M

)

2/3

Term

Storage & Feed System up to Pod Feed
Mass

63.85

2.5770%

1.5232%

Power

6.3

7.7371%

Parts Count

Pod Feed System (1 per Pod)
Mass (Includes Gimbal)

13.90

Power

0.0

Parts Count

Thruster Feed System (1 per Thruster)
Mass

14.22

Power

0.0

Parts Count

10. Thruster Number Calculation Methodology

For both the Ion and ALFA

propulsion systems, the number of thrusters (and their corresponding PPUs,

feed systems, etc.) required in each pod will be a function of the total “bus” power and the power-per-thruster of
each thruster. Specifically, we first calculate the net total power available to the thrusters by subtracting all the
various system electric powers (e.g., systems, payload, RCS catalyst bed heaters, propellant storage and feed system
heaters, etc.) from the total “bus” power. We then determine the power-per-PPU for the thrusters, which is equal to
the thruster’s power-per-thruster (typically a function of the thruster’s I

) divided by the PPU’s efficiency. The

number of operating thrusters is then simply the rounded-up integer of (available power)/(power-per-PPU). For the
Ion thruster system with two pods, it may be necessary to add one additional thruster so that each pod has the same
number of operating engines so as to ensure overall thrust balance (e.g., left/right symmetry). Also, for thrusters
with short lifetimes (i.e., low propellant throughput-per-thruster), it may be necessary to add extra complete sets of

American Institute of Aeronautics and Astronautics

thrusters so as to consume all of the propellant for the mission. Finally, one extra thruster (and its PPU) is added to
each pod as a redundancy spare. As a specific numerical example, if we have a 6,000-lb

-s/lb

Ion propulsion

system for a 1,000-kW

total “bus” power vehicle, the power available to the thrusters is 998.2 kW

. The power-per-

PPU is 23.9 kW

, so there are (998.2/23.9)=41.7 operating, which is rounded up to 42 thrusters. (This corresponds to

each of the 42 thrusters operating at 99.3% of their full rated power.) For the Ion system, we have two pods, which
implies a need for an even total number of operating thrusters. In this case, we already have an even total number
(i.e., 42 total, with 21 in each pod), so we do not need to add an additional thruster for thrust balance. (If there were
only one pod, then an even or odd number of operating thrusters would be allowed based on the assumption that any
gimbaling required to compensate for thrust imbalance would be small, and thus would not impact overall
performance, because all the thrusters would be relatively near the vehicle’s thrust centerline. By contrast, thrusters
in separate pods would be far from the vehicle thrust centerline, and thus produce a thrust imbalance moment-arm
that could not be corrected by gambling without unduly impacting performance.)

The Ion thrusters have a long lifetime, so only one set of operating thrusters is required to consume all of

the Xe propellant. Otherwise, it would be necessary to add additional sets of 42 thrusters (again distributed evenly
between the pods) until a cumulative total throughput (lifetime) was reached that consumed all of the required
propellant. In this case, the number of sets (with 42 thrusters in each set) would be the rounded-up integer value of
(total propellant mass)/(total throughput of 42 thrusters). Again, one extra spare thruster (and PPU) would be added
to each pod for redundancy. Finally, when we consider MW

-class electric propulsion vehicles, systems using

thrusters with modest power-per-thruster and modest lifetimes (throughput) will only require one set of thrusters
because the large total number of thrusters required to consume the available MW

of power naturally results in

sufficient numbers of thrusters to consume the available amount of propellant. Typically, the need for additional sets
of thrusters arises only when we have the combination of high power-per-thruster and low throughput-per-thruster,
as can be the case with magnetoplasmadynamic (MPD) thrusters.

11. ALFA

Propulsion System

The ALFA

thruster was described in detail above. For these analyses, we have assumed a thruster mass of

129 kg (including the applied-field magnets and Li vaporizer of the ALFA

engine system). The thruster has a

diameter of about 0.3 m. The ALFA

nominal (baseline) I

assumed for these analyses is 6,000 lb

-s/lb

, with an

efficiency of 60.0% at 6,000 lb

-s/lb

. We also evaluate the ALFA

operating at I

values ranging from 4,000 to

7,000 lb

-s/lb

, with efficiencies assumed constant at 60.0%, over this range. The throughput (lifetime) of the

thruster is assumed to be 8,300 kg per thruster independent of I

However, the thruster mass quoted above does not include a power processing unit (PPU) that converts the

NEP “bus” electric power (e.g., high-voltage AC) to the low-voltage DC power required by the ALFA

thruster.

Based on mass and efficiency scaling models provided by Alexander Kristalinski (Aerojet, Redmond WA),

have estimated the mass of an ALFA

NEP PPU as 360 kg with an efficiency of 98.3% (including power for the

applied-field magnets on the thruster counted as a “loss” in determining PPU efficiency). Lastly, we have added a
“generic” shielding mass for the Ion and ALFA

PPUs that corresponds to 0.92% of the NEP PPU mass.

This

shielding is intended to protect the PPU electronics from the general space environment, and not the much more
severe Jupiter radiation environment encountered in a JIMO-type mission.

Finally, a mass list for a lithium propellant storage and feed system was developed based on a design by

Joseph Lewis (JPL).

For the Li-propellant ALFA

, we can store the propellant before launch as a room-

temperature solid in the propellant tank, and then melt the Li once in space. (This step can occur at an arbitrarily
slow rate so as to not represent a major electric power impact, even though the full energy [enthalpy] of melting
would need to be provided.) For calculation purposes, the thermal control (heating) power requirement for
maintaining the propellant tank and feed system was determined by Robert Miyake (JPL)

based on an assumed

storage temperature of 230

C. (Li melts at 179

C.) Also, the high power-per-thruster of the ALFA

makes it

possible to have a small number of individually-gimbaled ALFA

engines in a single (non-gimbaled) thruster pod.

A schematic of the Li-ALFA

propellant storage and feed system is shown in Figure 8. A summary of the

various components is given in Table 3. In this system, a low-pressure (few psi) sequential-blowdown helium
pressurization system is used to pressurize the molten Li; this ensures that any Li vapors that could diffuse into the
pressurization system cannot cause a contamination-induced failure because the vapors are allowed to “see” only
already-discharged He pressurization modules (and are isolated from un-used units by normally-closed pyro valves).
With sufficient head pressure, the liquid metal is then forced into the electromagnetic pumps, which are used to push
the liquid through the feed system and into the thrusters where the liquid is vaporized. Interestingly, the
electromagnetic pumps serve the same general purpose as pressure regulators and valves in the Xe system; similarly,
Li flow sensors are analogous to gas pressure gages in a Xe system. Note that significant electric power is required
to maintain the system at an elevated temperature, as well as additional electric power corresponding to the heat of

American Institute of Aeronautics and Astronautics

vaporization of lithium (i.e., the power is equal to the heat of vaporization [21,279 J/g] multiplied by the mass flow
rate [g/s] of propellant into the ALFA

thrusters).

Figure 8. Li-Propellant ALFA

Storage and Feed System Schematic.

Table 3. Summary of Propellant Storage and Feed System Mass and Power Scaling for the Li-ALFA

System.

(All masses in kg and powers in W

Component

Fixed Mass, Power, or Parts Term

% of M

Term

% of (M

)

2/3

Term

Storage & Feed System up to Pod Feed
Mass

66.10

3.6341%

7.5425%

Power

69.0

155.3613%

Parts Count

Pod Feed System (1 per Pod)
Mass

1.67

Power

17.3

Parts Count

Thruster Feed System (1 per Thruster)
Mass (Includes Gimbal)

30.77

Power

17.3

Parts Count

C. Mission Analysis Results for NEP Robotic Outer Solar System Exploration Missions

We evaluated two robotic missions in these analyses. The first was a Saturn Orbiter Mission that includes

an extensive tour of Saturn’s moons. The second mission was an Interstellar Precursor Mission to 200 Astronomical
Units (AU) with a Solar System escape velocity (V

inf

) of either 5 or 10 AU/year. In each case, the trajectory data

used for the analyses did not include any gravity assists. Gravity assists could potentially reduce flight time for a
given mass but would limit the applicability of the analyses to very specific launch date opportunities. Thus, the
non-gravity-assisted results here are broadly applicable to approximating many potential future launch dates. Also,
he analyses for these missions were generated assuming a starting low-Earth orbit (LEO) at 1000 km altitude.

The

associated propulsive “spiral out” from this orbit to C

=0 added approximately 7.35 km/s to the total mission

∆

For each of these missions, we considered two primary factors in evaluating the mission benefits of the

ALFA

propulsion system:

American Institute of Aeronautics and Astronautics

1. Total Initial Mass in Low Earth Orbit (IMLEO) and Trip Time
2. Propulsion System Storage and Feed System Complexity (as represented by a parts count)

1. Initial Mass in Low Earth Orbit (IMLEO) vs Trip Time for the Saturn Orbiter with Moon Tour Mission

Figure 9 illustrates the IMLEO versus trip time performance for the Ion (Herakles) system over a range of

from 6,000 to 8,000 lb

-s/lb

, and the ALFA

thruster option over a range of I

from 5,000 to 7,000 lb

-s/lb

These analyses show the importance of a high I

for high-

∆

V missions; for example, there is a pronounced

difference between the 5,000- or 6,000-lb

-s/lb

curves and the 7,000- or 8,000-lb

-s/lb

curves. However, for

this high-

∆

V (ca. 41 km/s total leaving from a 1,000-km low Earth orbit [LEO]

) mission, the ALFA

results in

slightly longer trip times than the Ion system, having a roughly 1-year longer trip time than the Ion system at a fixed
initial mass. In this case, the superior thruster efficiency, I

, and especially cumulative throughput of all the running

thrusters of the Ion system provide superior performance in terms of initial launch mass and trip time. In particular,
the ALFA

system is severely penalized because of its modest throughput and small number of running thrusters.

For example, at an I

of 7,000 lb

-s/lb

, the ALFA

points in Figure 9 represent a system with two sets of thrusters

(each set run in series) to accommodate the total propellant throughput requirement; at lower values of I

(i.e., with

even more propellant), as many as four sets of ALFA

thrusters may be needed at the lowest power levels.

Figure 9. Variation in IMLEO and Trip Time for the Saturn Orbiter with Moon Tour Mission.

2. Initial Mass in Low Earth Orbit (IMLEO) vs Trip Time for the Interstellar Precursor Mission

Figure 10 illustrates the IMLEO versus trip time (to 200 AU) performance for Interstellar Precursor

mission with the Ion (Herakles) and ALFA

thruster types over a range of I

from 6,000 to 8,000 lb

-s/lb

, and

5,000 to 7,000 lb

-s/lb

, respectively. Performance for two cases is given; the first is for a slower final velocity (V

inf

)

of 5 Astronomical Units (AU) per year, and the second is for a faster V

inf

of 10 AU/year. Not surprisingly, the

slower mission is less demanding in overall mass and required power. As with the Saturn Mission, these analyses
show the importance of a high I

(and high cumulative throughput) for high-

∆

V missions; for example, there is a

modest difference between the higher I

values for the 5 AU/Yr mission which has a

∆

V on the order of 35 km/s

leaving from a 1,000-km LEO (comparable to the Saturn Mission total

∆

V); by contrast, there is a strong I

dependence for the 10 AU/year mission with a

∆

V around 60 km/s (leaving from a 1,000-km LEO).

American Institute of Aeronautics and Astronautics

However, at 750 kW

, the three running ALFA

thrusters are insufficient to consume all of the propellant, so an

extra set of three thrusters (along with their corresponding PPUs, feed systems, gimbals, etc.) must be added so as to
accommodate all the propellant, resulting in a total of 7 thrusters (including 1 spare). Similarly, at an I

of 5,000 lb

s/lb

, two or more sets of ALFA

thrusters are needed throughout the range of powers, whereas at 6,000 lb

-s/lb

two sets of thrusters are needed at power levels less than 2,250 kW

The results show that, for a given initial mass, the ALFA

system achieves very comparable trip times

(within 1 to 2% difference) to the Ion system for the lower

∆

V case (i.e., 35 km/s at 5 AU/year) at higher power

levels and I

. However, for the very high-

∆

V 10 AU/year mission, the superior thruster efficiency, I

, and

especially cumulative throughput (i.e., number of running thrusters * throughput per thruster) potential of the Ion
system provides clearly superior performance (approximately 10% reduction in trip time for a given initial mass).

3. Propulsion System Complexity

Thus far we have concentrated on mass and trip time as the traditional figures of merit in determining the

benefit (i.e., feasibility) of the new propulsion technologies embodied in the ALFA

system. However, another

element of mission feasibility is the overall system “complexity.” In these high-level system analyses, “complexity”
is quantified as a measure of propulsion system parts count. No matter how it is quantified, high “complexity” is
generally considered undesirable because of its perceived impact on decreasing system reliability and increasing
flight system integration and test costs.

Figure 11 illustrates the propulsion system parts count (e.g., thrusters and propellant storage and feed

system components [valves, regulators, filters, etc.]) as a function of the total or “bus” power level for the 5 AU/year
200 AU mission. (However, as discussed above, the situation for other missions at the same power level could be
different depending on the number of sets of thrusters required for total throughput.) In this case, the parts count is
used as a measure of system “complexity.” For example, propulsion systems like the high-power (e.g., 245 kW

per

thruster) ALFA

have a relatively small parts count, and thus ultimately “complexity,” for components like the

number of thrusters, valves, etc. By contrast, the Ion system, with its relatively low power-per-thruster (e.g., 30.4
kW

per thruster at 7,000-lb

-s/lb

) has an enormous parts count. Also note the increase in parts count for the Ion

thruster at 6,000 lb

-s/lb

; this is due to the dependence of power-per-thruster on I

in Ion thrusters, as illustrated

in Table 4.

Note also that the “jog” in the 7,000-lb

-s/lb

curve for the ALFA

points between 750 and 1,000 kW

Figure 11 corresponds to the same “jog” in the mass and trip time shown in Figure 10. As discussed above, this is
due to the need to add additional sets of thrusters to accommodate the total propellant throughput requirement. The
situation is even worse at lower values of I

, with a correspondingly larger total propellant mass. In these systems,

many sets of thrusters are required to consume the required propellant mass, although at higher powers, the larger
numbers of thrusters needed to consume the available total “bus” power can result in sufficient numbers of running
thrusters to consume the available propellant. For example, the 6,000-lb

-s/lb

curve for the ALFA

shows that

for “bus” powers less than 2,250 kW

, two sets of thrusters are required, but only one set is needed at 2,250 kW

and

above.

Figure 11. Electric Propulsion System Parts Count vs Total “Bus” Electric Power for the NEP 200 AU Interstellar

Precursor Mission (V

inf

= 5 AU/Year).

American Institute of Aeronautics and Astronautics

Table 4. Power-per-Thruster and Number of Thrusters Required for a 1-MW

NEP System.

Thruster

Ion (Herakles)

ALFA

(lb

-s/lb

)

6,000

7,000

8,000

6,000

Power into Thruster (kW

)

22.7

30.4

30.1

245.0

Power into PPU (kW

)

23.3

31.2

30.9

249.2

Number Running at 1 MW

Total “Bus” Power

Number per Pod

22 + 1 Spare

16 + 1 Spare

17 + 1 Spare

4 + 1 Spare

Number of Pods

Total Number of Thrusters

Also, although not explicitly considered in detail in this study, there is the non-trivial issue of packaging a

large number of thrusters in the Earth-launch vehicle launch shroud. For example, the Herakles Ion thruster system
requires approximately 40 times the surface area of the ALFA

system; when we take into account the need to have

roughly 8 times as many Herakles as ALFA

thrusters (e.g., due to the difference in power-per-thruster), we find that

there is a significant packaging and integration challenge for Ion thrusters in MW-class electric propulsion systems,
as illustrated in Figure 12 below.

Figure 12. Size Comparison Between Ion (Herakles) and LFA (ALFA

) Thrusters for a 1-MW

Total “Bus” Power.

D. Parametric Investigation of ALFA

Performance for NEP Robotic Outer Solar System Missions

As described above, we found that the performance of the ALFA

system for high-

∆

V missions is

somewhat less than that found for the Ion system for the nominal ALFA

thruster. In this section, we will investigate

approaches that could improve the performance of the ALFA

system in order to make it comparable to or superior

to the Ion system trip time performance, while maintaining the dramatic benefits in reduced packaging volume and
parts count associated with the ALFA

system.

1. Variation in ALFA

Thruster and PPU Specific Mass

For example, at a nominal I

of 6,000 lb

-s/lb

, the specific masses of the ALFA

thruster and PPU are

0.527 and 1.444 kg/kW

, respectively, for a total of 1.971 kg/kW

. Figure 13 illustrates the performance impact that

reductions in the thruster and PPU specific mass would have on the Saturn Mission described above. In this case,
both the thruster and PPU specific mass were reduced proportionally to the total shown in the Figure. We see here
that even with a thruster and PPU having zero mass, the ALFA

system does not surpass the performance of the Ion

system. In part, this is due to the need for multiple sets of feed systems for each thruster, even when the thruster and
PPU have no mass. Also, the propellant tankage for the low-density, high-temperature Li propellant is higher than
that for Xe in the Ion system, adding even more mass to the ALFA

system. This suggests that only limited

performance leverage can be obtained by attacking only the overall thruster and PPU mass, and this is
fundamentally due to the lower assumed thruster efficiency (60%) of the ALFA

system as compared to the high-

efficiency Ion system.

American Institute of Aeronautics and Astronautics

Figure 13. Variation in IMLEO and Trip Time for the Saturn Orbiter with Moon Tour Mission with ALFA

Thruster

and PPU Specific Mass Variations.

2. Variation in ALFA

Efficiency

The nominal ALFA

efficiency is 60% independent of I

. (For comparison, the Ion thruster has an

efficiency of 76.7% at an I

of 8,000 lb

-s/lb

.) Figure 14 illustrates the performance impact that improvements in

the thruster efficiency would have on the Saturn Mission. In this case, improvements in the ALFA

efficiency have

modest impact on performance. In contrast to the specific mass analysis above, this evaluation of thruster efficiency
suggests significant benefits for aggressively pursuing major improvements in ALFA

efficiency.

3. Variation in ALFA

Lifetime (Throughput)

As discussed above, the ALFA

system requires multiple sets of thrusters for the various robotic outer solar

system missions because of the ALFA

’s limited throughput (e.g., 8,300 kg per thruster) and small number of

running thrusters. For example, at 1 MW

total power, the cumulative throughput of the four running ALFA

engines is only 33,200 kg as compared to 176,000 kg for the 32 running 7,000-lb

-s/lb

Ion (Herakles) engines at

1 MW

. Figure 15 illustrates the performance for the Saturn and Interstellar Precursor missions with the ALFA

throughput increased sufficiently to allow operation with one set of thrusters at the lowest power level (i.e., with the
fewest number of thrusters required for power). We see the general trend that the required lifetime increases with
increasing propellant mass (i.e., with increasing mission

∆

V and decreasing I

). For example, the 5 AU/year

Interstellar Precursor mission, with the smallest

∆

V (ca. 35 km/s), only needs a 45% increase in ALFA

throughput

at an I

of 7,000 lb

-s/lb

, as compared to the 10 AU/year case, with the largest

∆

V (ca. 60 km/s), that needs a 2.75-

fold increase in ALFA

throughput (at 7,000 lb

-s/lb

4. Combinations of Variation in ALFA

Throughput, Thruster and PPU Specific Mass, and Thruster Efficiency

Thus far we have treated ALFA

specific mass, efficiency, and throughput as independent variables. In

Figure 16, we assume that the ALFA

throughput is increased by a factor of 2.0 for the Saturn mission (at an I

7,000 lb

-s/lb

) so that only one set of thrusters is required. We then vary specific mass or efficiency to evaluate the

benefits in mission performance that might be realized by advancing these technologies.

American Institute of Aeronautics and Astronautics

Assuming all the systems shown in Figure 16 have adequate throughput, we see that decreasing the ALFA

thruster and PPU specific mass by a factor of two has only modest benefit. By contrast, increasing ALFA

efficiency

from its nominal 60% to 70% results in mass and trip time performance comparable to the Ion system, suggesting
that a combination of improvements in ALFA

throughput and efficiency can have major mission performance

benefits.

Figure 16. Variation in IMLEO and Trip Time for the Saturn Orbiter with Moon Tour Mission with ALFA

Thruster

and PPU Specific Mass, and Thruster Efficiency Variations for a Thruster Throughput Large Enough to Enable a

Single Set of Thrusters.

E. Summary of Mission Analysis Results of ALFA

for NEP Robotic Outer Solar System Missions

As shown above, these high-

∆

V outer Solar System missions, where the vehicle spends most of its flight

time in heliocentric space, tend to optimize towards high values of I

and favor longer mission life (thus higher

thruster throughput). However, as will be shown below, missions within the inner Solar System tend to favor lower
I

values, because a greater fraction of the time is spent in planetary gravity wells and less time is available for

thrusting in transit to the target, thus driving up the need for higher thrust acceleration. In these cases, the lower I

produces more thrust (at a given power level), so as to potentially reduce the trip time. Nevertheless, even if limited
to an I

of 7,000 lb

-s/lb

, the Li-ALFA

system is still competitive with a high-performance Ion thruster system,

with the very significant advantages of easier spacecraft integration within the launch vehicle (i.e., smaller thruster
pods), and especially reduced system complexity as evidenced by a nearly order-of-magnitude reduction in the
number of system components for the ALFA

system as compared to the Ion system. Finally, there are several Li-

ALFA

technology improvements that could significantly enhance mission performance by combinations of reduced

thruster and PPU specific mass, increased thruster efficiency, or increased thruster propellant throughput.

IV. MISSION ANALYSIS RESULTS FOR THE NEP MARS CARGO MISSION

A. Introduction

Although often considered for high-power NEP robotic planetary exploration applications, high-power-per-

thruster systems like ALFA

can also be used for high-power electric propulsion (EP) Cargo Missions supporting

American Institute of Aeronautics and Astronautics

Human exploration of the Moon or Mars. In these “split” mission scenarios, time-insensitive cargo (e.g., propellants,
landers, surface habitats, etc.) is transported by a high-I

(i.e., fuel-efficient), although slow (i.e., low-T/W), EP

vehicle from LEO to an orbit about the target body. A fast (i.e., high-T/W) vehicle is used to transport the crew from
LEO to a rendezvous with the Cargo vehicle, where the pre-delivered supplies are then used for exploration of the
target. The crew then returns to Earth; in some scenarios this may be accomplished by using propellants delivered by
the Cargo vehicle. For lunar missions, the Cargo vehicle is typically re-used; by contrast, Cargo vehicles for Mars
missions are typically left in Mars orbit.

For the Mars Cargo Mission, we have assumed the use of a one-way (expendable) megawatt-class NEP

Cargo Vehicle for transport of payload from LEO to a 6,000-km altitude low Mars orbit (LMO). Typical

∆

Vs for

low-T/W LEO-to-LMO transfers are on the order of 16 km/s.

This LMO is at the same altitude as the inner, and

larger moon of Mars, Phobos. This high-altitude LMO, rather than a low-altitude (e.g., 400-km altitude) LMO was
chosen to make it possible to support exploration of Phobos, with special emphasis on Phobos as a potential
extraterrestrial resource of water that could be processed to produce chemical (O

) propellants. For example,

after delivery of the cargo payloads, the EP Cargo Vehicle could land on Phobos and use its power to support
mining, ore processing, water electrolysis, and so on. (Note that the chemical RCS thrusters might be needed for the
landing, however, because the vehicle acceleration from the electric thrusters might be too small even for the micro-
gravity surface gravity of Phobos.) Finally, the mass of Phobos could provide shielding mass to prevent radiation
from the NEP vehicle from damaging other assets in Mars space.

Also, we have chosen an Earth-to-Mars trip time goal of 2.2 years to match the Earth-Mars synodic period.

This makes it possible to launch the Cargo Vehicles during one trans-Mars injection (TMI) opportunity, travel to
Mars, perform Mars orbit insertion (MOI), and check out all the payload systems prior to launching the crew during
the next Mars TMI opportunity.

B. Nuclear Electric Propulsion Mars Cargo Vehicle Assumptions

1. NEP Systems

The assumptions made for the NEP vehicles described above are again used in this section.

2. Mars Cargo Payload

As with the Human missions to the Moon, architectures for Human exploration of Mars are still under

study. For our EP Cargo Vehicle analyses, we have assumed a 63.892-MT payload derived from the NASA Human
Exploration of Mars Design Reference Mission (DRM) Version 3.0.

This payload corresponds to delivery of an

Earth Return Vehicle (ERV) into Mars orbit. In the nominal DRM 3.0 Mission scenario, a Nuclear Thermal
Propulsion (NTP) stage is used for Earth escape and trans-Mars injection (TMI); the NTP stage is then jettisoned,
and the total payload (74.072 MT for the ERV and entry Aeroshell) is aerocaptured for Mars orbit insertion (MOI).
In the analyses here, we assume this Cargo transport function is performed by an electric propulsion Cargo Vehicle
instead. For the EP options, the Aeroshell (10.180 MT) required for the nominal NTP ERV Cargo Mission is
removed, because the EP Cargo Vehicle places the ERV directly into Mars orbit. There is also a second NTP Cargo
Vehicle launch with a payload consisting of a 66.043-MT Cargo Lander (CL) (to place an Ascent Vehicle and other
elements on the surface) that is aerobraked directly to the martian surface. Finally, the Crew Vehicle (with landers)
is sent by NTP on a fast trajectory to Mars where the Crew Vehicle aerocaptures into Mars orbit.

Interestingly, if a slow, minimum-energy (Hohmann) trajectory is acceptable for the crew, an aerobraked

chemical (O

) propulsion Crew Vehicle can provide comparable IMLEO to the NTP Crew Vehicle. However, the

real advantage of NTP is its combination of high-T/W and high-I

(projected

to be in the range of 940-960 lb

s/lb

); this makes it possible to fly fast, high-energy trajectories that have much shorter flight times than that for the

minimum-energy trajectory (e.g., 130-180 days versus the ideal 259 days, respectively, for the Earth-to-Mars step)
without suffering from an excessive IMLEO.

C. Mission Analysis Results for the NEP Mars Cargo Mission

In these systems analyses, we considered the following factors:

1. Total Initial Mass in Low Earth Orbit (IMLEO) and Trip Time
2. Propulsion System Complexity (as represented by a parts count)
3. Vehicle Power Level Required for a Given Earth-to-Mars Trip Time (nominally 2.2 year)

American Institute of Aeronautics and Astronautics

1. Initial Mass in Low Earth Orbit (IMLEO) vs Earth-to-Mars Trip Time

There is a strong similarity in performance between the ALFA

and Ion thruster systems for an NEP Cargo

Mission, as shown in Figure 17. For this mission, with a

∆

V typically on the order of 16 km/s, the optimum I

around 6,000-7,000 lb

-s/lb

; higher I

values result in the need for a higher power to achieve a given trip time such

that the increase in power system mass (and corresponding increase in thruster, PPU, etc. mass) essentially negates
any propellant (and propellant tankage) mass savings afforded by the higher I

. Also, it is interesting to note that in

the ALFA

NEP system, the optimum I

is around generally around 6,000 lb

-s/lb

rather than the 7,000 lb

-s/lb

the Ion system; this effect is probably due to the interaction of the thruster efficiency and thrust. More specifically,
vehicle thrust (and thus trip time) is proportional to the propulsion system’s exhaust or “jet’ power divided by I

;

thus, the ALFA

, with its lower efficiency, needs a lower I

to have the same thrust as an Ion thruster at a higher

efficiency and I

Figure 17. Variation in IMLEO and Trip Time for the NEP Mars Cargo Mission.

2. Propulsion System Complexity

Figure 18 shows the general trends seen previously where the low power-per-thruster of Ion thrusters

results in almost an order-of-magnitude increase in propulsion system storage and feed system parts count and thus
complexity over the Li-ALFA

system. Also shown in Figure 18 are the power levels required to achieve the target

Earth-to-Mars trip time of 2.2 years. Finally, for this mission with its modest

∆

V and thus propellant load, the

ALFA

throughput is adequate to enable a single set of thrusters at all but the lowest I

and power levels, so the

parts count curves for the three ALFA

cases generally fall on top of each other.

3. Vehicle Power Level Required for a 2.2-Year Earth-to-Mars Trip

As discussed above, we chose an Earth-to-Mars trip time goal of 2.2 years to match the Earth-Mars synodic

period. Figure 19 illustrates the general trend of requiring higher power at higher I

values in order to achieve a

desired trip time. We also see the similarity in IMLEO and power between the ALFA

and the Ion thruster systems.

Finally, it is worth noting the different contributions to dry mass in each vehicle. For example, the NEP systems
have a significant fraction of their dry mass tied up in the electric power system. However, the actual propulsion
system (e.g., thrusters, tankage, etc.) is relatively modest. By contrast, a Chemical or NTP Cargo Vehicle would

American Institute of Aeronautics and Astronautics

E. Summary of Mission Analysis Results of ALFA

for the NEP Mars Cargo Mission

Electric propulsion in general and ALFA

high-power electric propulsion in particular holds the promise of

providing significant mass savings for Cargo Missions in support of Human missions to Mars. Also, ALFA

propulsion technology improvements in throughput or efficiency are not required to match the mission performance
of Ion systems for Cargo missions. Finally, as with the robotic missions, the high power-per-thruster of the ALFA

system can provide major reductions in propulsion system complexity as compared to low power-per-thruster Ion
systems. The greatly reduced packaging volume and parts count of an ALFA

system dramatically simplifies the

integration complexities of a 1-2 MW-class EP Cargo Vehicle.

V. SUMMARY AND CONCLUSIONS

In these analyses, based on preliminary estimates of future ALFA

technology capabilities, we found that

the nominally proposed ALFA

propulsion system typically requires a modest trip time penalty (for a given IMLEO)

compared to an advanced Ion (Herakles) system for the limited set of NEP outer solar system robotic missions
considered. Moderate improvements in ALFA

throughput and efficiency can provide significant benefits by

enabling performance comparable to Ion thrusters. For the Mars Cargo mission, the ALFA

and Ion systems have

comparable mission performance even without further advancements in nominal ALFA

throughput or efficiency.

Further analyses including investigation of addition mission destination scenarios and sensitivities to mission-level
parameters (payload mass, science “tour” delta-V at the destination, etc.) are recommended.

For all missions examined, in response to one of the primary goals of the original NRA solicitation,

the

inherently high power-per-thruster of the ALFA

engine can result in almost an order-of-magnitude reduction in the

number of thrusters as compared to the inherently low power-per-thruster Ion engine. This reduction yields a
corresponding dramatic decrease in the packaging volume of the thrusters and reduction in the parts count of an
ALFA

propellant storage and feed system. This reduction in the ALFA

system complexity may ultimately prove

more attractive than any mass or trip time benefits of this technology by allowing the implementation of a more
reliable propulsion system with much simpler demands on the system integration and test process, and packaging
into a launch vehicle payload shroud volume.

VI. ACKNOWLEDGEMENTS

The work described in this paper was carried out at the Jet Propulsion Laboratory (JPL), California Institute

of Technology, under a contract with the National Aeronautics and Space Administration (NASA).

The authors would like to thank John Warren of the Prometheus Advanced Systems and Technology Office

for support of this work. Additionally, we would like to thank Daniel Parcher at JPL and Jim Gilland at the Ohio
Aerospace Institute (OAI) for their support on system modeling and trajectory analysis.

VII. REFERENCES

Warren, John, “Advanced Electric Propulsion (AEP) Technologies,” NASA Office of Space Science, Accessed

from http://research.hq.nasa.gov/code_s/nra/current/nra-03-oss-01/AppendA4_4.html, 23 January 2004.

Tikhonov, V., Semenikhin, S., Alexandrov, V., Dyakonov, G., and Popov, G., “Research of Plasma Acceleration

Processes in Self-field and Applied Magnetic Field Thrusters,” IEPC Paper IEPC-93-076, Presented at the 23rd
International Electric Propulsion Conference, Seattle WA, 1993.

Popov, A., and Tikhonov, V., “Theoretical and Experimental Research on Magnetoplasmadyanamic Thrusters,”

Russian Conference on Electric Propulsion

, No. RGC-EP-92-16, MAI, Moscow, May 1992.

Popov, A., Kim, V., Tikhonov, V., Semenikhin, S., and Tibrina, M., “High-Power Lithium-fed MPDT

Research,” The Fourth (Final) Quarterly Report on Contract No. 960938 between RIAME and JPL, Technical
Report, RIAME MAI, Moscow, December 1998.

Parcher, Daniel, NASA JPL, Personal Communication, 2005.

Sauer, Jr., Carl, NASA JPL, Personal Communication, 2002.

Mason, Lee, NASA GRC, Personal Communication, 2003.

Noca, Muriel, NASA JPL, Personal Communication, 2003.

Frisbee, R.H., Das, R.S.L., and Krauthamer, S., “Power Processing Units for High Powered Nuclear Electric

Propulsion with MPD Thrusters,” Presented at the AIAA/DGLR/AIDAA/JSASS 23rd International Electric
Propulsion Conference, Seattle WA, 13-16 September, 1993; and Krauthamer, S., et al., "Power Processing Units

American Institute of Aeronautics and Astronautics

for High Powered Solar Electric Propulsion Using MPD Thrusters," Presented at the 28th IECEC, Atlanta GA, 8-
13 August, 1993.

Frisbee, R.H., and Hoffman, N.J., "Near-Term Electric Propulsion Options for Mars Cargo Missions," AIAA

Paper AIAA 96-3173, Presented at the 32nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference, Lake Buena
Vista FL, 1-3 July, 1996.

Fiehler, D., et al., “Electric Propulsion System Modeling for the Proposed Prometheus 1 Mission,” AIAA Paper

AIAA 2005-3891, Presented at the 41st AIAA/ASME/SAE/ASEE Joint Propulsion Conference, Tucson AZ, 10-
13 July, 2005.

Gani, Ganapathi, NASA JPL, Personal Communication, 2004.

Miyake, Robert, NASA JPL, Personal Communication, 2004.

Kristalinski, Alexander, Aerojet, Redmond WA, Personal Communication, 2004.

Lewis, Joseph, NASA JPL, Personal Communication, 2004.

Frisbee, R.H., Leifer, S.D., and Shah, S.S, “Nuclear Safe Orbit Basing Considerations,” AIAA Paper AIAA-91-

3411, Presented at the AIAA/NASA/OAI Conference on Advanced SEI Technologies, Cleveland OH, 4-6
September, 1991.

Drake, B.D., Editor, “Reference Mission Version 3.0, Addendum to the Human Exploration of Mars: The

Reference Mission of the mars Exploration Study Team,” NASA Report EX13-98-036, June 1998.