FLIGHT DEMONSTRATION OF FEEP ON GET AWAY SPECIAL

AIAA 98-3332

Flight Demonstration of FEEPon Get Away Special

S. Marcuccio, L. Paita, M. Saviozzi, and M. AndrenucciCentrospazio, Pisa, Italy

33rd AIAA/ASME/SAE/ASEEJoint Propulsion Conference and Exhibit

July 13-15, 1998 / Cleveland, OH

For permission to copy or republish, contact the American Institute of Aeronautics and Astronautics1801 Alexander Bell Drive, Suite 500, Reston, VA 22091

AIAA-98-3332

2American Institute of Aeronautics and Astronautics

FLIGHT DEMONSTRATION OF FEEPON GET AWAY SPECIAL

Salvo Marcuccio*, Luca Paita†, Marco Saviozzi†, and Mariano Andrenucci‡

Centrospazio, Via A. Gherardesca 5, 56014 Ospedaletto, Pisa, ItalyTel. +39 050 985072 - Fax +39 050 974094

* Program Manager. Member, AIAA. E-mail: [email protected]† Research Engineer.‡ Director, Centrospazio. Professor, Department of AerospaceEngineering, University of Pisa. Member, AIAA.Copyright © 1998 by Centrospazio/Consorzio Pisa Ricerche.Published by the American Institute of Aeronautics and Astronautics,Inc., with permission.

Abstract

Primarily intended for verification of the thruster designand assessment of actual performance, the FEEP thrusterflight demonstration (EMITS - Electrical MIcrothrusterTest in Space) is aimed at building confidence in this newtechnology among the potential commercial and scien-tific users. EMITS will exploit an existing ESA-NASAagreement for the utilization of the Get Away Special(GAS) facility to fly a package including two thrusters ofdifferent thrust levels (30 µN and 1 mN), the thrusterpower and control electronics unit, a computerized ex-periment control and data storage unit, and a battery. Themain goal of the experiment is the verification of in-orbition beam production, neutralization and throttling. Othergoals include ion beam shape investigation and monitor-ing of propellant deposition on the nearby surfaces. Thispaper presents the EMITS configuration and goals andoutlines the experiment main features.

Introduction

FEEP is a promising electric engine concept that offerssound advantages upon traditional propulsion systems fora wide range of missions. The application field of thisthruster includes several commercial applications such assmall satellite attitude and orbit control1, along with drag-free control and fine pointing of scientific spacecraft2. Thelatter require highly controllable thrust in the 1 to 100 µN

thrust range, while 0.5 to 2 mN thrust is adequate for per-forming the former. In this thrust range, the high specificimpulse, high power-to-thrust ratio characteristics of FEEPmay be exploited best. Therefore, FEEP thruster devel-opment is presently focused on these two thrust ranges,with thrusters of two different sizes (2 mm and 70 mmslit length, respectively) being manufactured and tested.

FEEP is currently baselined for LISA3 (Laser Inter-ferometer Spaceborne Antenna, low-frequency gravita-tional wave detector, a Cornerstone mission in ESA’s Ho-rizon 2000+ programme), OMEGA4 (Orbiting MediumExplorer for Gravitational Astrophysics, a JPL proposalfor the NASA MIDEX programme) and GG-GalileoGalilei5 (a small satellite for testing the equivalence prin-ciple, under pre-phase A study at ASI), and is presentlyconsidered for several other missions (e.g., ESA’s Dar-win and GAIA, etc.). As for the commercial applications,many of the future LEO telecommunication constellationsare based on 200 to 1000 kg mass satellites, at orbitalaltitudes of 700 km or higher. For this category of mis-sions, the use of FEEP for such tasks as attitude controland fine pointing, drag makeup and orbit maintenance mayresult in substantial mass savings and performance en-hancement when compared to traditional chemical or coldgas thrusters.

Ground tests of FEEP thrusters have been extensivelycarried out for years on different thruster configurationsin several European laboratories. Although thruster per-formance has been thoroughly characterized6, it is clearthat there are some aspects of the FEEP technology thatcannot be validated with sufficient reliability other thanin space. A proper propellant feeding of the thruster bycapillarity in microgravity conditions, the absence ofemitter slit clogging due to possible formation of propel-lant compounds with residual atomic oxygen, any possi-ble induced contamination from the thruster on the space-craft surfaces, are only the most important of the issuesthat need to be investigated by means of a FEEP system

AIAA-98-3332


flight test. A flight demonstration of FEEP is the requiredstep towards the operational use of this technology. Byassessing the proper operation of the thruster in space,the flight test will provide the confidence in this technol-ogy that is needed to open the way to commercial appli-cations, as well as to long term, ambitious scientific ap-plications.

With this goal in mind, ESA has recently awarded acontract to Centrospazio, aimed at flying a FEEP flightdemonstration in a Get Away Special (GAS) canister inthe Space Shuttle bay. This experiment, called EMITS(Electrical MIcrothruster Test in Space), will be carriedout using a canister equipped with a Motorized Door As-sembly (MDA). The GAS-MDA hardware, as well as theflight opportunity, are provided by NASA through an ex-isting agreement with ESA. The FEEP thrusters and theassociated diagnostics are developed by Centrospazio,while the thruster power conditioning and control elec-tronics unit is provided by LABEN (Milan, Italy). Theexperiment controller electronics is provided by TechnoSystem Developments (Naples, Italy).

EMITS is aimed at firing the FEEP thruster in spacefor the first time. Although of short duration (2 to 3 days),this experiment will provide valuable information on thethruster behaviour. In particular, the capability of operat-ing FEEP in less-than-ideal space environment will beverified, as the Shuttle bay environment (high-pressure,water vapour- and oxygen-rich atmosphere) is one of themost demanding a field emission thruster may operateinto. All operational features of FEEP will be checked,including propellant feeding by capillarity, beam neutrali-zation, thrust throttling and repeated restart capability. Thediagnostics will include ion beam electrostatic probes andquartz crystal microbalances for propellant depositionmonitoring. Full electrical characterization of the thrusterwill be performed, and thrust will be evaluated from therecorded electrical parameters.

The experiment will host two different thruster assem-blies, each including its own sealed container, heater, tem-perature sensor, emitter and accelerator. In order to coverthe thrust ranges of interest of both commercial and sci-entific applications of FEEP, slit lengths of 2 mm and 70mm were selected, providing nominal thrust levels of 30µm and 1 mN, respectively. Two thermionic neutralizerswill be installed, each close to one of the emitters, allow-ing for cross-neutralization tests.

EMITS has been given the G-752 Get Away Specialpayload identification number by NASA. While the tar-get launch date is September 1999, the actual STS flightnumber depends upon the Shuttle schedule, which is of-ten subject to much re-arrangement during the pre-launchmonths. Due to the unpredictability of the actual launchdate for secondary payloads such as GAS canisters, theexperiment had to be designed in such a way as to be able

to be stored for several months in a dormant state.

The Get Away Special Facility

The Small Self-Contained Payload (SSCP) Program,popularly known as the Get Away Special (GAS) pro-gramme, was initiated in the mid seventies to provide ex-tremely low cost access to space to a diverse user com-munity. It provides an opportunity for individuals andorganizations, both public and private, of all countries tofly a small experimental payload on a Space Shuttle mis-sion for a nominal fee. GAS payloads must contain scien-tific or engineering oriented experiments, although NASAdoes not judge the scientific merit of the payloads. Theymay not be used for profit or for commemorative pur-poses. These payloads are flown in NASA provided con-tainers following a Flight Certification process which in-cludes both a Launch Agreement and a Safety Reviewthat lasts about one year, depending on payload complex-ity.

The Space Shuttle has a capacity for approximately40000 lb (18200 kg) of payloads carried in theunpressurized payload bay. The payload bay is 60 feet(18.3 m) in length and 15 feet (4.6 m) in diameter, and isdesigned to carry up to four large payloads mounted bymeans of sliding “trunion” bearings which engage spe-cial fittings on the sides and bottom of the payload bay.This arrangement is appropriate for payloads weighingbetween about 1800 kg and 18200 kg. NASA has madeprovisions for accommodating smaller payloads in the 50lb (23 kg) to 4000 lb (2270 kg) range by means of "carri-ers" which provide mechanical and electrical interfacessuitable for small payloads.

NASA maintains two carrier programs for accommo-dating small attached payloads in the Space Shuttle pay-load bay. These are the Small Self Contained Payload(SSCP) program more commonly known as Get-Away-Special (GAS), and the Hitchhiker program. GAS andHitchhiker are developed and operated by the GoddardSpace Flight Center Shuttle Small Payloads Project (SSPP)for the NASA Office of Space Flight. The GAS carrier(Fig. 1) provides standard, very simple, mechanical andelectrical interfaces for self-contained experiments. Sim-ple crew control functions may be performed but power(battery), data recording and sequencing systems, ifneeded, are provided by the user. GAS payloads are lim-ited in weight and volume ranging from two and one-halfcubic feet, 60 pounds to five cubic feet, 200 pounds. Sincethe program became operational in 1982, about 150 GASpayloads have been flown. They are flown on a first come,first served, space available basis. The Hitchhiker pro-gram is intended for customers whose space activity re-quires power, data, or command services. The Hitchhiker

AIAA-98-3332


system provides for real-time communications between acustomer in the Hitchhiker control center at Goddard andhis payload and can also provide crew control/displaycapability if necessary. Mechanical and electrical inter-faces and integration procedures are more complex thanthose used for GAS. Hitchhiker was not considered forthis study, as it is not included in the ESA-NASA existingagreement for small payloads.

GAS payloads are carried in standard canisters whichcan be mounted in a large number of locations on the sideof the Shuttle payload bay. Up to twelve canisters canalso be mounted on the cross-bay “GAS Bridge” carrierif required by the mission configuration. The GAS car-rier consists of a canister which can accommodate cus-tomer equipment in a volume 19.75 inches (501.6 mm) indiameter and 28 inches (711.2 mm) high. The customerequipment, which can weigh up to 200 lb (90.7 kg), at-taches to the canister top plate and connects by means ofcustomer supplied cables to the GAS electrical interfaceon the canister bottom plate. The user must provide thestructure which supports his experiment hardware andwhich will be cantilevered to the experiment mountingplate. At least three equally spaced bumpers are to be in-stalled at the free end of the experiment structure to sup-port radial loads between the structure and the internalsurface of the GAS container.

Each canister is equipped with an electrical systemwhich has a 25 A power relay and two, 2 A signal relaysfor controlling customer equipment. The canisters con-nect to a common signal line in the Orbiter which con-nects them to a crew controller in the cabin. The crew canindividually address relays in up to 25 canisters to turn

the relays on or off or to determine the state of a relay.The relays can also be simultaneously reset by a masterreset command. A baroswitch can optionally additionallybe used to turn the power relay on and off on ascent anddescent.The Experiment Mounting Plate (EMP) serves the fol-lowing three purposes:

• seals the upper end of the standard GAS container (notthe case for openable canisters);• provides a mounting surface for the experimentalequipment;• can act as a thermal absorption or radiation surface.

The inner surface of the plate has a hole pattern adaptableto mounting a variety of hardware. Forty-five stainlesssteel, internally threaded inserts are available for mount-ing. The experimenter may use any of them in any com-bination required. The same experiment mounting plateis used for all payload sizes.

The GAS can will be purged with dry nitrogen. Twopurge ports are located on the experiment mounting plate,and at least one of these must be unobstructed to allowpurge gas flow through the GAS can. The line from thecenter of the plate through the two purge ports will al-ways be aimed out the starboard (right) side of the Orbiter,perpendicular to the Orbiter centerline.

The Motorized Door Assembly (MDA) is an optionof the standard Get Away Special which allows for theexposure of the GAS payload directly to the space envi-ronment. A closed GAS-MDA canister provides thermalisolation to the payload from the Orbiter bay and the space

Fig. 1 - The GAS Facility; left, container concept; right, a GAS can mounted in the Shuttle bay.

AIAA-98-3332


environment. Moreover, it can be vented to 1 psi differ-ential or it is able to maintain vacuum conditions. TheMDA door opens and closes in about 35 seconds with anopening angle of 111°. The MDA assembly mass is 18kg. Pressure relief valves vent the canister atmosphereduring ascent to less than 2 psid prior to door opening bycrew command.

The MDA Experiment Mounting Plate (MDAEMP)is basically a standard GAS experiment mounting platewith has a 15.375 in diameter opening located on itscenterline. The mounting of the experiment is accommo-dated by 24 holes equally spaced on a 19 inch diameter.Some of them are to be left unobstructed for battery vent-ing (if required) and gas purge purposes. Each GAS-MDAcontainer has two payload control units supporting somebasic needs of the user and controlling the MDA. Essen-tially there are six latching relay contacts which are equiva-lent to toggle switches, three available for user and threeavailable to NASA. The state of these relays can bechanged remotely by the astronauts at predetermined timesor flight conditions. They are rated at 2 A, but a supple-mental unit, the Payload Power Contactor (PPC), providesthe user with two additional relays rated at 25 A. Thesecontrols simply allow to change the operational mode ofthe payload (to turn power on and off, initiate a specialsequence, change data rate and to open or close the MDA).The PPC also provides self-control of payload power viathe “malfunction inputs”. If a malfunction arise in the ex-periment, PPC can remove the payload power and reap-ply it when the malfunction condition disappears. If de-sired, MDA operations can be controlled by the GAS usersupplying 12 V DC at 20 mA to the MDA through a switch.The payload control units are located at the bottom of thecanister and electrical connections to the payload are viathe NASA Interface Equipment Plate (IEP).

Safety requirements, which are specified in a dedi-cated NASA document, are carefully reviewed during theexperiment acceptance phase and constitute one of themain design drivers. Obviously, safety requirements for aGAS-MDA payload are more stringent than for a con-ventional GAS canister payload. In particular, the pay-load structure must undergo a Fracture Mechanics Analy-sis according to NASA GSFC Fracture Control Plan, andthe user has to demonstrate that if any accident occursduring the experiment, the door of the canister will maybe closed before the re-entry phase. Outgassing of thewhole experiment must be kept to a minimum to avoidcontamination of other payload or the Orbiter. Since theMDA option violates the EMI shielding integrity of theclosed GAS container, the user has to design his experi-ment so as all forms of electromagnetic radiation are re-duced at the minimum.

The EMITS Experiment

No FEEP slit emitter has been fired in space yet. There-fore, the main objective of the proposed experiment is tovalidate the FEEP technology in space by checking thecorrect, in-orbit operation of a complete propulsion sys-tem, including the thruster and all of its subsystems.

The experiment goals include primary goals and sec-ondary goals. Primary goals are:

• firing the thruster in space for the first time;• demonstrating the propellant storage and container

sealing functionality;• demonstrating proper operation in microgravity of

propellant feeding by capillarity;• demonstrating ion beam neutralization;• testing the power conditioning unit in an operational

environment.

Secondary goals are:

• assessing the plume current distribution;• evaluating the propellant backflow and deposition on

surrounding surfaces;• checking the evolution of thruster performance with

time.

Thrust measurement is not within the scope of this ex-periment. Direct measurement of micronewton level thrustcan be achieved only by means of sophisticated, expen-sive and probably bulky balances or by sensing the timecumulated effect of the very low thrust on the dynamicsof an autonomous spacecraft, via body-mounted acceler-ometers or via precision orbit tracking. None of this sys-tems is viable on GAS. Nevertheless, the thruster perform-ance will be evaluated by means of the electric param-eters recorded through the test; in particular, the actualthrust delivered and the thruster specific impulse will beindirectly evaluated using well known formulas, alreadyfully proven by a long record of earth-based tests.

Evidence of the ion beam presence will be gained bymeans of electrical measurements in the thruster’s powersupply lines and, independently, by means of ion beamelectrostatic probes reading.

Thruster tests will include investigation of:

• emitter threshold voltage;• steady state operations at different emitter voltages;• switch on/off capabilities;• throttleability over the entire characteristic curve.

Experiment configuration

As the GAS canister interface to the external world is lim-

AIAA-98-3332


ited to a small number of telecommands, all of the neces-sary experiment hardware and support equipment mustbe carried within the canister. EMITS will therefore in-clude the following subsystems:

• the battery pack;• the GAS Electronics Unit (GEE), including the raw

power conditioning system, drivers for the diagnos-tics, the data acquisition and storage system, and theexperiment controller;

• the thruster power and control electronics (FEEP Elec-tronics Unit - FEU);

• two thruster assemblies of different size;• two neutralizers;• the experiment diagnostics, including a UHV pressure

probe, a set of movable electrostatic probes and thepropellant deposition monitors.

A block diagram of the experiment, schematically show-ing the interconnections between the main subsystems andbetween the GAS canister and the Shuttle interface, isshown in Fig. 2 (the external GEE EGSE - ElectricalGround Support Equipment is not part of the flight hard-ware).

The GAS will be equipped with an internal experi-ment support structure, consisting of a simple aluminum

frame with horizontal mount plates (Fig. 3). The experi-ment will host two thruster assemblies, each including itsown sealed container, heater, temperature sensor, emitterand accelerator. In order to cover the thrust ranges of in-terest of both commercial and scientific applications ofFEEP, the following two emitter slit lengths have beenselected:

• 2 mm slit length: nominal thrust = 30 µN, maximumpower consumption = 2 W;

• 70 mm slit length: nominal thrust = 1 mN, maximumpower consumption = 60 W.

All emitters will have a slit height of 1.5 µm. The liquidmetal used as propellant will be stored in the emitter in-ternal reservoir. Propellant feeding to the emitter tips willbe performed by capillarity. The emitters will be loadedwith about 2 grams of propellant each.

Traditionally, FEEP has been operated with cesiumpropellant. Recently, the use of rubidium has been pro-posed and emitter performance with this propellant hasbeen experimentally assessed. While both metals havesimilar properties with respect to handling and operationalbehaviour, rubidium has the advantage of a higher melt-ing point, 39 °C vs. 29 °C of cesium. This may ease theon-ground thermal conditioning procedures needed to

Gas Experiment Electronics(GEE)

FEEP Electronics Unit(FEU)

Battery Box

NASA Interface

Shuttle CrewTelecommand I/F

GEE EGSE PC

GEE EGSEExternal

Power supply

Pres

sure

an

dte

mpe

ratu

res

QCM

sig

nals

and

com

man

ds

Wire

pro

besi

gnal

s an

d co

mm

ands

RS422/1553BData/commands

Power line

Tes

t lin

es

RS2

32

Dat

a/Com

man

ds

FEEP DiagnosticsEmitter/container/neutralizer

assemblies

HV

Lin

es(e

mit

ters

, ac

cele

rato

rs)

LV L

ines

(neu

tral

izer

s,

heat

ers)

Tem

pera

ture

si

gnal

s

Cov

er r

elea

se c

omm

ands

and

mon

itor

s

GAS Canister

Fig. 2 - EMITS block diagram.

AIAA-98-3332


store the propellant in the solid phase, while presentingno significant drawbacks for in-orbit operation. On theother hand, cesium has a higher atomic mass, resulting inabout 20 % better power efficiency. Therefore, rubidiumseems best suited for low power applications of FEEP.On EMITS, cesium will therefore be used for the 1 mNthruster, and rubidium will be used for the 30 µN one.

Emitter ContainerEach emitter will be located into a special container pro-viding a sealed envelope (Fig. 4), filled with inert gas,which protects the emitter from interaction with air, toavoid slit clogging due to formation of cesium (or rubid-ium) oxide or hydroxide from the moment of propellantfilling to experiment start in orbit. Propellant is introducedin the emitter internal reservoir via a capillary connectedto the rear part of the emitter. The capillary is connectedto a funnel with a Swagelok connector; propellant drop-lets flow into the funnel from the propellant feeding sys-tem mounted on the support equipment. When the pre-paratory activities in the vacuum chamber are completed

and the system is returned to atmospheric pressure, thefunnel is disconnected, exposing the residual liquid metalpresent in the capillary to air. In this way, a short plug ofoxide is formed at the end of the capillary, which resultsin effective isolation of the emitter internal reservoir. Aclosed Swagelok cap is placed on the fitting, to ensuredefinitive sealing.

The envisaged sequence of events, prior to thrusterfiring, is as follows:

1) In the vacuum chamber:• emitter internal reservoir filling with cesium or rubid-

ium propellant;• emitter firing for at least 2 hours; recording of I/V

curves and beam profiles, to check proper wetting ofthe emitter tips and verify correct performance of heat-ers and isolators;

• container filling with inert gas and lid sealing.2) In the laboratory:• container dismount from the chamber, closing of the

capillary;

Accelerator

Container

Openable lid

Isolator

Mounting flange

Propellant filling duct

Heater

Emitter

Latch

Top plate

Thruster electronics (FEU)

Experiment controller (GEE)

Battery

Electronics plate

2 mm thruster70 mm thruster

Neutralizers

Ion beam probe QCM

Pressure probe

Fig. 3 - EMITS Configuration: left, overall view; right, top plate close-up.

Fig. 4 - Emitter container: left, schematics; right, prototype during a thermal vacuum test.

AIAA-98-3332


• integration with the GAS experiment and shipping.3) In orbit:• emitter heating at the operating temperature;• container lid opening;• start of the flight test.

Lid closing in the vacuum chamber is performed by meansof a dedicated device (Fig. 5), remotely controlled fromoutside the vacuum chamber. The lid closing sequence isshown in Fig. 6. This device was recently tested with suc-cess at Centrospazio. Lid opening will be performed bymeans of a simple, commercially available paraffin ac-tuator, provided by M3D (Carouge-Genéve, Switzerland).This actuator is a scaled-down version of a space-quali-fied device7 developed for another ESA project. Actua-tion is initiated by supplying electrical power to heat theparaffin at about 80 to 100 °C. The total actuation time isabout 4 minutes for a pin extraction of several millime-tres.

NeutralizersThe neutralizer must prevent negative charge accumula-tion due to the emission of ions; one neutralizer is suffi-cient for all thrusters. Two different types of neutralizerhave been considered: the field emission electron sourceand the thermionic neutralizer. The field emission neu-tralizer presents two main advantages respect to thethermionic one:

• the source does not require high temperature;• the source unit is intrinsically redundant, being formed

by a large number of emitting microtips.

On the other hand, the field emission electron sources arenot yet fully developed. In particular, space operation ofthese units is not reported, and the mating of such sourcesand FEEP emitters must be further investigated. There-fore, the experiment will mount simple thermionic neu-tralizers, based on the BaCO3 technology. Power con-sumption per unit extracted current for these devices is inthe order of 0.16 to 0.33 W/mA; this is acceptable to matchthe 6 W/mA of the FEEP emitter.

Experiment ControllerThe thrusters electrical supply (emitters and acceleratorsHV, neutralizer voltages) and the thruster thermal controlwill be performed by the FEEP Electronics Unit (FEU).The FEU will interface with the thrusters, from one side,and with the GAS Experiment Electronics (GEE) fromthe other side. The GEE is a microcontroller-based unitwhose tasks are:• to drive the FEEP subsystems and diagnostics through

a pre-stored sequence of events;• to acquire and store the experiment data in a non-vola-

tile memory bank for post-flight retrieval.

To this end, the GEE has to:

1) store the experiment time-line, i.e. the sequence of eventto be performed during the entire experiment duration,from experiment activation to shutdown. Typical eventsinclude FEU commands, data acquisition operations, bat-tery checks, etc. Execute the stored sequence of events inorder to perform activities 2. and 3. below. Monitor theprogress of the experiment and perform alternative, re-covery actions in case of detection of malfunctions or fail-ures.2) Provide commands to, and accept telemetry from, theFEU. The FEU will be equipped with a MIL-STD 1553/B interface; however, an RS 422 line is presently consid-ered as a possible option. As an example, commands in-clude container lid opening command, switching on / offthrusters, setting the emission level and varying it in time,etc.3. Drive the FEEP Diagnostics. The GEE shall provideall the necessary electric power supply lines to feed theFEEP diagnostics, condition and acquire the output data,with appropriate D/A converters, at a proper rate and store

Fig. 5 - Lid closing device.

Fig. 6 - Closing sequence.

AIAA-98-3332


them for post-flight analysis.4. Interface with the EGSE during pre- and post-flightactivities.

As an input, the GEE will receive a telecommand fromShuttle crew to initiate the experiment, and will be readyto accept an emergency shut-off command. Standard elec-trical and mechanical requirements for electronics sub-systems to be flown as a part of a GAS payload shall bemet. The onboard software will include recovery routinesto be performed in case of non-nominal operation or fail-ures; the recovery actions will range from experiment pros-ecution with a reduced scope, to partial switch off, to to-tal abortion. Shuttle safety issues will be the primary driv-ers for experiment design.

Temperature of most of the experiment componentsmust be monitored. The temperature of emitters, neutral-izers, battery pack, electronics, stepper motors will besensed by appropriate gauges. The microcontroller willperiodically check the data which are stored in the massmemory for post-flight analysis. Should any of the pa-rameters exceed the allowed values, the microcontrollerwill execute emergency sequences.

A total pressure gauge will be provided in order tomonitor the ambient pressure conditions after the MDAopening and during GAS interior outgassing. When suit-able vacuum conditions are reached, the microcontrollerwill start the procedure for thruster ignition. Pressure willbe measured until the end of the experiment.

The experiment will be attempted in any case, even ifnon-ideal pressure conditions are indicated (due to highambient pressure or to gauge malfunctioning), in ordernot to lose the flight opportunity. In this case, a proper,alternative sequence of operations will be followed, in-cluding prolonged waiting time before thruster lid release,reduced switch-off periods, and limited high emissioncurrent operation.

The GEE must be able to store all the experiment pa-rameters measured throughout the test. The followingparameters will be recorded via a set of multiplexed sig-nal acquisition lines and ADC converters:

• emitter voltage (10 data/min)• accelerator voltage (10 data/min)• emitter current (10 data/min)• accelerator current (10 data/min)• neutralizer voltage (10 data/min)• ion probe measurements (150 data per scanning)• emitter temperature (8 data/min)• neutralizer temperature (8 data/min)• FEU temperature (2 data/min)• battery temperature (2 data/min)• ambient pressure (2 data/min)• contamination sensors (2 data/min)

The thruster electrical parameters (in particular, those atHV) and temperature will be provided to the GEE throughthe built-in monitoring functions of the FEU. As the com-plete test run will last about 24 hours, a total measure-ment data quantity of about 120000 single values is to bestored. This may be easily accomplished with a standardbank of 2 MB EEPROMs.

DiagnosticsThe diagnostics package, whose purpose is to monitor theperformance of the FEEP system and its interactions withthe surrounding environment, will include, as a minimum,the following items:

- a set of electrostatic probes. The probes are tungstenwires immersed in the FEEP ion beam in several loca-tions, and kept at ground potential with respect to thethruster electrodes. The current drained by the probes,usually in the order of a few microamps, gives qualitativeindications of the shape and current distribution withinthe ion beam. The probes will be designed and manufac-tured at Centrospazio using the usual arrangement exten-sively used in past FEEP experiments. Trade-off betweenmotorized probes and fixed wires will be performed dur-ing the early experiment definition phase. The motorizedprobe, consisting in a tungsten filament, is mounted onan arm which rotates with the shaft of a stepper motor, sothat the probe can scan the beam few centimetres down-stream the accelerator. The motors are driven by dedi-cated power electronics, located in the GEE. Hardwareend travel sensors will be provided. Standard motors likethe Caburn B14.1 Miniature UHV Stepper Motors, withangular resolution of 0.9 deg, can be employed. The ionprobes will be operated in a sequence such to acquire firstinformation on the divergence in the plane perpendicularto the slit, which is known to be the most important con-tribution to thrust loss;- a cold-cathode pressure gauge, to monitor the environ-ment pressure in which the FEEP system operates;- a set of temperature sensors, to monitor temperatures inseveral parts of the experimental assembly. Standard plati-num thermoresistors will be used;- a set of Quartz Crystal Microbalances (QCMs), to moni-tor the rate of deposition of backflowing propellant onthe surfaces surrounding the thrusters. These devices canmeasure deposits from less than 10-9 g cm-2 up to severalhundreds of mg cm-2. Theoretical evaluation of the amountof cesium that evaporates from the slit surface duringthruster switch-off periods is in the order of 10-7 g s cm2.Several QCMs will be placed in various locations on theexperiment assembly. A specialized vendor (QCM Re-search of Laguna Beach, CA) will provide flight-quali-fied QCMs and technical assistance.

AIAA-98-3332


Thruster Power and Control ElectronicsThe FEEP Electronics Unit (FEU) is basically a set ofpower supplies (both low and high voltage) with a digitalinterface, and a control unit driven by a microcontroller.It will perform the following tasks:

• high voltage supply to the emitter and the accelerator;• low voltage supply to the neutralizer;• thermal monitoring and control of the thruster compo-

nents;• sensor signal conditioning and interface with the GEE.

Since the power supply provides +28 V DC regulatedpower on one power line, voltages other than this valuewill be produced internally by the FEU. The FEU has in-ternal DC/DC converters used to decouple the experimentpower source.

BatteryThe battery pack is the most critical of the experimentsubsystems. It poses several design challenges due to thesafety requirements associated to its use on a GAScontainer. The battery has to provide 28 V on two separatestrings to power the GEE and the FEU. A special, sealedcontainer for a battery pack already flown on a previousGAS experiment is available at ESTEC and may be reusedafter refurbishment. Re-flight of this unit has been assumedas a baseline. This container is able to accommodate atotal of 76 Ag-Zn cells with a total energy storage capac-ity of about 1.8 kWh. This quantity of energy is sufficientfor a total experiment duration of about 2 days. Anenhanced battery design, using larger cells, is presentlybeing considered. This option would permit the experimentto last for about 3 days, at the cost of a moderate massincrease. A trade-off study between the increasedcapability and the relevant design modifications isunderway.

EMITS Design Constraints

The GAS facility provides a very special environment forspace experiments, and quite an unusual one for a thruster.The necessity to avoid ion beam impingement on the GAS,as well as the requirement of allowing MDA closing re-gardless of the ion beam probe arms position, have dic-tated the choice of the geometrical arrangement of thethrusters and the diagnostics on the experiment top plate.Fig. 7 shows the ion beam emerging from the GAS in anextreme beam divergency case. The MDA is shown inboth the open and the closed position.

The EMITS experiment design has been heavily in-fluenced by the peculiar environmental constraints posedby the GAS-MDA canister. These constraints are reviewed

in the following sections.

GAS Thermal EnvironmentDuring normal thruster operation, the propellant must bein the liquid phase. This implies that the emitter tempera-ture must be not lower than 29 °C or 39 °C for cesiumand rubidium, respectively. On the other hand, if the tem-perature is too high propellant evaporation may causesparking between the electrodes and may induce contami-nation. Therefore, emitter temperatures of 35 ± 5 °C and45 ± 5 °C for cesium and rubidium, respectively, havebeen assumed as design points for in-orbit operation.

A careful design of the emitter thermal control systemis necessary in order to avoid exposure of the emitter tothe demanding thermal conditions experienced in the Shut-tle cargo bay. The in-orbit temperature of a GAS payloadstrongly depends on the following factors:

• the angle between the solar vector and the orbit plane(beta angle), which determines the eclipse duration;

• the Shuttle attitude relative to the earth and the sun,which determines the amount of sunlight or earth ra-diation entering the orbiter bay;

• the position of the payload within the orbiter and rela-tive to other payloads.

None of these factors is known in advance, nor can theybe controlled by the GAS user. If a GAS payload needsdirect exposure to the space, as the FEEP experiment, thethermal environment is more complex and dynamic thanin a conventional GAS container (without MDA).

The nominal orientation of the Shuttle during GASexperiment operation is payload bay to the Earth, and the

70 cm thruster

254

213

390

25°

Fig. 7 - Ion beam geometry (dimensions in mm).

AIAA-98-3332


beta angle is equal to 33 deg. In this condition, the ther-mal environment is considered benign: when the cover isclosed, steady state temperature of the GAS payload isabout -10 °C.

Since the prelaunch phase lasts an average 3 months,the risk that payload temperature reaches the highest val-ues during this period must be evaluated. During the phasein the assembly building (for installation of GAS bridgein Shuttle and Shuttle preparation) the temperature is con-trolled at around 20 °C. After installation on the Shuttle,the payload cargo bay is purged with clean air at 20 °C.

Shuttle Bay Gaseous EnvironmentPressure and composition of the background atmospheremay strongly influence the FEEP thruster performance.Start-up is the most critical phase of thruster operation,since emitter ignition depends on the wetting of the innersurface of the emitter blades by the propellant. Properwetting can be seriously jeopardized by the presence ofwater or other impurities absorbed on the emitter surfaces,that may lead to the clogging of the slit due the formationof a layer of cesium oxide. Although best wetting is ob-tained when pressure is in the range of 10-9 mbar, labora-tory tests have shown that the thruster can be ignited at apressure as high as 10-6 mbar, provided that the partialpressure of water be lower than about 10-7 mbar. The lackof these conditions may result in an initial emission sitesdistribution highly non- homogeneous, thus yielding poorthruster performance. The initial inhomogeneity mayeventually smooth out slowly with time.

The flight experiment will be performed with thrust-ers filled with propellant and fired for acceptance onground, in an optimal ultra-high vacuum. This will elimi-nate any danger of improper wetting of the slit which couldoccur as a consequence of attempts to fill the emitter inspace.

In-flight pressure requirements are less stringent. Oncecorrectly prepared, FEEP emitters are tolerant with re-spect to background pressure increase8. Residual atmos-phere in orbit is made up of two contributions: local am-bient pressure and contaminants. At the Shuttle orbit alti-tude, the background pressure ranges from 10-8 to 10-7

mbar. Atomic oxygen is present with a partial pressureranging from 10-9 to 10-7 mbar, while water vapour pres-sure varies from 10-8 to 10-7 mbar. The largest importantcontaminant is water vapour and its quantity seems to bedirectly controlled by temperature-induced outgassing ofspacecraft surfaces. Events like thruster burns and waterdumps cause the H2O concentration to increase; for in-stance, Shuttle attitude and maneuver engines burns seemproduce pressure pulses of 1.10-6 to 2.10-6 mbar risingimmediately with engine initiation and falling to back-ground level within a few seconds of engines cutoff. Otherimportant contaminants are He, H2 and N2.

Therefore, pressure conditions seem not be very fa-vourable to FEEP experiment. However, high pressureexperiment made on ground provide enough confidencein the thruster operation even in that unfavourable envi-ronment. Emission tests have been successfully carriedout at a pressure as high as 10-4 mbar, with a water va-pour partial pressure in excess of 10-7 mbar.

It is well known that the Orbiter early desorption ratedecreases by approximately two order of magnitude afterthe first 40 hours in orbit. Thus, the severity of the pres-sure environment may be partially alleviated by lettingthe experiment start during the last days of the Shuttlemission. The emitter will be fired immediately after con-tainer lid opening, as it has been observed that contami-nation is prevented by ion emission. Finally, if the Shut-tle mission profile allows to do so, a favourable orbiterattitude will be preferred, avoiding direct exposure of theemitters to the ram direction, and emitter firing will betimed as to avoid engine burns.

Mass Budget and Experiment ProfileThe experiment mass budget is shown in Tab. 1.

Item Mass (kg)

Thruster No. 1 (1 mN)* 1.20Thruster No. 2 (30 µN)* 0.25Neutralizers (2 units) 0.20FEU 5.80GEE 4.50Heaters 0.05Motorized ion probes 0.50QCMs (5 pieces) 0.10Pressure probe 0.10Battery 30.40Structure and harness 12.00

TOTAL 55.10

(*) including propellant

Table 1 - EMITS Mass Budget

Since the maximum allowed mass of GAS payloads is 72kg (having taken into account 18 kg for the MDA assem-bly), there is a mass margin largely sufficient to accom-modate for possible design changes. In particular, it ispossible to consider flying a larger battery to extend theexperiment duration. In this case, the critical design issueis the constraint on the first vibration mode frequency,which must be not lower than 50 Hz. This constraint maybecome difficult to meet with a larger battery, and mayrequire significant rearrangement of the subsystem blockson the structure plates.

AIAA-98-3332


The envisaged experiment profile is as follows:

• experiment preparation at Centrospazio, including ac-ceptance firing of the emitters in the vacuum chamber,sealing, integration on the support structure, electricalchecks;

• hardware integration on GAS. Subsequent integrationof GAS on the Shuttle bay is performed by NASA per-sonnel;

• launch and orbit acquisition;• MDA opening via Shuttle crew telecommand;• experiment initiation via Shuttle crew telecommand.

From this moment on, the experiment is autonomouslyexecuted following a preset sequence of events withthe timeline stored in the onboard computer software.The following tests will be executed for each of theemitters:- switch on- fixed voltage operation, with ion beam scanning- variable voltage operation, I/V characteristic curverecording- prolonged switch-off, followed by switch on and rep-etition of the tests.

• autonomous experiment termination, followed by MDAclosing via Shuttle crew telecommand;

• return from orbit, GAS recovery, post-flight hardwareinspection, data analysis.

Conclusions

The EMITS experiment will be the first space test ever ofa field emission electric propulsion system. Developedby Centrospazio under ESA sponsorship, EMITS intendsto demonstrate the viability of FEEP for millinewton thrustlevel commercial missions, as well as for micronewtonthrust level scientific applications. Providing a unique lowcost, high reliability opportunity for access to space, theGet Away Special facility will play an important role inthis achievement. After more than two decades of basicresearch, FEEP will eventually enter the operational sce-nario and contribute to broaden the field of applicationsof electric propulsion.

Acknowledgments

The EMITS experiment is funded by the European SpaceAgency (Contract No. 12466/97/NL/PA). FiammettaFichi, Fabrizio Rugo and Lorenzo Serafini at(Centrospazio) participated in the EMITS study. The con-tributions of Giorgio Saccoccia and José Gonzalez delAmo (ESTEC), Dario Fossati (LABEN), DanieleTitomanlio and Giuseppe Capuano (Techno System) aregratefully acknowledged.

References

1. Marcuccio, S., Giannelli, S., Andrenucci, M., “Atti-tude and Orbit Control of Small Satellites and Constella-tions with FEEP Thrusters”, IEPC-97-188, Proceedingsof the 25th Electric Propulsion Conference, Cleveland,OH, 1997, pp. 1152-1159.

2. Klotz, H., Strauch, H., Wolfsberger, W., Marcuccio, S.,and Speake, C., “Drag-Free, Attitude and Orbit Controlfor LISA,” Proceedings of the ESA/ESTEC 3rd Interna-tional Symposium on Spacecraft Guidance, Navigationand Control, ESA SP-381, Noordwijk, The Netherlands,1996, pp. 695-702.

3. Bender, P., et al., “LISA - Laser Interferometer SpaceAntenna for the detection and observation of gravitationalwaves - Pre-Phase A Report”, 2nd edition, Max-Planck -Institut für Quantenoptik, MPQ 233, Garching, Germany,1998.

4. Hellings, R. W., “OMEGA - Orbiting Medium Explorerfor Gravitational Astrophysics,” MIDEX Proposal, JPL,California Institute of Technology, Pasadena, CA, 1995.

5. Nobili, A. M., et al., “Galileo Galilei - GG: Flight Ex-periment on the Equivalence Principle with Field Emis-sion Electric Propulsion”, Journal of the Astronautical Sci-ences, 43, 3, July-Sept. 1995, pp. 219-242.

6. Marcuccio, S., Genovese, A., Andrenucci, M., “Experi-mental Performance of Field Emission Microthrusters”,to be published in Journal of Propulsion and Power, Vol.14, No. 5, September-October 1998.

7. Mallet, O., Kornmann, M., Marirrodriga, C. G., “De-velopment of a Paraffin Actuator”, Proc. 7th EuropeanSpace Mechanisms and Tribology Symposium, ESA SP-410, Noordwijk, The Netherlands, 1997.

8. Genovese, A., Marcuccio, S., Dal Pozzo, D.,Andrenucci, M., “FEEP Thruster Performance at HighBackground Pressure”, IEPC-97-186, Proceedings of the25th Electric Propulsion Conference, Cleveland, OH,1997, pp. 1138-1144.

FLIGHT DEMONSTRATION OF FEEP ON GET AWAY SPECIAL

Documents