NASA/TM--1999-209436

/1/2

AIAA 99-2949

I ">i.]_-" i_l :" "U

/,v --_ o

High Power Hall Thrusters

Robert Jankovsky

Glenn Research Center, Cleveland, Ohio

Sergey Tverdokhlebov

TsNIIMASH Export, Korolev, Moscow, Russia

David Manzella

Dynacs Engineering Company, Inc., Brook Park, Ohio

December 1999

https://ntrs.nasa.gov/search.jsp?R=20000025489 2020-04-04T01:22:51+00:00Z

The NASA STI Program Office... in Profile

Since its founding, NASA has been dedicated tothe advancement of aeronautics and spacescience. The NASA Scientific and Technical

Information (STI) Program Office plays a key partin helping NASA maintain this important role.

The NASA STI Program Office is operated by

Langley Research Center, the Lead Center forNASA's scientific and technical information. The

NASA STI Program Office provides access to theNASA STI Database, the largest collection of

aeronautical and space science STI in the world.

The Program Office is also NASA's institutionalmechanism for disseminating the results of its

research and development activities. These resultsare published by NASA in the NASA STI Report

Series, which includes the following report types:

TECHNICAL PUBLICATION. Reports of

completed research or a major significantphase of research that present the results of

NASA programs and include extensive dataor theoretical analysis. Includes compilations

of significant scientific and technical data andinformation deemed to be of continuing

reference value. NASA's counterpart of peer-

reviewed formal professional papers buthas less stringent limitations on manuscript

length and extent of graphic presentations.

TECHNICAL MEMORANDUM. Scientific

and technical findings that are preliminary or

of specialized interest, e.g., quick releasereports, working papers, and bibliographiesthat contain minimal annotation. Does not

contain extensive analysis.

CONTRACTOR REPORT. Scientific and

technical findings by NASA-sponsored

contractors and grantees.

CONFERENCE PUBLICATION. Collected

papers from scientific and technical

conferences, symposia, seminars, or othermeetings sponsored or cosponsored byNASA.

SPECIAL PUBLICATION. Scientific,

technical, or historical information from

NASA programs, projects, and missions,

often concerned with subjects having

substantial public interest.

TECHNICAL TRANSLATION. English-language translations of foreign scientific

and technical material pertinent to NASA'smission.

Specialized services that complement the STIProgram Office's diverse offerings include

creating custom thesauri, building customizeddata bases, organizing and publishing research

results.., even providing videos.

For more information about the NASA STI

Program Office, see the following:

• Access the NASA STI Program Home Pageat http://www.sti.nasa.gov

• E-mail your question via the Internet tohelp@sti.nasa.gov

• Fax your question to the NASA Access

Help Desk at (301) 621-0134

• Telephone the NASA Access Help Desk at(301) 621-0390

Write to:

NASA Access Help Desk

NASA Center for AeroSpace Information7121 Standard Drive

Hanover, MD 21076

NASA/TM--1999-209436 AIAA 99-2949

High Power Hall Thrusters

Robert Jankovsky

Glenn Research Center, Cleveland, Ohio

Sergey Tverdokhlebov

TsNIIMASH Export, Korolev, Moscow, Russia

David Manzella

Dynacs Engineering Company, Inc., Brook Park, Ohio

Prepared for the

35th Joint Propulsion Conference and Exhibit

cosponsored by AIAA, ASME, SAE, and ASEE

Los Angeles, California, June 20-24, 1999

National Aeronautics and

Space Administration

Glenn Research Center

December 1999

NASA Center for Aerospace Information7121 Standard Drive

Hanover, MD 21076Price Code: A03

Available from

National Technical Information Service

5285 Port Royal Road

Springfield, VA 22100Price Code: A03

High Power Hall Thrusters

Robert JankovskyNational Aeronautics and Space Administration

Glenn Research Center

21000 Brookpark RdCleveland, Ohio 44135

Sergey TverdokhlebovTsNIIMASH Export

4 Pionerskaya StKorolev, Moscow Region

141070, Russia

David Manzella

Dynacs Engineering Company Inc.2001 Aerospace ParkwayBrook Park, Ohio 44142

Abstract

The development of Hall thrusters with powers ranging from tens of kilowatts to in excess of one hundred kilowattsis considered based on renewed interest in high powel; high thrust electric propulsion applications. An approachto develop such thrusters based on previous experience is discussed. It is shown that the previous experimentaldata taken with thrusters of 10 kW input power and less can be used. Potential mass savings due to the design ofhigh power Hall thrusters are discussed. Both xenon and alternate thruster propellant are considered, as aretechnological issues that will challenge the design of high power Hall thrusters. Finally, the intplications of such adevelopment effort with regard to ground testbtg and spacecraft integration issues are discussed.

Introduction

It has long been the goal of both US andRussian scientists to use high power electric pro-pulsion for primary spacecraft propulsion. Missionsto Mars, reusable space tugs, and other propulsionintensive missions have all been considered since

the early 1960's. The opportunities to conduct thesemissions have never arisen due to the lack of on-

board spacecraft power. Additionally, due to acombination of technical problems, politicalpressure, and unfortunate timing the possibility forspace nuclear power has been virtually eliminated.As a result the major electric propulsiondevelopment efforts over the last several decadeshas concentrated on devices with powers of only afew kilowatts or less.

With regard to Hall thrusters, these efforts,which were for the most part conducted in whatwas then the Soviet Union, have resulted in

successful operational deployment of these devicesfor stationkeeping purposes on Soviet and laterRussian spacecraft. The early engines operated at0.6 kW and later engines operated at 1.35 kW.Since the end of the cold war this very successfulHall thruster technology became available to therest of the world. The first implementation of this

technology on a Western spacecraft occurred in1998 when a device of the anode layer type(TALk derated for operation at 0.6 kW, was flownon a U.S. government experimental spacecraft.There are a multitude of additional spacecraftbeing planned that will use higher power Hallthrusters in the near future. The highest powersystem being developed for near term flight is a5 kW system utilizing a TI60E SPT type engineslated for flight in early 2000.

Spacecraft power system designs have steadilyevolved. Spacecraft with power levels approaching20 kW are currently being implemented. As aresult, electric propulsion, and specifically Hallthrusters, are being considered once again for hightotal impulse missions that have not beenconsidered since the possibility for space nuclearpower died.

The combination of advanced power systemsand advanced orbital trajectories have missiondesigners emphasizing a significant need for highpower Hall thruster technology with a high ratio ofthrust-to-power. This technology promises thepossibility of significant reductions in launch massmaking new missions possible with the existinglaunch vehicle fleet. Non-nuclear trans-lunar and

NASA/TM--1999-209436 1

trans-Marsinjections,spacesolarpowersatellites,SpaceStationReboost.LEO-GEO transfers in lessthan 90 days are a few examples of these newmissions._-"34 The thruster power levels required forthese missions range from 30 to 100 kW, inexcess of what is currently available. As a result,

the objective of this paper is to give preliminaryconsideration to the general issues associated withthe development of high power Hall thrusters, sug-gest reasonable limits for near term extrapolationof state-of-the-art (SOA) Hall technology, and tosuggest a development path for the even higherpower thrusters.

Nomenclature

T thrust

P powerI_ ion currentJ_ ion current densityI,t,, permeability of free space_ average ion exhaust velocitym mass flowrate (use typical sign with upper dot)e electron chargeM, molecular wt. ionB magnetic field strengthIH Hall current caused by electric drift of

electrons

D average diameter of thruster channelr_ ion Larmour radius

V_ discharge voltageb,. channel width

Ij discharge currentj_. electron current densityk_ constantk, constant

Sch channel cross section areaI m equivalent mass flow rate in current unitsI_,.. electron cyclotron frequencyag_.,,electron-neutral collision frequency

Background

There is limited published data relating to theperformance of high-power Hall devices. Thosedata that are available were taken with laboratorymodel hardware and are generally not compre-hensive in nature. However, a review of this work

indicates the current state-of-the-art with respect tohigh power Hall thrusters. Beginning in the early1980"s laboratory models of high power xenonstationary plasma thrusters with outer dischargechamber diameters up to 290 mm were developedand tested in Russia at Fakel in cooperation withMoscow Aviation Institute (MAI) and theKurchatov Institute. __ At the same time large-scale

anode layer thrusters (TALs) were developed andtested in Russia at TsNIIMASH in cooperationwith RSC "Energia." Initially these TAL testsinvestigated bismuth as a propellantfl Later tests

considered xenon as the propellant. In both casesthe objective of these tests was to demonstrateoperation with high specific impulse/ Morerecently the performance of one of these highpower TALs developed at TsNIIMASH for opera-tion on xenon, given the designation TM-50, wasevaluated at NASA GRC for operation in a highthrust, low specific impulse modefl Most recently,Jankovsky et al. reported the preliminary resultsfrom testing an SPT type thruster, designated theT-220 at NASA GRC. _' This thruster was

developed in the United States by TRW incooperation with Space Power Incorporated. Aportion of these data encompassing the range ofoperation for each these devices is summarized in

the Table 1 and shown graphically in Figure 1.

Physical Limitations

The operational features of low-power thrusterswith closed electron drift have previously beencomprehensively investigated. Therefore theanalysis below focuses on the physical constraintslimiting the maximum power of a single Hallthruster. Arrays of lower power Hall thruster are analternate strategy for obtaining high power Hallthruster propulsion systems, however neither thisapproach nor multi-channel Hall thrusters areconsidered in this analysis. Both of theseapproaches may offer certain advantages, but theyare evident design derivatives from the singlethruster channel which is considered.

The general desire to evaluate high-power Hallthrusters assumes their capability to operate withthe highest possible thrust-to-power ratio. Becausethis ratio is inversely proportional to the average

ion exhaust velocity, T/P~I/'o i, operation with

relatively low specific impulse, Isp, at lowdischarge voltages is preferable. Thus a thruster'sability to accelerate a maximum ion current I i is ofgreat interest. Correspondingly for a thruster ofgiven cross section, the maximum achievable ion

current density Ji represents a physical constraintwhich must be considered.

The maximum permissible value for J, in Hallthrusters can be estimated by utilizing themagnetohydrodynamic approximation for the

momentum flux density of ions accelerated by Hallcurrent. This approximation, first suggested by A.Zharinov and Y. Popov over thirty years ago,neglects electron pressure and the influence ofself-induced magnetic field:

NASA/TM--1999-209436 2

JiM_xh/e<_ B2/2JJo

Therefore, to increase ion current it is

necessary to increase the cross section of thethruster channel or increase the magnetic fieldstrength. When the magnetic field is increased tosatisfy this condition the ion current density is

proportional to the Hall drift current In:

Ji = Ire/2 l-d)ri

Where D is the average diameter of thrusterdischarge chamber and r, is the ion Larmour or

gyroradius.Because of the electron losses to the walls.

anomalous electron mobility caused by thescattering attributed to oscillations, and due toazimuthal non-uniformity's, the Hall current valueis always less than its theoretical maximum.

X,,

I.= fj_6J,(1/o,.)a.,0

Another potential limit to scaling of Hallthrusters is the distortion of the external magneticfield, caused by the diamagnetic influence of theHall current. If the self-induced magnetic field ofthe Hall current is significant, under certainconditions, a significant distortion of the electricfield distribution along the thruster channel and

subsequent unpredictable thruster performance maydevelop. Although all aspects of this phenomenaare not well understood, the strong gradient of

magnetic field in the thruster channel may help tominimize this distortion.

Another potential limit to the maximumdischarge current is Joule heating of the anodecaused by the back-streaming of electrons. Theaverage electron energy at low mass flow rates fora TAL may be as high as eva/5. This requires upto 20% of the input power be radiated or conductedaway from the anode. However, generally as theneutral density increases the average electronenergy decreases, but the total power increasesand the anode energy dissipation remains constantover a relatively wide range of thruster operation.This is referred to as the "power plateau". Furtherincreases in propellant flow rate above this "powerplateau" results in a substantial increase inelectron back-streaming evident by large increasesin discharge current with incremental flow rateincreases. Substantial flow rate, or current

increase, in this regime of operation are generallylimited by the thermal design of the anode whichserves as a practical limit for high power operationof a given device.

The choice of propellant also impacts thenecessary ion current density for efficient thrusteroperation. While the previously presented equationshows that a lower magnetic field is all that isrequired to accelerate the same flux of lighter ionsas compared to a heavier ion, an acceptable valueof propellant utilization (or efficiency) requireshigher current density for the lighter propellantJ jThus, while the thermal limitations of a giventhruster are not affected by the choice of

propellant, the necessity to increase the dischargecurrent to maintain the efficiency while operating

with lighter propellants does reduce the marginwith respect to the thermal design. The increase indischarge current also results in a reduction in theattainable discharge voltage at fixed power.

Because these considerations are general in

nature they are applicable to both SPT and TAL-type Hall thrusters. A quasi-neutral mode ofoperation is assumed, and the so called vacuummode of TAL operation which has been previouslydescribed by others _' is not considered primarilydue to the unfavorable thrust-to-power ratioencountered during operation in this mode.

Design Parameters

Simple correlations between power and size oflarge-scale Hall thrusters can be easily madebased on existing laboratory, engineering and flightmodels of SPTs and TALs. To assess the target

performances for high power Hall thrusterdevelopment comparative analysis was made withuse of published data v-12131_15on power, character-istic size and mass of 6 SPT models and 5 TAL

models. For comparison convenience the values of

power and mass flow rates were taken from datacorresponding to an l_p of approximately 2500 sec.Also, for convenience an average diameter of thethruster channel was used instead of the outer

diameter which is the convention in SPT-related

publications. Thrusters with average diameter of38-250 mm were considered.

Figure 2 illustrates the general scaling trend ofHall devices as power increases with channel size.For a discharge chamber with an average diameterof 400 mm power levels of 37 kW and 48 kWwould be predicted for operation at 2500 secondspecific impulse with an SPT and TALrespectively. This result is consistent with the factthat TAL is characteristically smaller device than

SPT for a given power level (for example an SPT-100 and TAL D-55 have the same nominal power

capability). As is also seen in this figure with datafrom TAL testing, by increasing the specific

NASA/TM--1999-209436 3

impulserequirementfora giventhrustersizethemaximumpredictedpowerlevel also increases,howeverattheexpenseof thrust.Thisdependenceof specific impulseon dischargevoltage ispredictablebasedon themassflow rate.thrusterdimensions,and the exchangeparameter_,definedas the ratio of the dischargecurrentdividedby thecurrentequivalentmassflow rateforasinglyionizedpropellant.

With somemathematicalmanipulationonecan write the thrusterpower in termsof thedischargevoltage,the exchangeparameter,thethrusterdiameter,andseveralvariablesdependingontheareaandwidthof thedischargechamber:

P = kiD2(l+k2/D) _, Va .where k_ _ re�Sen and k2 _ m/D

In SPT thrusters the parameter of exchange _,

has been shown to increase linearly with theincreased channel average diameter D and widthbe. For instance, with an increase of averagediameter from 50 mm to 250 mm, the increase in

the exchange parameter from 0.9 to 1.4 wasreported. _6 However, the ratio of m/Sch remainedconstant. Previously it has been suggested that theratio of tolD remains constant with thrustersize, _6_7 however, based on the data shown in

Figure 3 there is some variation with thruster sizeand. therefore, it is believed that in general m/D¢constant. More specifically, those cases in whichm/D is held constant are not of significant interestbecause they require an increase of the thrusterdiameter while holding other discharge chamberparameters constant, thus sacrificing performance.Finally, while it appears that for the most partm/sch is constant it is possible to increase thisratio at the expense ot" thruster lifetime.

Similar considerations may be used toestimate thruster mass. The required magnetic fieldstrength B is directly proportional to the dischargecurrent la for a wide range of mass flow rates (fordischarge voltages in the range of 200-600 V). Themass of the magnetic system usually representsabout 75-80% of total thruster mass. The difference

between the mass value for existing engineeringmodels and the targets for flight models was alsoconsidered to estimate the thruster mass reduction

trend during the development efforts from labor-atory to flight status. The values of thruster masswere corrected to exclude cathode(s) and orificeblock. The results are presented in the Figure 4.

The mass reduction trend is generally similarfor both SPT and TAL, so the approximate valuesof 50 and 40 kg may be predicted for both an SPTand TAL 50 kW, 2500 sec laboratory and flight-weight thruster. Thruster specific mass goes downwith the increase of power. Therefore, for

comparison purposes, it can be seen that a single50 kW thruster would be 30'7c lighter than ten5 kW thrusters.

A simple parametrical approach like the onedescribed above does not involve a thorough con-sideration of all the physical constraints such ascritical ion current density or thruster thermalmode. Nevertheless, it is useful for initial engineer-ing estimates and it gives an opportunity toimagine the sequence of attempts to employ thetraditional Hall thruster design in the extremeconditions. For instance, it may be shown that a

100 kW Hall thruster with a specific impulse of1600 s (300 V for xenon) may be as large as onemeter in average diameter.

Technological Issues of Thruster Design

The development of high-power Hall thrusters

as well as of any large-scale device representscertain design and technological challenges. Whilethere are numerous possible design configurationsfor Hall-current devices only the most generalcharacteristics will be considered. First. a linearincrease in Hall thruster dimensions involves

certain difficulties with providing an azimuthallyhomogeneous magnetic field and gas distributionin the thruster channel required for normal driftmotion of electrons. Local non-uniformity's mayresult in potential eccentricities in the thrusterplume. Potential reasons include, but are notlimited to, local changes in the discharge chamberwalls conductivity (issue of ceramic materialhomogeneity), mechanical alignment, and cathodelocation. These may lead to several effects such asnon-symmetric wear of the discharge chamberwalls and shifts in the thrust vector. All these

features that are generally inherent to any size ofHall thrusters may be strongly amplified whileoccurring in a large design.

Increase of thruster size may also complicatethruster design from a thermal-mechanical stand-point. This is because absolute values of linearexpansions expected for thermal stressed elementsof the thruster are proportional to its averagediameter and are comparable with the character-istic length of acceleration zone. Another signi-ficant issue is the temperature gradients originatingacross the channel walls. This will necessitate a

transient as well as steady-state thermal analysisof critical components such as thruster ceramics. Apreliminary consideration has already been givento transient analysis of high power hall thrusters. Ithas been demonstrated experimentally that it may

take multiple hours to reach thermal equilibrium.In one experiment a non-monotonic increase intemperature was observed during a transient mode.

NASA/TM-- 1999-209436 4

Thisindicatespotentiallydifferentcontributionintime of the thermalconductivityand radiationmechanismsof heat exchange.Correspondingchangesin performancesmayaccompanythesedifferentthermalregimes,whichtherefore,willneedto beaconcernof futurehighpowerdesigns.

Generally,if theapproximateheatflux distri-bution in the thrusterchamberis knownthesimulationof athrusterthermalmodemaybedonewith useof finite elementsmethods.The onlyproblemis to determinea boundaryconditionsonheat-exchangingand radiatingsurfaces.As fordetailedthermalanalysisof a thruster,oneof theremaininguncertaintiesis associatedwith thecorrectdeterminationof theenergyassociatedwithplumeemissionin theVUVrange.

Anotherissueis thedevelopmentof a specialpropellantinsulator.Becauseof increasedvolt-ages,largemassflow rates,highanodetemper-ature,presenceof highfrequencyelectricfieldandresidualmagneticfield,etc.,a newreliabledesignis required.So,it is clearthattherearepracticalimitationswithregardhowlargehighpowerHallthrusterscanbe made,basedon theindividualcomponentsusedto makethesethrusterswhichwill haveto beaddressedin orderto permitthedevelopmentofhighpowerHallthrusters.

Propellant

Propellant selection for High Power Hall is of

critical importance. Xenon is the only propellantunder utilization in the current and near term

missions employing Hall thruster propulsion. While

typical magnitudes of propellant mass required forprojected telecommunication satellites with 1 to5 kW Hall thrusters does not exceed 100 kg,hundred-kW space missions would require tons of

propellant. Assuming total thrust time for two roundtrips to a high elliptical point of trans-planetinjection, utilizing 500 kW of solar electric power,a propulsion system consisting of 10 Hall thrustersconsuming 50 kW power each could be imagined.Such thrusters might have a flow rate of 100 mg/sof xenon each with a specific impulse of 2500 sec.For this mission, a propellant mass of ~30 metrictons is required. Availability of xenon supply,therefore, becomes one of the major concerns.Potential alternatives under discussion include

krypton, xenon-krypton mixtures or condensablemetal propellants.

From a purely thruster physics standpoint, it isclear that to maximize thrust-to-power the best

choice of propellant is the highest molecularweight element that that does not pose an

excessively high cost for ionization. A fewexamples of alternate propellants are in Table 2.For comparison purposes changes in thrusterefficiency with alternate propellants is notincluded. While only considering the thrusterphysics in selecting a propellant is not arecommended approach, it is the basis from whichinitial considerations should start. From the initial

thruster physics basis other considerations such aspropellant management, spacecraft contamination,

ground testing, and environmental impact can betraded against trip time, spacecraft power systemand launch vehicle for each application. The trade"

for a geostationary spacecraft that desires to dolow Earth orbit to geostationary orbit transfer witha Hall thruster or the International Space Stationwhich desires to use a Hall thruster for reboost may

be quite different than a manned mission to Marsor a trans-Neptune stage.

If a propellant other than xenon is pursuedseveral issues will need to be addresses. Here we

will qualitatively consider only some of them forthe most commonly considered alternative,krypton. The first is with respect to the impact ofutilizing krypton as a propellant which may seemattractive based on it similarity to xenon. In

practice, however, the higher ionization potentialand lower atomic weight of krypton lead tosubstantial decreases in Hall thruster performance.

Preliminary investigations conducted with single-stage SPTs and TALs operated with kryptondemonstrated 15-20% less efficiency than withxenon, ts Even with a greater than 50% increase inmass flow rate and subsequent increase in the dis-

charge current the expected efficiency on kryptonin the specific impulse range of 2000-2500 sec willprobably not be in excess of 40-45%. The penaltyassociated with the operation at the increaseddischarge currents is a higher thruster thermal loadas compared with operation on xenon.

Mixtures of gases such as xenon and kryptonhave also been previously considered. For our con-sideration, however, the extra energy dissipation inthe thruster structure may still be the limitingtechnical issue because it constrains the maximum

input power of a thruster with given size. Certainly,the addition of 10-50% molar impurity of xenon toa krypton propellant will result in improvement ofthe thruster performance as compared to purekrypton. However, no principal "resonant"phenomenon facilitating the increase of partialkrypton utilization efficiency with xenon are

expected. This is verified by limited experiment-ation which demonstrated the additive character ofxenon's contribution to thrust and efficiency.Therefore. the attractiveness of use of the

NASA/TM--1999-209436 5

xenon-kryptonmixture"to resolvea cost andpropellantavailabilityissuesis still in directproportionof xenonfractionused.Ultimatelyeithercostorothersystemsconsiderationsarelikely todrivethe propellantselection.However.at thistime preliminaryanalysisbasedon combinedconsiderationof cost/performances/systeminte-grationimpactsshowsthat the mostreasonableand cost effectiveapproachfor the nearestdevelopmentof high-powerHall thrusteris tocompletethe evaluationand test programofhardwareon xenon.Also a dedicatedanalysiswhichgivescarefulconsiderationto the benefitsvs.environmentalimpactsof developmentof metalpropellantsystemsshouldbeundertaken.

Test & Integration

The testing and integration of high power Hallthrusters will have an impact on the results of anydesign consideration for this type of thruster. Ofcourse, the magnitude of the impact is dependenton such things as thruster type and propellantselection (i.e.. non condensable versus condens-

able). Also. the possibility for things like plumeimpingement onto spacecraft surfaces, or con-tamination of sensitive optical surfaces are highlydependent on a particular geometry and spacecraftconfiguration. In the following discussions theseissues will be considered in general for a singlethruster. If an array of thrusters is utilizedadditional complications, beyond those discussedhere may be manifest.

Although it may be possible to design a Hallthruster for a given application with favorableoperational characteristics in space, if the designcannot be qualified and acceptance tested in anacceptable fashion on the ground the design is ofquestionable utility. As a result, the issues relatedto the ground testing of large Hall thrusters are ofprimary importance. The main technical consider-

ation with regard to ground testing is what level ofvacuum is needed within the ground test facility toadequately simulate the environment in which thethruster would operate in space. The ability of atest facility to sustain a given vacuum pressure isrelated to its pumping speed. Typically a surfacewhich will condense the propellant upon it is usedas the pump. this often times requires cryogenictemperatures for gases such as krypton and xenon.

Note that obtaining of a pure Xe-Kr mixture may be even more

expensive procedure than direct production ojXe fi'om initial

pro&tct of industrial air -separating facilities, containing

0.15..0.25 % qfxemm and ko'pton.

Previously, for consideration of propellantselection, a xenon flow rate of 100 mg/s was con-sidered as a possibility for a single 50kW thrusterwith a specific impulse of 2500 sec. For thrustersof the SPT type the facility pressure was shown tohave an effect on the measured performance atpressures above 2x 10-_ torr. j5 In order for this effect.

thought to be caused by ingestion of backgroundgas into the discharge chamber of the engine, to beinsignificant, a xenon pumping speed in excess of600,000 liters per second would be required. Whilethis pumping speed is within the range of currentlyexisting electric propulsion test facilities, thosewith a order of magnitude higher pumping speedsare not. Additionally, thrusters of the traditionalanode layer type tend to have a smaller crosssectional area at the exit plane than do SPT typethrusters. As a result it is likely that chamber

pressure does not have an effect on performanceuntil slightly higher pressures are reached.

These considerations have focused on theeffect of mass flow rate, which to first order deter-mines the current in a Hall thruster and on vacuum

facility background pressure during testing. Whendiscussing high power operation the effect ofdischarge voltage should also be considered. Asdischarge voltage is increased the effect of theingested background gas on performance changes.This is because at higher voltage, the ingestedneutral gas has a lower probability of diffusingcompletely through the acceleration zone withinthe discharge chamber before being ionized byelectron impact and subsequently accelerateditself. Therefore these ingested, ionized, and sub-sequently accelerated background atoms do notproduce as much thrust as those propellant atomsintroduced at the rear of the discharge chamberthrough the anode. This is the basis for the varyingeffect of background pressure on performance withdischarge voltage. The over riding consideration isthat in any ground test the effect of the finitebackground pressure must be adequately under-stood. Corrections can be made to account for this

effect, but if the corrections are not adequate areal possibility exists for thruster operation inspace to be significantly different than duringqualification and ground testing.

A second consideration with regard to testingof high power Hall thrusters is the ability toadequately address relevant integration issues.While these individual issues will subsequently beconsidered, a significant number of these are

related to the distribution of effluent in the plume.Background gas within a test facility duringthruster operation may charge exchange withenergetic plume ions affecting the distribution andcomposition of the plasma generated by theengine. While estimates of this effect can be made

NASA/TM--1999-209436 6

basedonknownchargeexchangecrosssections,_brexample,experimentalevidenceindicatesthatevenatpressuresbelow2x10-_torrthisphenomenaisstillsignificantfor thoseregionsof theplumeatlargeangleto the directionof thrust._ For thesame100mg/sHall thrusterconsideredabove,thiscorrespondsto a 6,000,000l/s requiredpumpingspeedwhich,aspreviouslystated,is notcurrentlypossiblein existing electric propulsiontestfacilities.

Anothertestingissueof seriousconsequenceisrelatedto sputteringof thesurfacessubjectedtodirection impingementwithin the test facility.Conceptuallythis issueis easyto visualize.Theengineisdesignedtoproducehighenergyparticleswhichcantravelat velocitieswell in excessof20km/s.Dueto theextremeenergeticnatureoftheseparticles,uponimpactwithasurfacethereisafiniteprobabilitythattheywill removeorsputtera portionof theparentmaterial.Afterlongtermbombardmentof this type the magnitudeofmaterialliberatedcanbecomequite substantial.For this reason, sputter resistant materials such asgraphite are often used as beam targets. Theerosion of the facility itself posses no real con-sequence to the test as long as the actualstructural integrity of the facility is not com-promised, however, the sputtered material willredeposit elsewhere within the facility, includingon the test hardware with potentially adverseresults. This effect can be minimized through

careful design of targets and baffles, but iscertainly a significant issue with respect to thelong term testing of high power Hall thrusters. Toillustrate the magnitude of this effect, again con-sider the 50 kW thruster with a xenon mass flow

rate of 100 mg/s. For a graphite target at 5 m fromthe exit plane of the engine the surface would beeroded 0.1 mm after 5000 hrs of operation. If thesurface was stainless steel it would be eroded to a

depth of 0.6 mm. While this may not seemsubstantial the total amount of graphite sputteredthroughout the tank would be as much as 46 kg andfor steel in excess of 1000 kg. Clearly theredeposition of this material may be a significantconcern for the test hardware.

The integration issues associated with highpower Hall thrusters are not significantly differentthan those associated with lower power Hallthrusters, although the impact on a particularspacecraft may be considerably greater due to thelarger size and higher fluxes. Direct impingementof particles accelerated out the engine transferboth momentum and thermal energy. For highpower thrusters this issue will likely be minimizedthrough spacecraft design, but this certainly will bean integration issue worthy of consideration.

Another issue of potentially significant impact

from a spacecraft integration issue is theaccommodation of the waste heat generated during

thruster operation. A thruster with a reasonablyefficient thermal design may still need to reject25% of its input power as thermal energy.Minimizing the effect of the conducted andradiated heat on the spacecraft will be a signi-ficant design challenge for the high power Hallthruster spacecraft integration specialist.

There are a number of additional issues for

consideration from a spacecraft integration per-

spective. These include the possibility of an offaxis thrust vector which could vary in time. If anengine does produce a time varying force in a nondesirable direction, the impact on a spacecraft's

attitude control system maybe significant. Otherissues include the effect of radiated electro-

magnetic emissions produced by the plasmagenerated within the operating high power Hallthruster. While this has proven to be a manageableissue for lower power thrusters, the way in whichthis scales to higher powers remains unknown.Finally, such things as contamination of sensitivesurfaces by deposition of engine erosion productsand possible optical interference from the lightemitted by the plasma produced within the enginemust also be considered.

Discussion

In order to successfully develop the nextgeneration of high power Hall thrusters it isnecessary to consider each of the various topics ofconsideration both separately and as they relate toone another. With this as a basis, the individual

technological aspects of such a program can beappropriately addressed. As previously mentionedthere is limited amount of prior data from highpower Hall thruster testing. What is noteworthy,however, is that these data were, in general.adequately predicted by previous data obtainedwith lower power Hall thrusters. This demonstratesthe efficacy of extending the current technology ina linear fashion which will likely provide signi-

ficantly increased performance with a modestoutlay of resources. This is primarily enabled bythe very substantial body of work on Hall thrustersat power levels of 10 kW and less.

There are physical limitations which dictatethe maximum power or power density possible withHall thrusters. These limitations which are not

accounted for by a scaling type of approach weredetailed previously. The most fundamental of these

physical limitations is the maximum ion currentdensity possible within the discharge chamber of a

NASA/TM--1999-209436 7

highpowerHall thruster.Thevalueof this limitdependson thepropellantandthrustertype(i.e.,SPTorTAL),butultimatelydeterminesthesizeofathrusterforagivenpowerlevel.Furthermore,duetotheannularnatureof Hall thrusters,asthesizeof thethrusteris increasedto permithigherpower.theactualpowerdensityorpowerperdiameterofthechanneldecreases.Due to this fact+as theouterdiameterof theHall thrusteris increasedinsizeto accommodatehigherpower,eventuallyanionthrusterwill theoreticallyapproachthethrustofa Hall thrusterfora giventhrusterdiameter.It isdifficult to predictwherethesetwocurveswillintersectbecauseof a sparcityof data,andtherealsoremainsa significantnumberof technicalissuesto be resolvedwithregardto thedevelop-ment of large griddedthrusters,but for highthrust/lowspecific impulseapplicationsit isexpectedthatthe Hall thrusterwill offerhigherthrustdensitiesforpowerlevelswell in excessof100kW.Ofcoursethisdoesnotconsidera multichannel approachor possible non circulargeometrieswhichwouldsubstantiallyincreasethethrustdensityin comparisonto a singleclassicannulardischargechamber.This leadsto theconclusionthatfor thrustersoptimizedto providehighthrustwithvoltagesontheorderof 300Volts,powersup to 100kW couldbe consideredwiththrustersizesupto 100cmindiameter.

Therearehoweverpracticallimitationswithregardto themaximumsizeof thrusterwhichcanbe fabricated,theselimitationsarebasedon theavailabilityof thenecessaryceramicsandrefrac-tory metalsin thesizesneeded.Whilea detailedassessmentonthemaximumsizesof rawmaterialswere not undertaken,anecdotalexperienceindicatesthatevenat powerlevelsof 10kW oneapproachesthepracticallimitsof whatis currentlyavailable.Therearealternatestrategieswhichcanbeadopted.Usingmultiplepiecesin placeof whathastraditionallybeena singlepieceis anobviousapproach.Theimplicationsof thiswith regardtosurvivingthe mechanicalloadsassociatedwithlaunchmayevenbefavorable.However,thereis adegreeof riskassociatedwithsucha strategythatwill needtobeaddressed.

Theissueof propellantchoiceis perhapsthesinglebiggestconsiderationwith regardto thedirectionof future high power Hall thrusterdevelopment.Froma costto develop,technicalandhistoricalbasis,theexperiencecurrentlyexist-ingwithxenoncoupledwithfavorableperformancemakethis propellantan overwhelminglyprefer-ablechoice.Problemswithpriceandavailability,however,maymakethischoiceuntenablefromacostandlogisticsstandpoint.If otherpropellantsareconsideredadditionaldevelopmentis needed.Someof this couldbe conductedwith smaller,

lowerpowerthrustersto minimizecost. Theadditionalissuesthat will likely be determiningfactorsforpropellantchoicesuchasenvironmentalconsiderationstendto be somewhatpoliticalinnatureandwill notbediscussedas partof thispaper.

Theissuesrelativeto testingwill alsohaveaverysignificantimpacton thedirectionof futurehigh powerHall thrusterdevelopment.This isprimarilybecausethe cost of upgradingtheinfrastructureof thevariouselectricpropulsiontestfacilitiesrequiredto testhighpowerHall thrustersmayexceedthe costof the developmenteffortitself. At NASAGRC,whichhasone of thehighest pumping speed dedicated electricpropulsiontest facility, the pumpingspeedofxenonisapproximately1 millionliterspersecond.Fora pressureof 2x106torrthiscorrespondsto aflowrateof 17mg/swhichat 300Voltsis onlya5kWthruster.Inorderto conductresearchonhighpowerHall thrustersat higherpressurestheimplicationsof the effectof backgroundpressurewill havetobemorecompletelyunderstood.

Concluding Remarks

In summary, the implications of this study arethe following: Based on past data there areperformance benefits to be gained by increasingthruster size. The wealth of previous experimentaldata taken with thrusters of 10 kW input power andless form an excellent basis upon which to develophigher power Hall thrusters. There is a masssavings in using fewer number of higher powerthrusters as compared to a larger number of smallerthrusters to obtain a given power level. The desiredpropellant is xenon, however from a cost. avail-ability, or system perspective high power Hallthrusters using alternate propellants may bedeveloped requiring a greater development effort.There are practical limitations with regard to howlarge high power Hall thrusters can be made.Creative engineering can stretch these limits, butultimately development in the such things as rawmaterial fabrication issues may be required. Therequirements of the ground test programs

necessitated by the development of high powerhall thrusters will represent a real limitation in the

practical limit for thruster development. Finally,all these considerations have assumed a traditional

linear or incremental development of the tech-nology. History indicates that a technical break-through permitting a non linear technology jump ispossible and even likely during the developmentcycle for such engines. If not, even now. Hallthrusters with power levels up to 100 kW can beconsidered which will enable missions to Mars,

NAS A/TM-- 1999-209436 8

reusablespacetugs,andotherpropulsionintensivemissionswhichhavebeenunderconsiderationfordecades.

References

I. Oleson, S.R., Advanced Propulsion for SpaceSolar Power Satellites, AIAA 99-2872

2. Gefert, L.P, Hack, K.J., Kerslake, T.W.,

Options for the Human Exploration of Marsusing Solar Electric Propulsion. proceedings ofthe Space Technology and ApplicationsInternational Forum--1999.

3. Oleson, S.R., Influence of Power SystemTechnology on Electric Propulsion Missions,NASA CR-195419

4. Oleson, S.R., Myers. R.M., Launch Vehicleand Power Level Impacts of Electric GEOInsertion, AIAA-96-2978

5. Arkhipov, B.A. et al., "Development andInvestigation of Characteristics of IncreasedPower SPT Models", IEPC-93-222, Proceed-

ings of 23-rd International Electric PropulsionConference, Seattle, September 1993.

8. Arkhipov, B.A. et al., "Investigation of SPT-200 Operating Characteristics at Power Levelsup to 12 kW", IEPC 97-132, Proceedings of25 'h International Electric Propulsion Confer-ence, Cleveland, August 1997.

7. Ageev V.P., Safronov I.N., Tverdokhlebov S.O.:Performance Characteristics and Peculiarities

of Two-Stage Accelerator with Anode LayerOperation with Low Voltages and MagneticField, Sov. Journal of Appl. Mech.&Tech.Phys. No. 6, 1987. pp. 3-7.

8. Semenkin, A., Tverdokhlebov, S. et al., "TAL

Thruster Technology for Advanced ElectricPropulsion System". 96-a-3-26, 20 m ISTS,May, 1996, Gifu, Japan.

9. D. Jacobson, R. Jankovsky, "PerformanceEvaluation of a 50 kW Hall Thruster", ALAA

99-0457, Presented on 37a AIAA AerospaceSciences Meeting and Exhibit, Reno, January1999.

10. R.S. Jankovsky, C. McLean, J. McVey,"Preliminary Evaluation of a 10kW HallThruster", AIAA-99-0456, Presented on 37 t"

AIAA Aerospace Sciences Meeting andExhibit, Reno, January 1999

11. Melikov, I. et al., Soviet Journal of Technical

Physics, 1974, Vol. 44, No3.12. Zhurin, V. et al., "Physics of Closed Drift

Thruster", IEPC-97-191. 25 'h IEPC, August,1997, Cleveland.

13. Arkhipov, B. et al. "Extending the Range ofSPT Operation: Development Status of 300and 4500 W Thruster". AIAA-Paper.Presented on 32 ndJoint Propulsion Conference,Orlando, 1996

14. Gavryshin, V.M., Kim, V. et al., "Physical andTechnical Bases of the Modern SPT

Development", IEPC 95-38, Proceedings of24 _h International Electric PropulsionConference, Moscow, 1995.

15. Sankovic. J., Hamley, J., and Haag, T.,"Performance evaluation of the Russian SPT-

100 Thruster at NASA LeRC," AIAA-93-094,

September. 1993.16. Bugrova, A. et al., "Scaling Laws for Integral

Petformances of SPT", Soviet Journal ofTechnical Physics. 1991. Vol. 61. No. 6,pp. 35-51.

17. Bugrova, A., Kim, V.. "Modern State ofPhysical Investigation in the Accelerators withClosed Electron Drift and ExtendedAcceleration Zone "--Paper in the book:Plasma.

18. Marrese, C. et al., "D-IO0 Pelformance andPlume Characterization on Krypton ",AIAA-96-2969, 32 _J AIAA JPC, July, 1996,Lake Buena Vista, FL.

19. Manzella, D. and Sankovic, J "Hall Thrusterion beam Characterization," AIAA-95-2927.

July, 1995.

NASA/TM--1999-209436 9

Table1:PerformanceofpreviouslydevelopedlaboratorymodelhighpowerHallthrusters

Thrustertype

Averagediameter,mm

Isprange,sInputpower,kW

Thrust.mN

SPT-290Ref.5250

1500-300012-30

T-220Ref.10

188

1500-24005-11

TALTM-50Ref.9200

1500-330010-25

TAL-200[Bi]Ref.7200

2000-520010-34

SPT-200Ref.6175

1500-30006-11

1500 524 1114 1130 498Efficiency 0.7 0.62 0.66 0.67 0.63

Table2:ComparisonofalternateelementalpropellantsElement

Radon

MolecularWeight

FirstIonizationPotential,eV

%ThrustofXenon

222 10.7 130

% Ispof Xenon

77

Bismuth 208.98 7.3 126 79

Lead 207.19 7.4 126 80

Mercury 200.59 10.4 124 81Cesium 132.905 3.9 101 99

Xenon 131.30 12.1 100 100

Krypton 83.80 14.0 80 125

Argon 39.948 15.8 55 181

t500

ZE.,d'

[[,.

1000

50O

tl,

n

* i

AQ[]

[]

• U •m

00

[]

m _ •

• SPT-200•'T-220

• SPT-290

i I_TM-50I

• TAL-200

10O00 20000 30000 40000

Power, Watts

Figure 1 : Thrust versus power of previously tested high power Hall thusters

5O00O

NASA/TM-- 1999-209436 10

6O

5O

40

030a.

t"

2O

10

0

o SPT, Isp=2500 s

" TAL, I,sp=2500 s

[] SPT, ,l_p=2000 s1

• TAL, Isp=2000 s

+ TAL, Isp=3000 s

• TAL, Isp=1700 s

+

100 200 300 400

Average Diameter, mm

Figure 2: Projected Effect of Input Power and Specific Impulse on Thruster Size

5O0

NASA/TM-- 1999-209436 11

m

e,,

_E.o

4500

4000

3500 -

3000 "

2500 -

2000

1500 -

1000 -

500 -

o oo J-_ 1.1-11_

0 i_ ..f._ _ .lllfJ

f/1-

jt_

ft I I-

o SPT

• TAL

0 .............. T" _ - "

0 0.05 0.1 0.15 0.2 0.25

m/D, mg/s mm

Figure 3: The ratio of mass flow rate divided by average thruster diameter versus specific impulse

-i

0.3

5O

,,r 30

E 20

4O

10

, J

0 5 10

J I

JJ

J

i i15 20 25 30 35 40 45 50

Thruster Power, kW (|s[_/,5_ s)

Figure 4: Projected thruster mass versus power for flight and engineering model thrusters

NASA/TM--1999-209436 12

REPORT DOCUMENTATION PAGE Form ApprovedOMB No. 0704-0188

Public reporting burden for this collection of informalton is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources,gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding th,s burden estimate or any other as_ct of thiscollection of information, including suggestions for reducing this Our0en, to Washington Headquarters Serwces, Directorate for Information Operations and Reports, 1215 JeffersonDavis Highway, Suite 1204, Arlington, VA 22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188), Washington, DC 20503

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED

December 1999 Technical Memorandum5. FUNDING NUMBERS4. TITLE AND SUBTITLE

High Power Hall Thrusters

6. AUTHOR(S)

Robert Jankovsky, Sergey Tverdokhlebov, and David Manzella

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)

National Aeronautics and Space AdministrationJohn H. Glenn Research Center at Lewis FieldCleveland, Ohio 44135-3191

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)

National Aeronautics and Space Administration

Washington. DC 20546-0001

WU-632- I B- 1B-00

8. PERFORMING ORGANIZATIONREPORT NUMBER

E-11912

10. SPONSORING/MONITORING

AGENCY REPORT NUMBER

NASA TM--1999-209436

AIAA 99-2949

11. SUPPLEMENTARY NOTES

Prepared for the 35th Joint Propulsion Conference and Exhibit cosponsored by AIAA, ASME, SAE. and ASEE,

Los Angeles, California, June 20-24, 1999. Robert Jankovsky. NASA Glenn Research Center, Cleveland, Ohio, Sergey

Tverdokhlebov. TsNIIMASH Export. Korolev. Moscow. Russia and David Manzella, Dynacs Engineering Company,

Inc., Brook Park. Ohio. Responsible person. Robert S. Jankovsky, organization code 5430, (216) 977-7515.

12a. DISTRIBUTION/AVAILABILITY STATEMENT

Unclassified -Unlimited

Subject Category: 20 Distribution: Nonstandard

This publication is available from the NASA Center/'or AeroSpace Information. (301) 621-0390.

12b. DISTRIBUTION CODE

13. ABSTRACT (Maximum 200 worde)

The development of Hall thrusters with powers ranging from tens of kilowatts to in excess of one hundred kilowatts is

considered based on renewed interest in high power, high thrust electric propulsion applications. An approach to developsuch thrusters based on previous experience is discussed. It is shown that the previous experimental data taken withthrusters of 10 kW input power and less can be used. Potential mass savings due to the design of high power Hall thrustersare discussed. Both xenon and alternate thruster propellant are considered, as are technological issues that will challengethe design of high power Hall thrusters. Finally, the implications of such a development effort with regard to groundtesting and spacecraft integration issues are discussed.

14. SUBJECT TERMS

Hall thruster

17. SECURITY CLASSIFICATIONOF REPORT

Unclassified

18. SECURITY CLASSIFICATIONOF THIS PAGE

Unclassified

19. SECURITY CLASSIFICATIONOF ABSTRACT

Unclassified

15. NUMBER OF PAGES

1816. PRICE CODE

A9320. LIMITATION OF ABSTRACT

NSN 7540-01-280-5500 Standard Form 298 (Flev. 2-89)Prescribed by ANSI Std. Z39-18298-102