NASA Technical Memorandum 103227AIAA-90-2582

Performance Characterization of aSegmented Anode Arcjet Thruster

Francis M. CurranNational Aeronautics and Space AdministrationLewis Research CenterCleveland, Ohio

David H. ManzellaSverdrup Technology, Inc.Lewis Research Center GroupBrook Park, Ohio

and

Eric J. PencilNational Aeronautics and Space AdministrationLewis Research CenterCleveland, Ohio

Prepared for the21st International Electric Propulsion Conferencecosponsored by the AIAA, DGLR, and JSASSOrlando, Florida, July 18-20, 1990

ES

NASA

https://ntrs.nasa.gov/search.jsp?R=19910000805 2020-05-09T00:39:27+00:00Z

PERFORMANCE CHARACTERIZATION OF A SEGMENTEDANODE ARCJET THRUSTER

Francis M. CurranNASA Lewis Research Center

Cleveland, Ohio 44135

David H. ManzellaSverdrup Technology, Inc.

Lewis Research Center GroupBrook Park, Ohio 44142

Eric J. PencilNASA Lewis Research Center

Cleveland, Ohio 44135

ABSTRACT

A modular, I - 2 kW class arcjet thruster incorporating a segmented anodelnozzle was operated on athrust stand to obtain performance charcteristics of the device and to further study its operatingcharacteristics under a number of experimental conditions. The nozzle was composed of five axialconducting segments isolated from one another by boron nitride spacers. The electrical configurationallowed the current delivered to the arcjet to be collected at any combination of segments. Both the currentcollected by each segment, and the potential difference between the cathode and each segment weremonitored throughout the test period.

As in previous tests of a similar device, current appeared to attach diffusely in the anode when all of thesegments were allowed to conduct. Improvements to the device allowed long term (4 - 8 hour) operation atsteady-state and operating characteristics were repeatable over extended periods. Performance characteristicsindicated that the segmented anode reasonably simulates the behavior of solid anodes of similar geometry.Current distribution depended on flow rate as the arc attachment moved downstream in the nozzle withincreases in the mass flow rate. The current level had little effect on current distribution on the anodesegments. Thrust measurements indicated that the current distribution in the nozzle did not significantlyaffect performance of the device.

INTRODUCTION

The past seven years have seen an extensiveresearch effort directed toward the development oflow power (1 - 2 kW class) arcjet technology foruse in north-south stationkeeping ofgeosynchronous communications satellites.While maintaining compatibility with currentand next generation spacecraft subsystems, thesethrusters will provide significant performanceimprovements over existing, state-of-art,resistojet and chemical systems. The propellantsavings realized can be used to reduce launchmass, and/or increase payload, and/or to increasesatellite lifetime.

To date, on-going research and developmentprograms have met many of the technology goalsnecessary for successful application of arcjetsystems. These include development of a pulsewidth modulated power processing unit with an

integrated, pulsed, high voltage starting circuit;Idemonstration of stable, reliable operation onhydrazine decomposition products at specificimpulse levels between 450 and 500 seconds;2-5successful completion of an automated, cycliclifetest;6 and the recent assembly and testing of aflight-type arcjet system. 7,8 In addition to thistype of component and system level testing,attempts to assess the impacts of arcjet systemson spacecraft subsystems are also in progress.Langmuir probe surveys of the plume haveprovided electron number densities andtemperatures, 9 - 11 and these have been used tomodel the effects of the plume oncommunications signals. 12 , 13 Also, a NASAsponsored program aimed at documentingarcjet/spacecraft interactions is to be completedthis year. 14

Arcjet technology development programshave been successful in bringing the arcjet nearlyto flight readiness and an arcjet propulsionsystem has recently been baselined on acommercial communications satellite. Thereremain, however, questions about the physicalprocesses of the device. For example, althoughthe arcjet performance is significantly above thatavailable with existing technology, arcjetefficiency is in the 30 to 40 percent range.Efforts to increase this have met with littlesuccess. The fundamental energy input and lossmechanisms are not well understood, andsignificant performance gains, if they arepossible, will likely require a betterunderstanding of the fundamental operatingphenomena. Cathode emission processes have

been studied, 15 as have the effects of nozzle

contour. 16 A recent study of current attachmentin the nozzle provided insights into energy input

regions, 17 and extensive efforts have beendirected toward the development of a numerical

model of the arcjet. 18

This report details the results from acontinuing experimental program aimed atproviding insight into arc energy deposition inthe nozzle, the nature of the arc attachment, andits effects on performance characteristics of thedevice. A companion paper details a preliminaryspectroscopic analysis of the arcjet plume and the

effects of electrode configuration. 19 A modulararcjet incorporating a segmented anode/nozzlewas assembled and tested. This arcjet was

similar to one reported in a previous study, 17 butincluded design improvements that allowedextended operation. The segmented nozzle hadthe same critical dimensions as baseline nozzlesused in many previous tests in this

laboratory, 15,16,20 but the isolated segmentsallowed the examination of the effects of thecurrent distribution on both operatingcharacteristics and performance level.

APPARATUS

Arciet Thruster. Figure 1 shows a cutawayschematic of the arcjet thruster used in the testsdescribed in this report. Aside from thesegmented anode, this thruster is identical tothose used in many tests in this

laboratory. 15,16,20 A 3.2 mm diameter, 2percent thoriated tungsten rod about 190 mm inlength was used as the cathode. The tip halfangle was ground to 30 degrees to match theconverging side of the anode. The modifiedcompression-type fitting was used to feed the

cathode through the rear insulator and to lock itin place once the gap was set. This fitting washeld in place by a center-drilled holding bolt.

Propellant entered the arcjet from the side ofthe rear insulator. The propellant tube wasthreaded into a cylindrical anchor located insidethe rear insulator. The axial center bore of thisanchor allowed passage of both the cathode andan insulating alumina sheath. This designisolated the propellant system from the electrodes

The rear insulator, front insulator, andcompression plunger shown in Figure 1 were allmade from high purity boron nitride. An inconelspring was placed between the propellant tubeanchor and plunger to ensure compression at theforward seals (ie. front insulator to injector diskand injector disk to nozzle). Where necessary,stainless steel washers were inserted to adjust thetolerance. Longitudinal grooves were machinedinto both the compression plunger and frontinsulator to allow propellant passage.

A molybdenum propellant injector disk madewith two 0.51 mm injection holes, drilledtangent to the inner surface, was used. The anodehousing was stainless steel.

Graphite foil gaskets were inserted betweencritical sealing surfaces. The anode housing andthe rear insulator were clamped together by twostainless steel flanges connected by four steelbolts. For this application, four additional holeswere drilled in the front flange to acceptinsulating feedthroughs for the bolts needed tohold the anode clamping flange in place.

Segmented Anode. A cross-sectional schematicof the segmented anode is shown in Figure 2(a).The upstream segment was 2 percent thoriatedtungsten and comprised both the converging sideof the nozzle (30 degree half angle) and theconstrictor (Dc = 0.64 mm; Lc = 0.25 mm).This first segment had been run previously, andsome damage to the constrictor was noted beforethe initiation of the tests described in this report.This did not appear to seriously affect theoperating characteristics of the arcjet. In earlierversions of this device, the next three anodesegments were made from molybdenum. Forthese tests, tantalum was used in place ofmolybdenum because of its machining propertiesand higher melting point. A detailed drawing ofone of these segments is shown in Figure 2(b).The thickness of each segment at the nozzlesurface was 1.3 mm and a step was cut in each toalign the stack. The tabs used for electrical

connection were extended and small bolt holeswere drilled in them so that electrical attachmentscould be made with stainless steel bolts. Allboron nitride insulating disks were 0.25 mmthick at the nozzle surface and had steps machinedto mate with those on the conducting segments.A detailed drawing of one of these disks is alsoshown in Figure 2 (b). The downstream endanode segment was 2 percent thoriated tungsten.A molybdenum flange was used to clamp thesegments onto the body of the arcjet. Thisflange was 1.6 mm thick and four insulated rods,discussed previously, held it in place. Fourinconel springs were inserted between theclamping flange and the tightening bolts to allowfor thermal expansion and to hold the segmentstogether tightly during high temperatureoperauon.

When assembled, the segmented nozzle had a20 degree half angle in the diverging section andan area ratio of 225. A photograph of theassembled thruster is shown in Figure 3(a) and aview looking down the nozzle toward the throatis shown in Figure 3(b).

Assembly. Arcjet assembly was complicated bythe fact that the force of the springs on theforward flange could push the nozzle inserttoward the cathode. To avoid this, the interiorspring was stiffened and the compression on theexterior springs was adjusted. Even with theseprecautions, there was some question about theactual cathode/anode spacing when the unit waspressurized. This will be discussed further in theresults and discussion section. The anodesegments comprising the diverging section of thenozzle were disassembled and reassembledoccasionally over the course of the test period toallow insertion of new parts, the rest of thethruster remained assembled in an effort to ensurethat the cathode/constrictor positioning wouldnot change from test to test. The small size ofthe individual anode segments and spacers madeexact positioning difficult. Every effort wasmade to center the segments. It is likely,however, that some variation in the anode stackoccurred between tests due to the assembly andinstallation process. These will be discussed inmore detail in the Results and Discussionsection.

Electrical Configuration. The electricalarrangement used in the testing is shown inFigure 4 along with the anode segmentnumbering scheme. Each anode segment couldbe made conducting or non-conducting. In thecourse of testing it was found that three electrical

configurations covered a wide range of currentdistribution on the anode. These threeconfigurations are shown in Table 1 and wereused almost exclusively in the test matrix.

A pulse width modulated power processingunit (PPU) was used to start and run the arcjet.Manual switches were used to connect anddisconnect the anode segments from the PPU.Separate digital voltmeters and Hall currentprobes were used to monitor the potentialdifference between the cathode and each anodesegment and the current passing through thesegments, respectively. Outputs from each ofthe current probes were taken to an eight channelstrip recorder.

Vacuum Facili!_y. All tests described in thisreport were performed in a 0.91 m diametercylindrical test section piece attached to a 0.91m gate valve. The gate valve was mounted on avacuum tank that was 1.5 m in diameter and 5 mlong. Pumping was provided by four 30,000 1psdiffusion pumps backed by a rotary blower andtwo mechanical roughing pumps. Duringnormal arcjet operation, ambient tank pressureson the order of 0.65 Pa were maintained.

For performance measurements, the arcjetwas mounted on a flexure-type thrust standsituated in the spool piece. This thrustmeasurement device has been described in detail

elsewhere. 21 Small drifts (<0.0002 N) in thethrust zero were observed during testing. Themaximum magnitude was less than one percentof the total thrust measurement and could havebeen due either to thermal effects caused by theincreased temperature of the thruster or by theextra current carrying wires needed to connect allof the segments.

Thermal conductivity type mass flowcontrollers were used to regulate both thenitrogen and hydrogen mass flow rates.

Calibrations and Procedure. Prior to the start oftesting, the current probes were calibrated using acommercially available, laboratory model dopower supply. The thrust stand was alsocalibrated before and after the test sequence usingweights of known mass attached to the thrustercradle on the thrust stand. These could be raisedand lowered using an external actuator.

In all tests a 2:1 hydrogen-nitrogenpropellant mixture was used to simulate fullydecomposed hydrazine. The propellant systemincorporated a calibration tank for in-situ

flowmeter calibrations. Both 3.73E-5 kg/s and4.97E-5 kg/s mass flow rates were used in thetests.

RESULTS AND DISCUSSION

General Comments. As noted in the section onarcjet assembly, the small size and closetolerances of the nozzle segments made exactalignment through the assembly and installationprocess difficult. Also, the high heat loads onthe segments caused some damage to thesegments similar to that described in a previous

paper. 17 This damage mainly took the form oflocalized melting, particularly on segment 2.Because of these test-to-test variations in thephysical condition of the nozzle and the anodeconnections, small differences in the operatingcharacteristics of the device were observed. Forexample, on some startups, the currentdistribution observed in configuration 1 shortlyafter a start was near the steady state distribution,with approximately half of the current collectedon the segment furthest downstream. Duringother tests with the same electrical configurationmuch of the current (> 5 A) was initiallycollected on segment 2, and the redistribution tothe steady state condition took tens of minutes.Despite these complications, stable arcjetoperation was observed throughout most of thetest periods and the trends and conclusionsdiscussed in this report were not significantlyimpacted by test-to-test variations in operatingconditions.

In earlier tests of a similar device, 17 noallowance was made for differential thermalexpansion and this limited the maximumduration of test runs to approximately fourminutes. After this period, operation becameerratic. The design changes detailed in theApparatus section effectively addressed issuesrelated to thermal expansion and steady stateoperation was demonstrated over multi-hourperiods.

Startup and Transition. In most test sequences,the arcjet was started in configuration 1 with thepower supply current preset to 10 A and a massflow rate of 4.97x 10-5 kg/s. As in previoustesting, a brief current surge was seen on the firstsegment upon startup. This starting transient for

the arcjet has been examined previously 17 andwill not be discussed in any detail in this report.Most of the current moved downstream to attachin the diverging section of the nozzle veryrapidly. Following startup, fluctuations wereobserved on the current traces before steady

operation was obtained. Visible fluctuations inthe plume were also observed during this period,often accompanied by some sparks. Similarstarting transients have been observed in many

previous arcjet tcsts. 6 , 16 ,20,21 One possibleexplanation for this behavior is motion of the arcattachment points on the electrodes before thesteady state condition was reached. Examinationof segment 2 after testing always showed damagein the form of some localized melting whichprobably occured at startup. These irregularitiesin the surface may have caused localized currentattachment that persisted for some time and led tomotion of the attachment zone on both the anodeand the cathode. It was also found that byswitching all of the current to the segmentfarthest downstream, into configuration 2,immediately following startup, the fluctuationsdisappeared and stable operation was attained veryrapidly.

In some tests the temperature of segment 5was measured with a two color pyrometer. Itwas found that the nozzle reached thermalequilibrium very rapidly. Over 95 percent of thesteady state value was reached within twominutes of startup. The steady state value ofapproximately 1450 K was 200 - 300 K abovenormal operating temperatures for similar deviceswith non-segmented anodes. This was attributedto reductions in conductive cooling due to theinsulating spacers.

Operating Characteristics and Performance. Aspreviously noted, stable arcjet operation wasobserved in each of the electrical configurationsshown in Table 1. In typical test sequences, thearcjet was started in configuration 1 and run untilsteady state operation was attained. Once thesteady state condition was reached, the arcjet wasswitched to another configuration, data weretaken and the arcjet was returned to configuration1. This sequence was repeated for each operatingpoint and the current/voltagc characteristics werefound to be very repeatable over each test periodindicating that the thruster maintained itsstructural integrity throughout the tests. Thesame philosophy was also used in changingoperating points, i. e. the initial test point (massflow rate, current, and electrical configuration)was retaken at the end of each test period in orderto determine whether any major changes inoperating characteristics had occured during thetest period.

Typical voltage-current characteristics of thethruster running in configuration 1 at the highestmass flow rate are shown in Figure 5. Also

included in the figure is a data set taken fromprevious arcjet tests in which standard nozzleinserts of similar dimensions were used. Fromthe figure it can be seen that the curves are offsetby 25 to 30 volts across the current range. Thisvoltage offset is not fully understood at present.It is possible that this offset was due todifferences in the anode to cathode spacing, or arcgap, between the tests with the solid andsegmented anodes. As noted in the assemblysection, the gapping procedure was complicatedby the springs inserted to allow for differentialthermal expansion in the nozzle. The magnitudeof the impedence change, approximately 3 ohmsat 10 A, makes it unlikely that the difference isdue to added resistance in the extra cabling andconnectors required to operate the segmentedanode. Thrust measurements taken during testingindicated that the noted differences in the current-voltage characteristics did not significantlyimpact the performance characteristics of thedevice.

Table 2 presents data from a typical test runin which the arcjet was operated at constantcurrent and mass flow rate in each of the threeconfigurations. It should be noted that due to thecombined uncertainties in the currentmeasurements for all of the segments, the totalcurrents for each configuration do not add toexactly 10 A in each case. During theexperiment, current was set to 10 A and thissetting was not changed as the configurationswere changed. Thus, 10 A was used to calculatethe power in each case. The data show that thearcjet ran at the lowest operating voltage whenall the segments were connected (configuration1). When all of the current was switched to thelast segment (configuration 2), the overalloperating voltage increased by about 5 volts.This same increase in voltage was observed whenonly the last segment was made non-conducting(configuration 3). As noted in a previousreport, 17 the increase in operating voltageobserved when the arcjet was switched fromconfiguration 1 to configuration 3 suggests thatthe anode fall voltage increases when the arcattaches upstream in the nozzle. The differencein potential between segment 4 and segment 5when segment 5 was isolated was about the sameas observed previously. This supports theconclusion that the anode fall in this region is onthe order of 10 - 20 volts.

The floating potentials of the segmentswhen only segment 5 was conducting provide arough estimate of the regions of energy input inthe device. This is plotted for a typical case in

Figure 6. Only about 40 percent of the totalvoltage drop occurred in the constrictor segment.This indicates that a significant amount of thetotal energy dissipated in the device was input inthe diverging section of the nozzle. The effectsthis had on thrust will be discussed in afollowing section.

To examine the effect of mass flow rate onthe current distribution the arcjet was run at twodifferent mass flow rates, 4.97x10 - 5 kg/s and3.73x 10 - 5 kg/s, in configuration 1 at 10 A. Theflow rate range approximates the worst caseexpected in a blowdown system on acommunications satellite. The current to eachsegment and the current distribution on eachsegment are shown in Figures 7 (a) and (b),respectively. To calculate current densities, itwas assumed that the current to each segmentwas distributed evenly across the interior surfaceof the segment except in the case of segment 1.For this, only the constrictor area was assumedto be conducting, (ie. it was assumed that currentwas not collected in the converging section of thenozzle, on the anode housing which was in directelectrical contact with segment 1, or on thedownstream face of the segment). Calculatedvalues of the current density on the constrictorsegment were high compared to those on theother segments and it is likely that thisassumption was incorrect. This implies eitherthat there was some current collected upstream inthe converging section of the anode or that theactual collecting area in the constrictor wassignificantly larger than calculated due to damagecaused during startup and/or steady stateoperation. Because of the uncertainties in thecurrent collecting area and the fact that onlysmall amounts of current were collected on thissegment, the current densities on this segmentwill not be discussed further. From Figures 7(a)and (b) it can be seen that almost half of the totalcurrent was collected on segment 5 and that thecurrent density profile, excluding segment 1,peaked at segment 2 at the higher mass flow rate.This current distribution changed significantly asthe mass flow rate was lowered by about 25percent to the second flow rate tested. For thisoperating point, the current to the last segmentwas decreased by half, and the largest currentincrease was seen on segment 3. The trendsobserved in this test indicate that the mass flowrate influences the current distribution in thenozzle and that the current distribution can beexpected to change with time in a blowdownsystem.

Finally, the total current to the device wasfound to have little effect on the currentdistribution in the nozzle. To study this, thearcjet was run in configuration 1 with a constantmass flow rate and the current was varied from 6to 10 A in 2 A increments. The fractions of thetotal current that appeared on each segment forthe three current levels used are shown in Table3. From the table it can be seen that there waslittle difference between the 10 and 8 A operatingpoints. While there were some differencesbetween these and the 6 A case, no clear trendwas evident. As there has been recent emphasis

on low power arcjet operation,20 further study atlower currents may be warranted. The uppervoltage limit of the power supply used in thisstudy prohibited operation at lower current levels.

It was also of interest to obtain thrustmeasurements with the thruster operating indifferent electrical configurations in order todetermine whether or not the current attachmentlocation significantly impacted the performanceof the device. In the course of earlier

experiments on segmented anodes, 17 significantchanges in plume characteristics were observed asthe electrical configuration was changed. Forexample, when the arcjet was switched toconfiguration 3 from configuration 1, theemission from the HR line (486.1 nm) becamemore prominent. When the arcjet was switchedto configuration 2 the entire plume became moreluminous. A detailed spectroscopic analysis ofthese changes is presented in a companion

report. 19 Briefly, the differences in visibleplume characteristics indicate that changes in theelectrical configuration result in changes in theexcited state populations of numerous plumespecies. Thus, it was concievable that theelectrical configuration might have some impacton physical processes important to the efficiencyof the device.

Typical performance data for the threedifferent configurations were presented in Table2. As noted earlier, the power input to the devicewas significantly above that observed in previoustests of thrusters with solid anodes. Figure 8shows performance data taken with the segmentedanode plotted along with data from previous tests

of similar arcjets with solid nozzles.6,16,20Specific impulse and efficiency are plotted vesusspecific power in Figures 8 (a) and 8 (b),respectively. From the plots it can be seen thatdifferences between nozzles are not significant.Thus, the use of the segmented anode to simulatea standard solid anode seems valid.

The data shown in Table 2 also demonstratethat the electrical configuration does notsignificantly affect the overall performance of thearcjet thruster. As the arcjet was switched fromconfiguration to configuration slight differencesin the thrust were observed, but these were notstatistically significant. As previously noted, thevisible changes in the plume characteristicscaused by switching the electrical configurationindicated that the excited state populations ofsome plume species are configuration dependent.

A companion report 19 advances the argumentthat the observed variation in intensity withconfiguration was the result of changes in theelectron energy distribution. From theperformance data it is clear that these changes didnot significantly affect the thrust characteristicsof the device. Thus, based on this investigationof a convential nozzle geometry, it appears thatarcjet performance cannot be significantlyimproved by changing the region of currentattachment.

CONCLUDING REMARKS

A segmented anode/nozzle was tested in a 1 -2 kW class arcjet thruster in order to study theeffects of current distribution on the operatingcharacteristics and performance of the device.The effect of current level and mass flow rate onthe anode current distribution was also examined.The electrical configuration was designed toallow current to be collected across anycombination of segments. Both the currentthrough each segment and the potential differencebetween each segment and the cathode weremonitored. Performance measurements indicatedthat this modular arcjet with the segmented anodeadequately simulated laboratory arcjetsincorporating solid anode inserts of similardimensions.

As in previous experiments, the current wasfound to attach diffusely in the diverging sectionof the anode when all of the segments wereconducting. When all but the segment farthestdownstream were isolated, the potentials observedbetween the cathode and the individual anodesegments indicated that a significant amount ofthe power input to the arcjet ( > 50 %) was addedin the diverging section of the nozzle. Thecurrent distribution in the nozzle was found to bedependent on the mass flow rate.

Finally, thrust measurements indicated thatthe electrical configuration does not significantlyaffect the performance of the device. Thisimplies that the changes in the electric field in

6

the nozzle that occur as a result of the changes inthe current distribution do not significantlyimpact the momentum transfer or lossmechanisms in the type of nozzle studied.

REFERENCES

1. Gruber, R. P.: "Power Electronics for a 1-kW Arcjet Thruster, AIAA Paper 86-1507,June 1986, (NASA TM-87340).

2. Hardy, T. L.; and Curran, F. M.: "LowPower do Arcjet Operation withHydrogen/Nitrogen/Ammonia Mixtures,"AIAA Paper 87-1948, June 1987. (NASATM-89876).

3. Knowles, S. C., et al.: "PerformanceCharacterization of a Low Power HydrazineArcjet. AIAA Paper 87-1057, May 1987.

4. Knowles, S. K.; and Smith, W. W.: "ArcjetThruster Research and Technology, RocketResearch Co., Phase I, Final Report, 87-R-1175, Aug. 1987.

5. Knowles, S. K.: "Arcjet Research andTechnology, Phase II, Final Report," NASACr-182276, Rocket Research Co., Redmond,WA (to be published).

6. Curran, F. M.; and Haag, T. W.: "AnExtended Life and Performance Test of aLow Power Arcjet," AIAA Paper 88-3106,July 1988 (NASA TM-100942).

7. Yano, S. E.; and Knowles, S. K.:"Simulated Flight Qualification Test of anEngineering Model Arcjet System,"presented at the 1989 JANNAF Meeting,Cleveland, OH, May 1989.

8. Knowles, S. K.; Yano, S. E.; and Aadland,R. S.: "Qualification and Lifetesting of aFlight Design Hydrazine Arcjet System" tobe presented at the 1990 21st InternationalElectric Propulsion Conference, Orlando,FL, July 1990.

9. Zana, L. M.: "Langmuir Probe Surveys ofan Arcjet Exhaust," AIAA Paper 87-1950,July 1987. (NASA TM-89924).

10. Carney, L. M.: "An ExperimentalInvestigation of an Arcjet Thruster ExhaustUsing Langmuir Probes," Master's Thesis,University of Toledo, NASA TM-100258,December, 1988.

11. Carney, L. M.; and Sankovic, J. M.: "TheEffects of Arcjet Operating Condition andConstrictor Geometry in the PlasmaPlume," AIAA Paper 89-2723, July 1989.(NASA TM-102284).

12. Carney, L. M.: "Evaluation of theCommunications Impact of a Low PowerArcjet Thruster," AIAA Paper 88-3105.July 1988.

13. Ling, H.; et al.: "Reflector PerformanceDegradation Due to an Arcjet Plume,"presented at the 1989 Antenna ApplicationsSymposium, Monticello, IL, September1989.

14. Zafran, S.: "Arcjet System IntegrationDevelopment Program," Final Report,NASA CR- 185266, TRW, Inc., RedondoBeach, CA, (to be published).

15. Curran, F. M.; Haag, T. W.; and Raquet, J.F.: "Arcjet Cathode Phenomena," presentedat the 1989 JANNAF Meeting, Cleveland,OH, May 1989.

16. Curran, F. M.; Sovie, A. L.; and Haag, T.W.: "Arcjet Nozzle Design Impacts,"presented at the 1989 JANNAF Meeting,Cleveland, OH, May 1989.

17. Curran, F. M.; and Manzella, D. H.: "TheEffect of Electrode Configuration on ArcjetPerformance," AIAA Paper 89-2722, July1989. (NASA TM-102346).

18. Butler, G. W.; King, D. Q.; and Kashiwa,B. A.: "Numerical Modeling of ArcjetPerformance," AIAA Paper 90-1471,presented at the 21st Fluid Dynamics,Plasmadynamics, and Laser Conference,Seattle, WA, June 1990.

19. Manzella, D. H.; Curran, F. M.; Myers, R.M.; and Zube, D. M.: "PreliminaryInvestigations of an Arcjet Plume," to bepresented at the 21st International ElectricPropulsion Conference, Orlando, FL, July1990.

20. Curran, F. M.; and Sarmiento, C. J.: "LowPower Arcjet Performance," to be presentedat the 21st International Electric PropulsionConference, Orlando, FL, July 1990.

21. Haag, T. W.; and Curran, F. M.: "ArcjetStarting Reliability: A Multistart Test onHydrogen/Nitrogen Mixtures." AIAA Paper87-1061, May 1987 (NASA TM-89867).

Thrust Isp P/m Efficiency

(N) (s) (kJ/kg)

0.219 450 27,000 35.2

0.220 452 27,900 34.3

0.22D 451 28,000 34.1

Table 1. Electrical configurations. (See Figure 4)

Switch positions

Configuration

1

2

3

(0 = Switch open; 1 = Switch closed)

1 2 4

1 1 1 1 1

0 0 0 0 1

1 1 1 1 0

Table 2. Typical operating characteristics with m = 4.97E -5 kg/s and I = 10 A.

Segment

Voltage, V Current, AConfia-umon 1 2 4 1 2 a 4 5

1 134.0 134.0 134.0 134.0 134.0 0.2 1.1 1.4 26 4.9395 23.1 15.1 18.1 28

2 56.1 93.6 116.8 1245 138.9 0.0 0.0 QO 0.0 10.00.O 0.0 0.0 0.0 5.8

3 1393 139.3 1393 L393 1235 0.4 27 60 12 0.079.0 570 63.7 85 GO

( Note: Second row under current for each configuration is current density, A/cm2).

8

Table 3. Fractional current per segment at three current levels.(Configuration 1; m = 4.97E-5 kg/s)

Fractional Current

ltotal = 6A 8A ] A

Segment #

1 0.0 0.01 0.02

2 0.16 0.12 0.11

3 0.21 0.15 0.14

4 0.17 0.22 0.26

5 0.45 0.49 0.49

r_o nou^TC cnu RACYCTC MI T CUV XXI J INGONEL

PLUNGER v

Figure 1. Cross-sectional schematic of the arcjet thruster.

9

r7= Tantalum 2% Thoriated Tungsten —1 Boron NitrideLLA El E

a) Cross-sectional schematic of the segmented anode.

Tantalum

b) Molybdenum anode segment and boron nitride spacer.

Figure 2. Segmented anode/nozzle schmatics.

10

a) Photograph of assembled arcjet.

9.5 mm

IL

b) View of assembled nozzle - looking upstream from the exit plane.

Figure 3. Photographs of arcjet and nozzle assembly.

11

0

0

1 2

od

4 5

Cathe -----------

Figure 4. Simplified diagram of electrical configuration.

200

180

160 0

> 140aito

120 13

0

100

80o Segmented

60 q Solid

40 -f5 6 7 8 9 10

11

Current, AFigure 5. Arcjet voltage-current characteristics.

12

140

120

100

aitoro

0

80

60

40

Segment number

2 3 4

Figure 6. Arcjet voltage gradient - configuration 3.

13

5

4

00 1 2 3 4 5 6

Segment number

a) Current versus segment number.

3Qc

U2

80

60

00

1 2 3 4 5 6Segment number

b) Current density versus segment number.

Figure 7. Current and current density to each segment - configuration 1.(I = 10 A)

NEU

T

40ba

U

20

14

500

0

- ,p0 0

`4 8:)

O^ o0 0

jQ0o

Q^o

^ 0 v o^o

o^ 0

o^ o Solid

0 n Segmented060

00

450

V)

400

U

350

300

250 -f0

0.50

0.45

0.40

U

0.35`W

0.30

0.25

0.200

10000 20000 30000

Specific Power (kJ/kg)

a) Specific impulse versus specific power.

00^ g o o

00091 ^ 00 O0 S&°$P O °

0 oR "

00o

`^Cp o

00 80

0 Solid

n Segmented

40000

10000 20000 30000 40000

Specific Power (kJ/kg)

b) Efficiency versus specific power.

Figure 8. Solid versus segmented anode arcjet performance.

15

NanNational Aeronautics and Report Documentation PageSpace Administration

1. Report No. NASA TM-103227 2. Government Accession No. 3. Recipient's Catalog No.

AIAA-90-25824. Title and Subtitle 5. Report Date

Performance Characterization of a Segmented Anode Arcjet Thruster

6. Performing Organization Code

7. Author(s) 8. Performing Organization Report No.

Francis M. Curran, David H. Manzella, and Eric J. Pencil E-5643

10. Work Unit No.

506-42-319. Performing Organization Name and Address

11. Contract or Grant No.National Aeronautics and Space AdministrationLewis Research CenterCleveland, Ohio 44135-3191 13. Type of Report and Period Covered

Technical Memorandum12. Sponsoring Agency Name and Address

National Aeronautics and Space Administration 14. Sponsoring Agency CodeWashington, D.C. 20546-0001

15. Supplementary Notes

Prepared for the 21st International Electric Propulsion Conference cosponsored by the AIAA, DGLR, and JSASS,Orlando, Florida, July 18-20, 1990. Francis M. Curran and Eric J. Pencil, NASA Lewis Research Center.David H. Manzella, Sverdrup Technology, Inc., Lewis Research Center Group, 2001 Aerospace Parkway,Brook Park, Ohio 44142.

16. Abstract

A modular, 1-2 kW class arcjet thruster incorporating a segmented anode/nozzle was operated on a thrust standto obtain performance characteristics of the device and to further study its operating characteristics under a numberof experimental conditions. The nozzle was composed of five axial conducting segments isolated from oneanother by boron nitride spacers. The electrical configuration allowed the current delivered to the arcjet to becollected at any combination of segments. Both the current collected by each segment, and the potential differencebetween the cathode and each segment were monitored throughout the test period. As in previous tests a similardevice, current appeared to attach diffusely in the anode when all of the segments were allowed to conduct.Improvements to the device allowed long term (4-8 hour) operation at steady-state and operating characteristicswere repeatable over extended periods. Performance characteristics indicated that the segmented anode reasonablysimulates the behavior of solid anodes of similar geometry. Current distribution depended on flow rate as the arcattachment moved downstream in the nozzle with increases in the mass flow rate. The current level had littleeffect on current distribution on the anode segments. Thrust measurements indicated that the current distributionin the nozzle did not significantly affect performance of the device.

17. Key Words (Suggested by Author(s)) 18. Distribution Statement

Space propulsion Unclassified — UnlimitedElectric propulsion Subject Category 20Auxiliary propulsionArcjet thrusters

19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of pages 22. Price'Unclassified Unclassified 16 A03

NASA FORM 1626 OCT 86 *For sale by the National Technical Information Service, Springfield, Virginia 22161