Results From SERT I Ion Rocket Flight Test

8/8/2019 Results From SERT I Ion Rocket Flight Test

1/43

TECHNICAL NOTE ks>*1 N A S A TN D-2718

FROM SERT IROCKET FLIGHT TEST

RonaldJ. Cybulski, Daniel M. Shellhammer,R. Lovell, EdwardJ. Domino, and Joseph T. Kotnik

A E R O N A U T I C S A N D S P A C E A D M I N I S T R A T I O N W A S H I N G T O N , D. C . M A R C H 1 9 6 5


2/43

N A S A TN D-2718

R E S U L T S FROM S E R T I ION ROCKET F L I G H T T E S TB y Ronald J . Cy b ulsk i , Danie l M . Sh el lham m er, R ob er t R . Lovel l ,Edward J. D o m i n o , and Joseph T. Kotnik

Lewis Research Cente rCleveland, Ohio

N A T I O N A L A E R O N A U T I C S A N D S P A C E A D M I N I S T R A T I O NFor sa le by the O f f i c e of Technical Serv ices , Department of C o m m e r c e,

Washington, D.C . 20230 -- Pr ic e $ 2 .0 0


3/43

R E S U L T S FROM S E R T I ION ROC K ET FLIG HT TESTb y Ronald J .C y b u lski, Daniel M. Shellhammer, Robert R.LovelI,E d w a r d J . Domino, a n d Joseph T . Kotnik

L e w i s Research CenterS U M M A R Y

The SERT I (Space Electric Rock et Test) spacecraft was launch ed at 0553:15 EDT on20, 1964, from Wallops Island, Virginia. Two ion thrustors were programed to

sequentially durin g the flight. One of these, a contact ion thrustor , could not bebecause of a high-voltage electrical short circuit. T he other thrustor, which

electron-bombardment type, operated as expected for a total of 31 minutes16 seconds during the flight. The primary flight objective of confirming neutraliza-of a high-perv eance ion beam was attained. Ion beam thrus t was mea sured with

independent thrust measuring systems. A ll voltages, currents, and other oper-characteristics of the electron-bom b ardment thrustor w ere close to the nominalin previous vacuum tank tests.

Telem etry data indicated no noticeable radio-frequen cy interfere nce generated byio n beam. Furthermore, all commands between th e ground t ransmit ter and space-we re transmitted and executed without interferen ce.

INTRODUCTIONThe SERT I (Space Electric Rocket Test) spacecraft was launched on a 46-minute

ballistic trajectory on July 20, 1964. Two ion thrustors were programed tothe flight; one of these was a contact ion thrustor and theth rus tor . The contact ion thrustor could not be

of a high-voltage electrical short, but the electron-bombardment io nwas operated for a total of 31 m inu tes and 16 seconds. This was the first tim e

an ion thrustor of any type had been operated in space. The performance data of thethrustor system obtained during th e flight are presented herein, and these data are

with similar data obtained in vacuum tank tests of the same io n thrustor sys-The primary objective of the SERT I flight experiment was to determine whether


4/43

high-perveance io n beams can be neutralized in space. T he secondary objectives of theflight were (1) to determine whether any unforeseen dif ferences in ion th rus tor perfor-mance occur in space as com pared with their operation in vacuum tanks, (2) to deter-mine whether different types of neutralizers are effective, and (3) to determine whetherrad io- frequency signals can b e transmitted to and from an ion-propelled spacecraft .

N eutralizat ion of ion beams had be e n effected in vacuum tank tests with various neu-tralizers, all of which used a heated filam ent as a source o f electrons. A flight experi-ment, however , was required to provide conclusive proof of the feasibility of ion beamneutralizat ion because of several uncertainties inherent in vacuum tank tests. These un-certainties exist b ecause of (1) the secondary electrons produce d when the ions strikemetal surfaces, (2) the dilute plasma that fills the vacuum tanks during ion thrusto r ope r-ation, and (3) the possible capacit ive coupling between the ion beam and the vacuum tankwalls.

In planning th e SERT I flight, it was decided that neutralization of the ion be a m scould b e demonstrated most conclusively b y proving th e existence of thrust produced b ythe ion beam s. A ccordingly, the two ion thrusto rs w ere mo unted so that their thrustwould change the spacecraft spin rate, and three independent system s for spin rate mea-surement w ere installed on the spacecra ft. In addition, extensive ion thrusto r instru-mentat ion was included on the spacecraft to perm it a direct com parison b etween flightand vacuum tank test results.

A lengthy d evelopment program w as required before the ion thru stor s, power sup-plies, and spacecraft compo nents w ere made com patib le and sufficiently reliable toassure a high probab ili ty of success in the flight. Du ring this development program ,performance data were obtained that provide the basis for comparison b etween vacuumtank and flight test results. In addit ion to the flight data, this report includes the per fo r -mance data of the e lec t ron -bombardmen t io n thrustor obtained in two flight s imula t iontests that were conducted with th e flight spacecraft in a 15-foo t-diameter vacu um tank atthe Lewis Research C enter .

A P P A R A T U S A N D I N S T R U M E N T A T I O NSpacecraft

Figure 1 shows th e SERT I spacecraft . A description of the design and developmentof this spacecraft can be found in re fe rence 1.

E l e c t r o n - B o m b a r d m e n t Ion Thrustor and P o w e r S u p p l i e sFigure 2 shows a f l igh t-model e lec tron-bom b ardment thrus tor . A detailed descrip-

2


5/43

T A B L E I. - THRUSTOR OPERATING CONDITIONS FOR SER T I FLIGHTComponent

BoilerMagnetic fieldCathodeDischargeN eutralizer heaterPositive acceleration

(screen) potentialN egative acceleration

(accelerator elec-trode) potential

Potential,V ,

volts

0 - 40 ac6 d c9 - 10 ac46 dc9 - 10 dc2500 dc-2000 dc

Current ,J,amperes

0 .60 - 2.6115 - 2621.9 - 2 2 . 92.08 - 5 .4522 - 26

0. 100 - 0. 377I T ~ I ^^ s o u r c e ~ ~ beam'0 .003 - 0.013

Nominaloperating

power,watts

10(as required)

18022025020 0725

10

Operating point

433 to 479 K30 Gfcathode temperature\ set to give 5-A

L dischargeneutralizer ~ beam

2 5 0 0 V-2000 V

of the general thrustor design and principle of operation can be found in referencesto 4. N om inal thru stor operating param eters are shown in table I. Figure 3 is an

identifying the location of pertinent thrustor voltage and currentand the corresponding significance of these measurements. The neutral-

was a tantalum filament heated electrically to a temperature sufficient to yield ae-limited electron curre nt. A ll the flight and vacuum tank test data pre-

the same electron-bombardment thrustor and the samesupplies.

The positive and negative high-voltage power supplies for the thrustor io n sourced accelerator electrode w ere operated from a 56-volt battery with a 28-volt tap.

output transistors in each of these power supplies received pulse-width-controlledfrom drivers and a master oscillator. The positive supply was current regulated,

d the negative supply had voltage and curren t regulation. Cu rrent overload sensors ine output of both of these power supplies turned off all outputs when overloads exceeded

to 20 milliseconds of duration. The overload sensors were set below the currentof the power supplies.

Thrust Detection S y s t e m sThe SERT I spacecraft was designed to measure thrust b y determining changes in

e spin rate and, hence, angular momentum of the spacecraft. The instrumentationhigh accuracy and wide response range in order to assure rapid measurements of


6/43

change in spin rate in the event of deviation from expected thrus tor perform ance or fail-ure of th rus tor or spacecraft after a short interval of t ime.

T he spin rate was m easured by three independent sy stems. Two of these sy stemsw e r e photovoltaic solar-cell spin-period detectors. T he third system used an acceler-ometer mounted so that its sensitive axis w as perpendicular to and intersected th e space-craft spin axis.

Each solar-cell system uti l ized a separate te lemetry l ink . The received pulses werefed into clock-con trolled electronic counters for per iod measurements . For pure spin,that is, no energy t ransfer b e tween axes, the system error in spin-period measurementwas calculated to be less than 0.1 percent for a single revolution.

The output of the accelerometer contained information concerning motion about allthree principal axes. T he ref or e, the accelerom eter provided necessary inform ationconcerning spacecraft precession if mo tion about th e t ransverse axes is large and ifthere is energy t r ans fe r f rom the t ransverse axes to the spin axis.

Hot-Wire CalorimeterFigure 4 shows th e f l ight-m odel hot-w ire calorim eter probe. T he theory of opera-

tion of this instrument can be found in r e f e r e n c e 5. The hot-wire ca lor imeter probe w aspivoted through the ion beam at a location 7 inches down beam f rom th e accelerator elec-t rode to measure th e distr ibution of ion beam pow er dens i ty (see f ig . 1). Pow er dens itycontour maps were cons t ruc ted and graphically integrated to obtain total b eam pow er.From this value of beam power and the measured accelerat ing potentials , a value of ionbeam cur ren t w as then calculated.

E - F i e l d MeterThe E-f ie ld m e t e r was mounted on the contact thrustor control box. It consis ted of

tw o sets of thin gold-plated vanes, 2 inches in diameter . The inner set of th ree vanesw as s ta t ionary and connected to vehicle ground through a precision resistor. T he outerset of three vanes rotated at a speed of 8000 rpm and intermittently shielded th e innervanes. If an electrostatic field were presen t , th e inner vanes would genera te a 400 cpssignal, which was fed into a solid-state alternating-current s ignal amplifier . T he ampli-fier output w as then rec t i f ied and fed into the telemetry signal condit ioning. The ampli-f i e r s w ere ad jus ted to pro vide data on electric fields up to a m a x im u m of 120 volts percen t imete r .


7/43

T A B L E n. - SERT I FLIGHT HISTORY FOR JULY 20, 1964(a ) Programed

Event Programer t ime ,min

Act iva te payload and start programerLift-offContact thrustor switched to full heat modeSeparationEngine dep loymentContact thrustor pod door openingHigh voltage applied to contact thrustorStudy of contact thrustor switched to t rap neutralizerStudy of contact trap neutralizer completedHigh voltage removed f r o m contact thrustorContact thrustor turned offElec t ron-bombardment thrustor turned onThrustor n eutral izer turned onBoiler heating, com ma nd being sent f rom ground

station to open magnetic field coil winding tot r igger thrustor start

Study of hot-wire calorimeter probe startedStudy of neutralizer voltage control startedStudy of neutralizer voltage control completedStudy of hot-wire calorimeter probe started

(neutralizer o f f )Study of hot-wire calorimeter probe started(neutralizer on)

00 : 5 31:595:315:335:355 : 4 0

20:2026:3426:3927:3827:4427:4527:46

39:4441:4347:3347:34

49:34

L e w i s 1 5 -F o o t V a c u u m T a n k FacilityVacuum tank tests of the spacecraft were conducted in a 15-foot-diameter b y 60-foot-

Cchamber a t pressures of the o rder of 5x10" torr. Figure 5 shows the flight space-installed in the vacuum facil ity . The spacecraft was mounted on a spin table capa-spinning at the rate expected during flight, nominally 90 rpm. The spacecraftould b e electrically insulated from ground.

R E S U L T S A N D D I S C U S S I O NThe data to be presented are divided into three parts:(1) The test results of the ion-bo mb ardm ent thrustor system during the 31 minutes

and 16 seconds of operation for the ballistic space flight of July 20 , 1964


8/43

T A B L E n. - Concluded. SERT I FLIGHT HISTORYFO R JU L Y 20 , 1964

(b ) ActualEvent Flight time,

m in afterlaunchSpin-up of payload to ~109 rpmPayload separation, precession dampers released,

Y O system timers activatedBoth thrustors and hot-wire probe deployedE lec t r on -bomDar d men t thrustor turned onThrustor producing thrustStudy of hot-wire calorimeter probe started

(neutralizer on)Study of hot-wire calorimeter probe completedStudy of neutralizer voltage control startedStudy of second hot-wire calorimeter probe started

(neutralizer o f f )Study of second hot-wire calorimeter probe completedStudy of third hot-wire calorimeter probe started

(neutralizer on)Study of third hot-wire calorimeter probe completedThrustor commandoffElectron- bombardment thrustor command on restartSecond thrust period and flight completed

2 : 4 84 : 3 84 : 4 0

13:511 7 : 4 925:52

27:0627:3333:2634:3435:2536:363 6 : 5 138:4246:58

(2) T he test results of the electron-bombardment thrustor system for the two systemtests of the f l ight spacecraft in the 15-foot-diameter vacuum chamber on June 7and June 9, 1964

(3) The comparison of vacuum tank and space-flight test data

Space-Flight T e s t D a t aThe spacecraft programer sequence is given in table n(a), and the actual flight

events are given in table n(b). The early portion of the flight was devoted to an attemptto operate the contact ion thrustor. This was unsuccessful due to a high-voltage elec-trical short circuit. The differences between the planned (table II(a)) and the actual(table n(b)) fl ight events resulted f rom command changes to the programer deemed ad-visable after the high-voltage short in the contact ion thrustor system.

Figures 6 and 7 show photographs of the real-time data recordings of the principalelectron-bombardment-ion-thrustor operating parameters during the fl ight. The ordi-6


9/43

T A B L E in. - E L E C T R I C A L BREAK D O W NS DUR ING S E R T I FLIGHTTrip

123456789

101112131415161718192021222324

Flight time19:4520:1420:4221:4722:3523:4725:2825:5526:5629:0929:1729:3929:4930:1530:1831:2731:4732:1832:2632:4932:5933:2133:2833:34

Operat ing condit ionsConstant voltages

1

Study of neutralizerstep voltage

i '

Trip25262728293031323334353637383940414 24344454647484950515253

Flight time33:4433:5033:5734:0334:1534:2334:3434:4734:5935:4040:0441:1741:2141:2642:1542:2042:4943:0043:0543:2343:3443:5944:2944:3445:4846:2546:3046:3846:53

Operat ing condit ionsNeutralizer off

Constant voltages -second thrustoroperat ing period

^r

n figu res 6 and 7 is the telem etry output voltage. The calibration curves for theare not linear; the re fore , no absolute values are shown on theFigures 8 to 11 show quantitative values of the thrustor parameters monitored dur-

space flight test of the electron-bombardment ion thrustor. The ion-chamberinitiated at 3:50 minutes after the thrustor sy stem w as turned on (flight t ime,

min). The ion thrustor s tar tup was norma l.The e lec t ron-bombardment io n thrustor was turned off by command at a flight t ime

36:51 minutes and was turned on again at 38:42 minutes. In this time interval a sec-nd attempt was made to operate the contact ion thrustor. During the second period of


10/43

operation of the electron-b omb ardment ion thrustor (flight time, 38:42 to 46:58 min), thevapo rizer overheated slightly, which gave higher than nominal discharge and b eam cur-rents as indicated in f igu re 9. This overheating can be attributed to the departure f r omth e planned flight test sequence and the resulting increased total test time on thee lec t ron-bombardment io n thrustor. A thermistor on the propellant vaporizer turnedo f f th e vaporizer heater power toward the end of the flight, b ut heat flux from the ionsource was sufficient to cause the overheating.

Many electrical b reakdow ns that caused autom atic shutdown of the power supply areindicated in the f igu res b y vertical lines superimposed on the curves. T he t ime of oc-currence of each electrical breakdown is shown in table in . No change in telemetry sig-nal strength was observed between those times when the thrustor was operating and thoset imes when it was not operating; it can therefore b e concluded that the ion beam did notin ter fere with radio-frequency transmission f r om the spacecraft .

During the total thrustor operating time of 31 minutes and 16 seconds, there were53 electrical breakdowns in the thrustor system. Eight of these electrical breakdownsoccurred in the initial 11 minutes and 53 seconds of operation. In this time interval thevoltages w ere m aintained constant and the thrusto r gradually w arm ed up to its designoperating conditions. Some electrical breakdowns would be expected in this period be-cause of high o utgassing rates during th e warmup of the thrustor.

One electrical breakdow n resulted in the course of the hot-w ire calorimeter prob estudies for a flight time of 25:52 to 27:06 minutes. A n arc between th e thrustor and theprobe is a likely possib ility , since such arcs have frequently occurred for similar probesurveys in vacuum tank tests.

Fifteen electrical b reakdowns occ urred d uring the neutralizer step voltage transientstudies for a flight t ime of 27:33 to 33:39 minutes. T en electrical breakdowns occurredin the flight time interval f rom 33:26 to 34:34 minutes while the ion thrustor neutralizerw as turned off . N um erous b reakdowns during some port ions of the neutralizer stepvoltage transient study and during th e t ime th e neutralizer was off would b e expected.Elec t rons were not available to neutralize the ion beam when th e neutralizer voltage w astoo high or when th e neutralizer heater current w as turned o ff . U nder these conditionsions could not leave th e thrustor as a beam. Instead, th e ions traveled to nearby low-voltage surfaces and w ere neutralized on contact. The result ing concentrat ion of ionsand propellant gas inside and adjacent to the th rus tor would then b e expected to cause fre-quent electrical breakdown.

The remain ing 19 electrical breakdowns o ccurred in the course of the second oper-ating period f rom a flight t ime of 38:42 to 46:58 minutes (end of flight). These 19 elec-trical breakdowns fo r almost s teady -state operation w ere m ore frequent than w ere usu-ally observed in the vacuum tank test pr o g r a m ; th e probable reason for this was theoverheating of the propellant vaporizer, as will b e discussed in the sec tion Com parison8


11/43

Vacuum Tank an d Flight Test Experimental D ata.Figure 10 shows neutralizer curren t and beam current as functions of t ime. The

o currents are equal, as expected, within th e accuracy of measurement, except attimes of 27:00 and 36:30 minu tes for the hot-w ire calorimeter probe surveys andto 46:58 minutes for the final 8 minutes of thrustor operation. The difference in

and beam current for the hot-wire calorimeter probe surveys can b ein part to ion interception and in part to secondary electron emission from th e

calorim eter probe. These secondary electrons contrib ute to neutralization ofbeam and do not show up as a measured current from th e neutralizer. The differ-neutralizer current and beam current during th e final part of the flight (beyond a

t ime of 38:42 min) can probably b e attributed to the unste ady thrusto r operation.electrical breakdowns occurred throughout this part of the flight, and current

and after each break-The neutralizer and beam cu rrents w ere comm utated measu rements and weremade at the same instant in time.

Figure 10 shows the results of the neutralizer step-voltage transient study, whichas repeated twice. Each study consisted of five discrete bias potentials applied to thewas accomplished b y successively placing five different

of resistance in series with th e neutralizer filament. The voltages applied to thefilament for the first half of the study w ere 190, 360, 6 50, 790 , and 1030

to the spacecraft. N eutralizer fi lament potentials of 190 and 360resulted in stable ion thrustor operation, which is indicative of effective ion beamThe neutralizer and beam currents decreased between 11 and 14 percent

these tw o neutralizer voltage steps (see fig. 10). A t the three highest neutralizerthe thrustor became unstable; frequent electrical breakdow ns occurred and the

electrode curren t increased rapidly each time the ion source current rosezero. This behavior is typical of an ion thrusto r operation without adequate neu-

of the beam. For the second half of the study wh ere th e neutralizer potentialas varied, the voltages applied to the neutralizer were 210, 375, 605, 886, and 1065to the spacecraft. T he same phenom ena occu rred as during the

half of the study; that is, the thrustor operation was stable at the two lower volt-the neutralization an d beam currents decreased about 12 percent, and the thrustoras unstable at the thr ee highest voltages. Re asons for this decrease in neutralizer andn beam currents are discussed in the section Comparison of Vacuum Tank and Flight

Figure 11 shows the ratio of accelerator current to ion beam current as a functiont ime. The accelerator current increases monotonically, as expected, because of

e exchange. Figures 10 and 11 indicate the beam current and accelerator currentzero during th e t ime the neutralizer was turned of f; how ever, close inspection of

9


12/43

th e data of f i g u r e s 6 and 7 indicates that the ion source and accelerator currents in -creased mo men tar i ly each t ime the anode-cathode discharge occurred for a br ie f inter-val. The tim e intervals are so short that they are not shown in f ig ure s 10 and 11. Forth e t ime interval when elec t rons were not available to neutralize the ions, the ions couldnot leave the thrustor as a b eam . Instead, the ions w ere drawn to the accelerator elec-t rode and possibly other metal surfaces, which caused automatic shutdown of the powersupply. This is expected to occur when the ion b eam is not neutralized.

Figure 12 shows the ratio of accelerator current J. to ion beam cu r ren t JR plot-ted as a function of

Reference 6 indicates that for an ion-chamber discharge voltage of 46 volts , 6. 5 per-cent doubly charged ions (13 percent of the current) can be expected. Allowing for thesedoubly charged ions produces a 4-percent reduct ion in thrust calculated f rom cur ren t andvoltage measuremen ts .

Io n beam spreading causes a reduction in thrust proport ional to the cosine of the ionbeam spreading angle. The ion beam curren t prof i le as a function of radius w as obtainedf rom the four hot-wire ca lor imeter probe surveys . I t was then assumed that th e sameprofi le existed at the exhaust of the ion th rus tor . T he average beam spreading angle w ascomputed b y using this assumption. T he result ing average reduction in thrust using thisapproximate method w as about 1 percent.

Figu re 14 is a plot of the delive red thrus t as a function of t ime as measured by thetw o sun sensor sys tems and the accelerometer sys tem. T he thrus t as computed f romthese three measurements agrees to within a few percent. The thrust increased mono -tonically with the ion beam curren t . Throughout each of the ho t-wire ca lor imeter probesu rveys at flight t imes of 25:52 and 35:25 minu tes th e thrust measured by the sun sensorand accelerometer sys tems decreased. The hot-w ire ca lor imeter probe projec tedfrontal area is 3. 5 square inches. When a beam spread angle which yields a 1-percentthrust reduction is assumed, th e pr o be body intercepts 27 percen t of the ion beam atth e midpoint of its traverse; consequently, a thrust reduction of approximately 27 per-cent would b e expected at that moment in t ime. The measured thrust reduction was20 and 24 percent , respect ive ly , for these surveys. The nonuniform current densitydistr ibution f rom the thru stor mig ht accoun t for this small discrep ancy . The sun sen-sor and accele rom eter data show no measurable change in spin rate (conseq uently , nothrust) for the portion of the flight when the neutralizer was turned of f ; this confi rmsth e results of all other m easurem ents during this t ime period.10


13/43

In figure 15, the thrust calculated from current and voltage measure m ents is com -with the thrust calculated from change in spacecraft spin rate as m easured b y the

n sensors and accelerometer. In the discussion that follows, "measured thrust" isas the value computed from changes in spacecraft spin rate; "calculated thrust"

defined as the value calculated from measurements of ion beam current and net accel-doubly charged ions and b eam spreading. The calcu-

thrust is about 25 percent above th e measu red thrust throughout most of the flight.or example, at flight t imes of 25:00 and 43:00 m inutes the calculated thrust is 22. 3 and

.1 percent, respectively , above the measured thrust. T he measured thrust i s be-accurate within 1 percent. This accuracy is b ased on the allowance for

of spacecraft momentum from precession into spin, th e estimated accuracy ofe measurement of spacecraft moment of inertia and moment arm of the thrustor aboute spin axis, and the accuracy of transmission and recording of the digital data from th en sensors. It must then b e concluded that the calculated thrust is erro neo usly high b y25 percent. This error is attributed to a backstreaming electron current

at the neutralizer, following some path around th e outside of the thrustor,d terminat ing at the high-voltage ion source. The solution to the problem of back-

in all vacuum tank tests of ion thrustors has been to completelyth e high -voltage portions of the thrustor with a screen. The flight thrustor had

a screen b iased at spacecra ft potential. The screen can b e seen in figure 2. Thewas tested in a small vacuum tank prior to flight and was found to be completely

Ho wev er, additional tests conducted subsequent to the flight showed th e screenb e inadequate in a larger vacuum tank. It can, therefo re , b e concluded that th e screenas also inadequate to prevent b ackstream ing electrons during flight. The details of thetank tests of this screen are further discussed in the section Comparison ofTank and Flight Test Data.

A s shown in figure 15, for those portions of the neutralizer step voltage transientwhere stable thrustor operation was obtained (flight t imes of 27:41 to 27:57, 28:13

28:29, 30:55 to 31:11, and 31:27 to 31:43 m in), the calculated thru st decrease d 17. 2,9, 12. 3, and 18.1 perce nt. For the same four time intervals, the me asured thrust

3. 5, 6. 5, 3. 2, and 4. 3 percent, while the calculated thrust was 2. 5 to 11. 0the measured thrust.The expected thrust reduction for the previous four t ime intervals was 4. 0, 7. 4,

2, and 7. 8 percen t, respectively . Th ese values are calculated from the change in netpotential, that is, the ion source potential m inus the neutralizer potential.

he reductions in measured thrust are in excellent agreement with th e theory except fore last of the four t ime intervals discussed. A n electrical breakdown had caused power

shutdown at a flight t ime of 31:27 minutes, and the thrustor was not operatingapproximately th e f irs t half of the neutralizer voltage step that was applied over

11


14/43

th e flight time interval from 31:27 to 31:43 m inutes. The time available to measurethrust during this voltage step may have been too short to yield accurate data.

The reduction in calculated thrust during each of the four time intervals is due onlyin part to the red uce d net acceleratin g potential discussed p reviou sly . M ost of the red uc -tion in calculated thrus t is due to the reduced apparent ion b eam cu rren t (see fig. 10).The reduced apparent ion beam current is, in turn, due to the elimination of the back-streaming electron current when th e engine screen was biased at negative potential rela-tive to the neutralizer.

From th e foregoing discussion, it is apparent that th e neutral izer and ion beam cur-rents we r e in error at all t imes throughout the flight except for the four brief intervalswhen steady thrustor operation was obtained with the neutralizer biased at a positivevoltage relative to the spacecraft. Hence, it is only for these four time intervals thatagreem ent can b e expected b etween calculated and m easured thrus t. A s noted pre-viously, the calculated thru st w as 2. 5 to 11. 0 percen t above the measu red thrust forthese four intervals.

The spacecraft spin rate increased from 85 to 94 rpm during th e flight due to ionthrust. The data obtained from th e accelerometer (ref . 1) showed th e spacecraft angularmotion to b e almost pu re spin. The total spacecraft ang ular m om entu m in precession atth e start of ion thrusto r operation was less than 1 percen t of the total angular m om ent umimparted to the spacecraft by the ion thru stor. The total spacecraft ang ular m om entu min precession decreased gradually throug hou t th e flight, and at the end of the flight it wasapproximately one-half it s initial value.Figure 16 shows th e recordings of the five hot-wire calorimeter probes for the threeseparate survey s. In each survey , the probe was moved across the ion b eam and thenback to its original position, thu s prod ucin g two sets of data for each survey . The sec-ond survey was with th e neutral izer turned off and verifies that no ion beam existed atthat time . The ion b eam power density profiles obtained from the hot-w ire calorimeterprobe during the f irs t an d third surveys are shown in f igures 17 and 18, respectively. Avalue of total ion beam power was obtained b y graphical integration of the data in thesefigures. In the first survey th e values of total b eam power were 492 and 482 watts forthe first and second probe sweeps, respectively. In the third prob e survey , the valueso f total beam power were 618 and 643 watts .

Figure 7 shows a photograph of the real-time oscillograph trace of the output of therotating vane electric field m eter. The oscillations in the E- field m eter output throug h-out thrustor operation are at the beat frequency resulting from the arithmetic differencebetween the te lemetry comm utating rate (120/min) and the spin rate of the spacecraft(85 to 94 rpm ). This beat frequency varies between 35 and 26 cycles per minute duringth e flight. It can, t he ref ore , b e concluded that the apparent E-field varies with a fre-quency equal to the spacecraft spin rate. If the E-field m eter readings are interpreted12


15/43

o truly indicate a surface electric field strength, the minimum and maximum portions ofe E-f ie ld data are 4 and 15 volts per centimeter, respectively. These values of E-fie ld

t rength, for reasonable assumptions of electron density and temperature, indicate apacecraft potential between 40 and 150 volts. There are, however, other possible ex-

for the observed behavior of the E-fie ld meter. These include photoemissionthe vanes of the meter and interaction between the moving spacecraft and the am-plasma.

For the two lowest neutralizer voltage steps (190 V at 27:41 to 27:57 min and 360 Vt 28:13 to 28:29 min), the E-f ie ld meter output ceased it s oscillation and gave an appar-

ent E-f ie ld magnitudeof 2. 5 and 3.3 volts per centimeter, respectively. Similar resultsobtained when the two lowest neutralizer voltage steps were repeated (210 V at

30:29 to 30:39 min and 375 V at 30:55 to 31:11 min). Again, th e E-fie ld oscillationseased; th e indicated field strength was 2. 5 and 3.1 volts per centimeter, respectively.

The meter does not provide a direct indication of the polarity of the E-field. It maye significant that the E-f ie ld oscillations ceased for the time intervals when the neutral-zer w as biased at a positive potential; th e presence or absence of E-f ie ld oscillations

may indicate a difference in polarity. N o definite conclusions can be made regarding thesignificance of the E-f ie ld data until th e data analysis and fur ther tests of a rotating-vaneE-field meter are completed.

Vacuum T a n k T e s t D a t aThis section will report electron-bombardment thrustor performance data obtained

n two flight simulation tests that were conducted with the flight spacecraft in the 15-foot-vacuum facility at Lewis Research Center on June 7 and June 9, 1964.

Figures 19 to 22 show thrustor data obtained in the vacuum tank test of June 7, 1964.The io n chamber discharge was initiated at 3:30 minutes. N o hot-wire calorimeter probesurveys or voltage step transient studies were attempted in this test. The ion beam cur-rent reached 190 milliamperes (fig. 21); this low ion beam current was the result of alow setting on the vaporizer temperature control. Figure 23 shows the thrust calculatedfrom the ion beam current and net accelerating voltage with no corrections for doublycharged mercury ions and ion beam spreading.

Figures 24 to 27 show thrustor parameters for the test run on June 9, 1964. A lloperating parameters were normal (see table T). The ion chamber discharge w as ini-

at 4:45 minutes. The ion beam current reached 280 milliamperes (fig. 25).During the step voltage transient study with the spacecraft isolated from ground, the

ias potentials on the neutralizer were 196, 360, 630, 830, and 994 volts with respect toth e spacecraft. Figure 27 shows that neutralizer potentials of 630, 830, and 994 volts

13


16/43

resulted in an abrupt increase in the accelerator imping emen t curre nt; this would be ex-pected for an unneutral ized ion beam. For neutralizer potentials of 196 and 360 volts,th e accelerator impingem ent current did not change. These results are in good agree-ment with the flight test results.

Figure 28 shows the thrust calculated from the ion beam current and the net acceler-ating potential with no corrections for doubly charged mercury ions and ion beam spread-ing.

Figure 29 presen ts the hot-w ire calorimeter probe data. The graphically integratedio n beam power obtained from th e contour maps is 326 and 375 watts for the f i r s t andsecond probe sweeps, respectively .

In th e vacuum tank tests, no quantitative thrust data could b e obtained from eitherthe sun sensor or the accelerom eter sy stem. The sun sensor and accelero me ter sys-tems were both functionally checked out durin g these tests.

C o m p a r i s o n o f V a c u u m T a n k a n d Flight T e s t D a t aFigures 17, 18, and 29 provide a comparison of beam spreading in flight and in a

vacuum tank test. The beam current in flight was 44 percent higher than in the vacuumtank tests. Nevertheless, the ion beam periphery, as determined by the hot-wire calo-rimeter probe, was at almost identical locations for the two tests. This result indicatesthat substantially all of the ion beam spreading is due to ion optics and not to the effect ofspace charge on the ion beam.

A s noted earlier, seven electrical breakdow ns occurred during th e f i r s t 11 minutesan d 53 seconds of the flight, and 19 electrical b reakdowns occu rred during the last8 minutes and 16 seconds of the flight. In the course of these tw o time intervals therewe re no prob e survey s or n eutralizer voltage changes that might cause the electricalbreakdowns. For comparison, there were no electrical breakdowns for the vacuum tanktest of July 7. There was a total of three electrical breakdowns for the vacuum tank testof June 9, but all three of these occurred for the neutralizer voltage study when th e neu-tralizer was biased at voltages m or e than 400 volts above spacecraft potential.

Extensive testing in vacuum tanks has demonstrated that io n thrus tors which havenot been operated for several weeks will exhibit frequent electrical breakdowns for sev-eral minutes to an hour , and there after operate with few , if any, electrical breakdowns.The phenomenon has always been attributed to the removal of adsorbed gas and dust fromth e high-voltage electrodes. The flight thrustor had not been operated for 41 days priorto operation in space. Therefore, th e frequent electrical b reakdowns that w ere encoun-tered in the flight we re expected. Co nve rsely , since th e thrustor had b een operated onJune 6, the absence of electrical breakdowns in the two vacuum tank tests on June 7 andJune 9 was also expected.14


17/43

-Thrustor

Screen

Moveable plate f Raised-'\Lowered - Thrustor mounting plate

Another probable cause of the frequent electrical b reakdowns during th e later portionthe flight is the high vaporizer tem perature for this portion of the flight as comparedthe vaporizer tem pera ture in vacuum tank tests (figs. 8 and 23). This high er vapo r-temperature, for fixed thrusto r conditions, w ould b e expected to produce a lower

efficiency. Figure 12 shows th e ratio J^/Jg is increasing atthan a linear rate with increase in Jg. This is an indication of a reduction inefficiency with increase in vaporizer temperature and beam cur-

A n increased frequency of electrical breakdowns is typically observed for vacuumtests where a low utilization efficiency is obtained; this is attributed to the in-

densities of neutral propellant atoms and charge exchange ions in the vicinity ofparts of the thrusto r.

To check th e adequacy of the thrustor screening for preventing b ackstreaming elec-a f light-mo del electron-bom b ardment ion thrustor was installed in the 15-foot-

NASA Lewis Research Ce nter. These tests w ere conductedth e flight test. The largest opening in the thrustor screens was around th e edges of

e thrustor mounting plate (see above sketch). A small moveable plate was thereforeto the sys tem as shown in the sketch. T his plate could b e raised to b lock most of

e opening b etween th e thrustor screen and the mounting plate, or it could b e loweredo provide essentially the same screen arrangement that w as used in flight. With theat cond itions giving an indicated 3 80-m illiampere io n beam, th e plate

as raised to observe the reduction in indicated b eam cu rren t, and the thrus tor screensthen biased at nega tive voltages to further reduce the indicated beam current. T he

are tabulated in table IV in the order in which the tests we re conducted.For these tests, a low-mass conical thrust target w as used to indicate relative

of thrust. During all of the tests there was no change in the thrust target read-which provided further evidence that only backstreaming electron current was

15


18/43

TABLE IV. - THRUSTOR SCREENING RESULTS FOR

PREVENTING BACKSTREAMING ELECTRONS

Indicatedbeam

current,mA38036 033 032032032036 0380

Test conditions

Plate lowered; no screen voltagePlate raised; no screen voltagePlate raised; - 100 V on screen and platePlate raised; -300 V on screen and platePlate raised; -400 V on screen and platePlate raised; -450 V on screen and platePlate raised; no screen voltagePlate lowered; no screen voltage

TABLE V. -COMPARISON OF RESULTS OF IONBEAM FOR

BOTH FLIGHT AND JUNE 9TANK TEST

Type of test

FlightFlightFlightFlightVacuum tank

(June 9)Vacuum tank(June 9)

Probesurvey

11331

1

Sweep

12121

2

aPVp,W4924 8 261(T643326

375

bpPJ'W53755 6632656444

471

cp*>W46146 6552563

VPJ0 . 9 20.870.980.980 .74

0.80

Pp/PF

1.071.031.121.14

aTotal io n beam power f rom graphical integration of beam pro-f i les obtained f rom hot-wire calorimeter probe data.

Total io n beam power calculated from measured io n beam cur-rent and net accelerating voltage with a 16-percent correctionfor backstreaming electrons.

cTotal io n beam power calculated from measured thrust and netaccelerating voltage.

16


19/43

altered and no change in t rue io n beam current was occurr ing. The 60-mil l iamperein indicated beam current is 16 percent of the original indicated b eam c urre nt.

are in reasonable agreement with th e flight data of figure 10, where a reduc-is shown in neu tralizer cu rren t and apparent ion b eam current of about 12 percentr th e four brief time intervals when the neutralizer was biased at a positive potential

to the spacecraft (and screen).A 16-percent error in indicated io n beam current due to backstrearning electrons can

e assumed for the flight data. This error should exist throughout th e flight, except fore brief time intervals when the neutralizer w as biased at a positive voltage. Co rrectio nr th e assumed error in beam current results in a reduction in calculated thrust equal to

percent of the m easured thrus t. Since th e calculated thrust was initially about 25 per-above the measured thrust , th e previous correction gives agreem ent between the two

within about 5 percent. A 5-percent error in sensing and recording th ecurrent is possible.A comparison of the ion beam power obtained b y different methods in both the flight

d the June 9 vacuum tank tests is shown in table V .The flight data show good agreement b etween th e hot-w ire calorimeter probe and

methods of measurement . In the vacuum tank test, th e beam power obtained withe probe was between 70 and 80 percent of the power calculated from current and voltage

this result is typical of similar probe studies in vacuum tanks.

S U M M A R Y OF R E S U L T SThe following major results were obtained from the SERT I flight of July 20 , 1964:1. The first successful flight test of an ion thrustor was achieved. The electron-

ion thrustor was operated for 31 minutes and 16 seconds during the flight.he contact io n thrustor could not be operated b ecause of a high-voltage electrical short

2. Effective neutralization of the ion beam was obtained; this was the primary objec-of the flight.3. No ma jor differences we re observed b etween io n thrustor performance in space

vacuum tanks.4. Only one type of neutralizer was tested during th e flight, but a significant varia-in its effectiveness was obtained b y biasing the neutralizer at various positive poten-relative to the space craft. Th is particular ne utralizer rem ained effective up to

of approximately 400 volts; it was ineffective at higher voltages. Th is samewas obtained in similar neutralizer voltage studies in a vacuum tank; it is there-

that meaningful beam neutralization experiments with other type neutral-17


20/43

izers can be conducted in vacuum tanks.5. There was no indication of the ion beam causing radio-frequency interference

with signal transmission to and f rom the spacecraft.6. All instrumentation functioned during the flight. Excellent agreement was ob-

tained with two independent sun sensors and a radial accelerometer for measuringchanges in spacecraft spin rate. Values of ion beam thrust determined f rom change inspacecraft spin rate have an estimated accuracy of 1 percent. Ion beam thrust calcu-lated f rom ion beam current and voltage measurements agreed within 5 percent with theother thrust measurements after corrections were made for backstreaming electrons,doubly charged ions, and beam spreading.

7. Measured ion beam current and neutralizer current were in good agreement dur-in g those portions of the fl ight when the thrustor was operating near design conditions.On the several occasions when the ion beam was turned off, the neutralizer current alsoimmediately dropped to zero.8. Frequent electrical breakdowns caused automatic shutdown of the power supplyduring the f l ight. However, th e frequency of these electrical breakdowns did not exceedthat expected in vacuum tank tests under similar conditions.

9. Spacecraft potential was indicated by a rotating vane electric field meter. Atime-varying E-field was indicated with a frequency equal to the spacecraft spin rate andan amplitude of 4 to 15 volts per centimeter. Doubt remains as to the interpretation tobe placed on the E-field data; fur ther analysis is under way.

10. Ion beam spreading was measured at a location 7 inches down beam f rom the ac-celerator electrode by means of a 5-sensor hot-wire calorimeter probe. The amountofbeam spreading was almost identical with that observed in vacuum tank tests.

11. The spacecraft angular motion consisted almost entirely of rotation about the de-sign spin axis. The total angular mome nt um about transverse axes (precession) wasless than 1 percent of the total angular mome nt um imparted about the spin axis by ionbeam thrust.Lewis Research Center,

National Aeronautics and Space Administration,Cleveland, Ohio, January 13, 1965.

R E F E R E N C E S1. Gold, Harold; Rulis, Raymond J. ; Maruna, Frank A. , Jr.; and Hawersaat,

William H .: Description and Operation of Spacecraft in S E R T I Ion ThrustorFlight Test. N A S A TM X-1077, 1965.

18


21/43

2. Kaufman, Harold R.: A n Ion Rocket With an Electron-Bombardment Io n Source.N A S A T N D - 5 8 5 , 1961.

3. Reader, Paul D.: Investigation of a 10-Centimeter-Diameter Electron-BombardmentIon Rocket. NASA TN D-1163, 1962.

4. Kerslake, William R.: Accelerator Grid Tests on an Electron-Bombardment IonRocket. N A S A T N D - 1 1 6 8 , 1962.

5. Baldwin, L. V.; and Sandborn, V. A . : Theory and Application of Hot-Wire Calorim-eter for Measurement of Ion Beam Power. Progress in Astronautics and Rocketry.Vol . 5 - Electrostatic Propulsion. Academic Press, Inc., 1961, pp. 425-446 .

6. Milder, Nelson L.: Comparative Measurements of Singly and Doubly Ionized MercuryProduced b y Electron-Bombardment Io n Engine. NASA TN D-1219, 1962.

19


22/43

C-70027Figure 1. - SERT I flight-model spacecraft.

C-68407CS-31814F i g u r e 2. - Flight-model electron-bombardment thrustor.

20


23/43

Boiler

V2

^ 5

V

]JA lagnet

v,J

>- Battery

Screen -anode

CathodenV4 V31

1i

^-Thrustor frame

r A C Oi

Neut i

ve1

Herator

a l izer-X\j7

V

r


24/43

W i re

C-71115

Figure 4. - Fl ight-model hot-wire calorimeter probe.

F igu r e 5. - S E R T spacecraft in Lewis Research C enter 15- foot-d iameter vacu um tank .

22


25/43

OLUJro1ag1rqCOo03D

TO"-oC1

~Z==

Co;Ccu

O>i_

cO

Co2

oa


26/43

0CCC"8t.czIS0C

'O|ofcl

O

-QS*C~


27/43

=1

SS88

9V4V

*

-1

25


28/43

Arccurentamp

0

P

g

ziAi-r /(s_i

k \e-&(X /oiw*o_s03

3


36/43

Angle of sweep80

( a ) First half of first probe survey.A n g l e o f s w e e p

30

20

4 3Wire number( b ) Second half o f first probe survey.

Figure 17. - P o w e r density distribution obtained with hot-wire calorimeter probe during f irst survey. S E R T Iflight.


37/43

Angle o f s w e e p1in.

5 4 3 2 1( a ) First half of th ird probe survey.

Angle o f s w e e p,1 in.

5 4 3Wire number( b ) Second half of third probe su rvey.

Figure 18. - P o w e r density distribution obtained with hot-wire calorimeter probe dur ing third survey. S E R T Iflight.

3 5


38/43

Vzemaue

V

v

.pc

JS

o

ro

c

0O0

C

////7

-tf

S>tart (hrust> e r i o d

~ ~

) fE n d o fthrnctinrusiperiod-^. -

) 2 4 6 8 1Run time, minFigure 19. - P ropellant vaporizer temperature. Tank testof 6-7-64.

2 4 r- 300

2 0 250

E 16O

o>1 12018

200

150auE

rau3 10 0

4 5 0

L o

r-1C

5tP

1,/

tart ohrusteriod

t_ C

,!f

i

/-

j

/S _

i

S _

t -^-

i C

-, -r

X

t.

r-

yc1

E

I n d othrusperioc

nd of

D0\^5

I

run

Arc currentB e a m currentC a t h o d e heatecurrentArc voltage

L O OS1~ 50

r 1 s \ , - * i *< j * S "^2 4 6 8 1 0Run time, min

Figure 20. - Ion thrustor voltage and currents . Tank test of 6-7-64.36


39/43

250i

200-

-_i_N

-rai

D O

150 - 3

100

3UreL_^Xu

5 C i 0

Startthrusperio

oVo

2o ftIJ

B e a m currentNeutralizer current -A c c e l e r a t o r current

Il\//.///

S_\r^

e^

X^E n d o lthrusperiod

r-c

^

I-x

.

10Run time, minFigure 21. - B e a m , neutralizer, and a c c e l e r a t o rcurrents. T a n k te s t o f 6-7-64.

Ratooaeaoceo

beamceJA/JB

f~

C

C

c

i1

i*

^5o

ro

C //i/ o

6 8 1R u n time, minFigure 22. - R a t io of a c c e l e r a t o ri m p i n g e m e n t current to ion b e a m

current. T a n k t e s t o f 6-7-64.37


40/43

ts ~ 3

2 4 6 8 1 0R u n time, m inF igure 23. - Ion beam thrust calculated from

beam current and net accelerating voltage.Tank test of 6-7-64.

V

zemaue

V>OJ

.o

Ig8 //J

/1 2 4

11

/"

/ Start of" thrust

periodE r idofthrustperiod-

6 8 10 12 14 16R u n time, min 1

F igure 24. - P ropel lant vaporizer temperatu re. Tank test of 6-9-64.


41/43

Acachheceam

O

^orocO

XJU

250

2001=

-|1500SCO inn

50

0

. & & & 1o. B e a m currentCathode heater c u r rA rc current

Ar c voltage

Starthriperit

lo fstxl &

/,/

ent

://

v*"*

/^

'

. -^> -^ ,

r \ ^.

- '6(

-

a

-

i

ftS

Jeutral izer !potential, -v 1 ,lJ po-4-1

rl r ;rustriod

> 1UUC P

o '0


42/43

00

Results From SERT I Ion Rocket Flight Test

Documents

Transcript of Results From SERT I Ion Rocket Flight Test