[American Institute of Aeronautics and Astronautics 36th AIAA Aerospace Sciences Meeting and Exhibit - Reno,NV,U.S.A. (12 January 1998 - 15 January 1998)] 36th AIAA Aerospace Sciences

Copyright© 1997, American Institute of Aeronautics and Astronautics, Inc.

GONDOLA DESIGN FOR VENUS DEEP-ATMOSPHERE AEROBOT OPERATIONS* £

Matthew Kuperus Heun, Jack A. Jones and Jeffery L. HallJet Propulsion Laboratory

California Institute of TechnologyPasadena, CA 91109

AbstractAerobots are balloon-based AEROnautical

RoBOTS with autonomous navigation capabilities. Anaerobot mission has been proposed for exploration ofthe upper atmosphere through near-surface regions ofVenus. The wide range of atmospheric conditions fromthe relatively benign upper atmosphere to the hot,high-pressure surface require thermal protection of thescientific instruments, and the mass constraints of aballoon system require that the thermal protectionsystem be lightweight. To meet the thermal controlchallenges, we propose use of lightweight vacuumdewar technology combined with a phase-changematerial (PCM) thermal damper. For the proposedaerobot mission, the total thermal control mass isestimated to be 7.1 kg out of a total gondola mass of25 kg. The design allows 7.3 kg for scienceinstruments and communication hardware. Fourkilograms of PCM are required to provide a repeatable15-hour mission sequence that includes a 3.5-hourdescent to the surface, 1 hour of surface operations, a2.5-hour ascent to the cooler upper atmosphere, and 8hours to refreeze the PCM.

Lastly, aerobot system design tradeoffs arediscussed, and the extension of vacuum dewar/PCMtechnology to outer planet probes is briefly explored.

IntroductionIn 1985 the Vega/Venera program, a

French/Russian/US initiative, sent two balloons andtwo landers to Venus. The balloons floated for 1-2days at 54 km and concentrated primarily onatmospheric science. The landers made a 60-mmutedescent to the surface where measurements were takenfor about 2 hours. The landers focused on surfacegeoscience at single sites and had limited operational

life due to the harsh surface environment at Venus. Nosubsequent in situ surface or atmospheric explorationsof Venus have been attempted despite a strongscientific motivation to improve our understanding ofVenusian volcanism, tectonic processes andatmospheric dynamics.' These issues would be ideallyaddressed by a vehicle with the capability for (a)repeated descents to the Venusian surface, (b)autonomous navigation toward specific targets ofscientific interest, and (c) long-term in situatmospheric measurements. One such vehicle is anautonomously navigated aerobot (AEROnauticalroBOT) consisting of a instrumented gondolasuspended below a zero-pressure balloon filled with aphase change buoyancy fluid. This concept is describedbelow. Although there are many aspects of Venusianaerobot design, this paper focuses on the structure andthermal control strategies for the instrumented gondolato enable its survival during repeated descents to thesurface.Aerobots

Aerobots with surface descent capabilities providea unique opportunity to advance our understanding ofthe surface and atmosphere of planets and moons withsignificant atmospheres. In particular, Venus, Earthand Titan have extensive tropospheric atmospheresthat allow the use of balloons containing phase-changefluids (PCF) to provide altitude control capabilities.2By using knowledge of wind headings at variousaltitudes and on-board computer-based path planningalgorithms, PCF balloons offer the possibility of zero-power autonomous navigation through theseatmospheres. The only power required is for on-boardelectronics and scientific instrumentation.

Altitude control requires a PCF that exists as avapor at wanner, low-altitude conditions and as a

Senior member, AIAA.

1American Institute of Aeronautics and Astronautics


liquid at cooler, high-altitude conditions. The aerobot'svertical motion is therefore caused by buoyancymodulation associated with balloon volume change asthe PCF vaporizes or condenses. When mostlyvaporized, the PCF provides enough buoyancy to movethe aerobot upwards; when mostly condensed, the PCFprovides insufficient buoyancy and the aerobot dropsdownwards. Note that this strategy only works in thetropospheric part of the atmosphere where thetemperature decreases with altitude.

Figure 1 shows a Van't Hoff (vapor pressure) plotof several candidate PCFs along with the pressurecurve for the Venusian atmosphere. The point at whichany PCF line crosses the atmosphere line is theequilibrium altitude for an aerobot employing thatparticular PCF. When the PCF is above its equilibriumaltitude it is a liquid, and when the PCF is below itsequilibrium altitude it is a vapor. It is seen from theplot that there are many possible fluid choicesdepending on the details of the aerobot and missiondesign. For example, a mixture of water (H2O) andammonia (NH3) has been proposed for use at Venus,where water would boil below 42 km and partiallycondense above 42 km.3'4'5 Alternatively, a more recentstudy at JPL has suggested that a helium-water mixturewould yield a lighter, more robust buoyancy system.6Whatever the final choice of fluids, the feasibility ofthe PCF balloon concept itself has been recentlydemonstrated in a successful campaign of flights overthe Southern California deserts.7

A nominal Venus aerobot mission profile is shownin Figure 2. The alternating evaporation andcondensation of the PCF causes oscillatory motion ofthe system about the equilibrium altitude. Note,however, that the aerobot does not require an on-boardenergy source to accomplish this vertical motion.Thermodynamically, the aerobot acts as a heat engineusing the planet's natural atmospheric temperaturegradient below the tropopause to drive the system'svertical motion in the gravitational field. The thermalenergy absorbed by the PCF during evaporation at lowaltitudes is used to propel the aerobot upward. Whenthe energy is released by the PCF during condensationat high altitudes, the aerobot descends and thethermodynamic cycle is complete.

The natural oscillating behavior of the aerobot canbe altered by capturing the PCF liquid in a smallreservoir, thereby delaying or preventing itsvaporization. In this situation, the aerobot will descenddeep into the atmosphere and potentially to the planet'ssurface. This will enable surface and near surfacescientific examination of the planet or moon. Upon

release of the liquid from the reservoir, the aerobot willre-ascend to the upper atmosphere.The Venus Environment

The Venusian environment is challenging for insitu scientific exploration because of its dense, opticallythick and corrosive atmosphere. The nominal flightprofile (Figure 2) for a Venus Aerobot spans the rangefrom 0 to 60 km altitude. On the surface, thetemperature is 460 °C and the pressure is .92atmospheres; at 60 km, the temperature is -10°C andthe pressure is 0.24 atmospheres. The Venusian cloudlayer extends from approximately 47 to 70 km withthick haze layers both below and above. Not only dothese layers hamper visual observations, but the cloudsalso contain a significant amount of sulphuric aciddroplets. Therefore, all exposed components of theaerobot must be resistant to sulphuric acid attack, aswell as be compatible with the temperature andpressure extremes encountered during the flightprofile. The remainder of this paper will focus on thedesign of a gondola that can safely house the onboardelectronics and scientific payload in this environment.Venus Aerobot Gondola Requirements

A recent JPL effort has examined low-powercamera configurations and thermal control strategiesfor Venus aerobot operations. The Venus AerobotSurface Science Imaging System (VASSIS) projectbegan by deriving survivability requirements from theexpected aerobot mission profiles and the Venusianenvironment. Table 1 summarizes preliminary Venusaerobot mission flight requirements, and Table 2summarizes the Venus atmospheric parametersassociated with the flight requirements.

It can be shown that each additional kilogram ofmass carried on the gondola requires roughly athreefold increase in the overall system mass becausethe increased balloon volume required to carry theadditional gondola mass. Thus, the primary gondoladesign challenge is to provide pressure and thermalprotection, provide an optical viewport, and allowexternal data communications with a minimumamount of mass. The following section describes agondola design that meets the VASSIS requirements.

VASSIS Gondola DesignPressure Vessels

A schematic of the VASSIS gondola is shown inFigure 3. It's a concentric sphere design employingboth passive and active thermal control devices. Thecentral spherical instrument housing is 30.5 cm(12 in.) in diameter and is surrounded by multi-layer

American Institute of Aeronautics and Astronautics


insulation (MLI) in a vacuum. The proposed MLI is 15to 20 layers of doubly-aluminized Kapton withfiberglass spacers. In addition, a couple of layers ofaluminum or gold foil will be required next to the outershell because Kapton starts to degrade in the range of430 to 450 °C. Note that this inner vessel is notsubjected to Venusian atmosphere pressure loading andtherefore is only 0.75 mm thick. In contrast, the outerpressure vessel is a spherical titanium shell that is 4.0mm (0.16 in.) thick. It is designed to withstand theVenus surface conditions of 460 °C and 92 bar, as wellas atmospheric entry force loads of up to 500 g.9>1° Thecentral instrument housing is held in place by a seriesof tension bands (Figure 4) fabricated from braidedpolybenzoxazole (PBO) fiber material.^ Despite themechanical complexity of using tension band supports,PBO's high strength and low thermal conductivitymake it an attractive choice for minimizing heatconduction from the hot outer shell to the cool innershell in the lower atmosphere.J

Gettering material will be placed in the annulusbetween the two shells to help maintain the vacuum.One material being tested at JPL is the zirconium-based getter St707 manufactured by SAES getters.Phase-change Material (PCM) and Heat Pipe

A heat sink is used in addition to the MLI toprovide temperature control for the instrumenthousing. It consists of a high heat of fusion phasechange material (PCM), LiNO3'3H:O, that absorbs 296kJ/kg of heat conducted through the MLI. Note thatthis phase change material is different than the phasechange fluid used as a buoyancy gas inside the balloon.The melting temperature of LiNO3'3H2O is 30 °C;therefore, during descents into the hot loweratmosphere, the PCM will maintain a 30°C internaltemperature until it is completely melted. Upon re-ascent to a 55 km altitude, the PCM will begin tofreeze again as it dumps heat to a now cooler than 30°Catmosphere. A simple, hollow stainless steel tube,1.27 cm (0.5 in.) diameter, partially filled withammonia, is the means by which heat is transferredfrom the PCM in the central instrument housing to theouter pressure vessel and atmosphere. When the

unpublished JPL data.

* This material is also being investigated for use as theprimary Venus balloon material."

* However, current experimental work at JPL is using analternative design based on rigid Ti-6Al-4V struts.

pressure vessel is cooler than the instrument housing,ammonia will condense in the pipe near the pressurevessel wall ,and will fall by gravity to the innerinstrument housing. The liquid ammonia will absorbheat from the PCM as it boils. The vapor returns to theouter pressure vessel and condenses again on the coolersurface, thereby transferring energy to the environmentand completing the cycle.

The heat delivered to the outer pressure vessel bythis gravity-fed diode heat pipe (also known as a refluxboiler) transfers conductively through the titaniumpressure vessel to a second, external heat pipe circuit,without the necessity of transferring the ammonia itselfthrough the pressure vessel wall. This outer, secondaryheat pipe circuit (Figure 5) is fashioned similarly as agravity-fed, 1.27 cm (0.5 in.) OD hollow tube, whereinammonia is condensed in an upper finned heat transfermatrix and falls by gravity to the outer pressure vesselwall area where it boils.

Note that when the outer shell is hot in the loweratmosphere, the ammonia does not condense on it andtherefore does not transfer a significant amount of heatbetween the vessels. The one-way nature of the heattransfer is why it's called a diode heat pipe.

Gondola Thermal PerformanceTo evaluate the thermal performance of the

system, it was necessary first to estimate the balloonascent and descent velocities. A transient thermalmodel was constructed for a 25 kg gondola with aballoon that was fabricated from 0.05 mm (0.002 in.)PBO film. At an altitude of 56 km, the balloon is 2.5 mdiameter and 25 m long, and it is filled with PCFmixture of 50 % ammonia and 50 % water. The totalfloating mass of this system is about 100 kg (25 kgammonia, 25 kg water, 25 kg payload, 15 kg balloonmaterial, and 10 kg structure and miscellaneous).Figure 6 shows the altitude profile for the aerobot, andFigure 7 shows the velocity profile for the aerobot.(The secondary effects of atmospheric thermalradiation were not included in the simplified aerobottrajectory analysis.)

As shown in Figure 7, the initial descent velocityis 12 m/s, but this slows to a nearly steady value of3 m/s from 30 km down to near the surface. Totaldescent time is calculated to be 3.5 hours, while totalascent time is slightly less than 2.5 hours. Thesecalculations assume only balloon drag is important,with a descending coefficient of drag of 0.8 and anascending coefficient of drag of 0.4, as has beenestimated for similar Earth balloon systems.12 Also,balloon heat transfer rates are assumed to be infinite so



that all balloon water vapor is condensed quickly ataltitude and boiled quickly near the Venus surfacewhen ascent is initiated.

A thermal model of the VASSIS gondola has beenconstructed in parallel with the above trajectory model.The resulting heat loads are shown in Figure 8 as afunction of altitude. The problem of calculating heattransfer rates through the MLI was finessed by simplyassuming that the total effective radiative emissivity (e)between the concentric spheres is 0.01. This value issomewhat lower than achieved on typical spacecraftinstruments (for which E ~ 0.03) but is believed to beachievable with careful design and installation of theMLI. Note that extremely low power dissipation rates(< 1 W average) are expected for the scienceinstruments during low-altitude operations. The aboveassumption is reasonable because a communicationsystem would be used only during upper-atmosphereoperation. Thus, no internal electronic powerdissipation is shown in Figure 8.

At altitudes below about 10 km, the radiation heatleak between the spheres dominates conduction heatleaks from the PBO band supports and the combinedconduction and radiation at the camera window(constructed from fused silica glass). Total heat load atthe surface is roughly 85 W. Approximately 1.5 kg ofPCM mass is necessary to absorb the integrateddescending heat load (Figure 9), a further 1 kg of PCMis required to absorb the heat load from 1 hour ofsurface operations, and another 1 kg is required toabsorb the integrated ascending heat load. The nominaldesign includes an additional 0.5 kg of PCM formargin.

An estimate has been made of the time requiredfor PCM cooling when the balloon bobs between 56 kmand 40 km during alternate condensation and boilingof the water in the water/ammonia balloon. The modelshows that two hours of bobbing time (about two fullround trips) is required for each kilogram of PCM to bere-frozen. Thus, one repeatable mission scenario couldconsist of a 3.5-hour descent, 1 hour on the Venussurface, a 2.5-hour ascent, and 8 hours bobbing timefor a total of 15 hours if 4 kg of PCM is used.Significantly longer low-atmosphere hover times couldbe achieved if the aerobot did not descend all the wayto the surface due to the compounding effect of rapidlyincreasing heating rates and transit times at the lowest,altitudes.

Table 3 presents a mass breakdown of the thermalcontrol system. The PCM and the two ammonia heatexchangers comprise the bulk of the 7.1 kg total. Table4 presents a mass breakdown for the entire gondola.

The thermal control system comprises 28%. the outertitanium shell 34%, and the onboard electronics andscience payload 29% of the 25 kg total.

Venus Aerobot Thermal/System Design IssuesThere exist many interesting design issues and

tradeoffs with the gondola thermal design and itsinteraction with the overall Venus aerobot balloonsystem design. First, it must be realized that becausethe aerobot is a floating system, small increases ininsulation or PCM mass lead to large increases in totalsystem mass. This drives the need for highly efficientinsulation (MLI) to minimize the heat leak into theinstrument container and hence minimize PCM mass.Similarly, if the heat of fusion of the PCM were to belarger, then less would have to be carried and theoverall system mass would shrink. Alternatively,higher heat of fusion PCMs would enable longer hovertimes near the surface for the same mass.

Another interesting tradeoff concerns the choice ofmelting temperature of the PCM. The lower the PCMmelting temperature, the higher the aerobot must fly torefreeze the PCM after surface operations. Higheraltitudes require larger (and heavier) balloon systemsand therefore heavier cruise and entry systems. On theother hand, higher melting temperature PCMs lead tohigher operating temperatures for the onboardelectronics and science instruments. This can be aparticular problem for imaging systems where thesignal to noise ratio is a strong function of temperature.

Another design tradeoff involves the shape of theballoon and the PCM mass. To limit transit timebetween the upper, cooler atmosphere and the lower,hotter surface (and therefore reduce the amount ofPCM necessary), the balloon should present a smallfrontal area to minimize drag. To achieve reducedfrontal area for a given volume, a cylindrical balloonwith large aspect ratio (length/diameter) can be used.(The balloon used herein for the gondola design has arather large aspect ratio of 10.) But, because of highsurface area to volume ratios, high aspect ratiocylindrical balloons are relatively massive compared tocylinders (or spheres) with near-unity aspect ratios.The minimum mass design for the overall system musttherefore balance these conflicting trends.

Applications for Outer Planet ExplorationThe vacuum dewar, PCM gondola concept can

also be used for deep atmospheric exploration of all thegas giant planets. Although the thermodynamicprofiles vary from planet to planet, all feature highpressure, high temperature environments similar to



that of Venus at sufficiently low altitudes. A verypreliminary evaluation has been performed byconsidering a point design for a deep-Jupiter probe thatwould make only one descent.

The non-repeating nature of this Jupiter designmeans that an atmospheric cooling heat exchanger isnot required. Therefore the gondola/probe can bestreamlined to increase the drop velocity and allowdeeper penetrations for a given thermal design. Withan aerodynamic tear-drop shaped inconel shell insteadof a spherical titanium shell, a 50-kg, 38-cm diametergondola would take approximately 1.5 hours todescend 414 km from the anticipated Jovian waterclouds (5.0 bar, 0 °C) to about 500 bar pressure and806 °C. Less than 2 kg of water serving as the PCMwould be required to maintain all science instrumentsat 0 °C for the entire descent. For comparison, the 118kg Galileo descent module only returned data fromroughly the 20 bar (70 km) level.

ConclusionThis paper presents a design for a low-mass,

thermally protective gondola that can be used forVenus aerobot operations near the surface. Theproposed aerobot mission scenario involves descent tothe surface followed by re-ascent to the upper, cooleraltitudes of Venus on a 15-hour repeatable cycle.Vacuum dewar technology with multilayer insulation isused to protect the science instruments andcommunication hardware during excursions into thehot lower atmosphere. A phase change material is alsoused as a heat sink to help limit Hhe temperatureincrease inside the gondola. Four kilograms of phase-change material will enable a mission sequencecomprised of a 3.5-hour descent to the surface, 1 hourof surface operations, a 2.5-hour ascent to the upperatmosphere (about 56 km), and 8 hours to freeze thephase-change material. Total aerobot mass is 100 kg.The gondola mass is 25 kg with 7.3 kg available forscience instruments and communication hardware.

The basic configuration of outer pressure vessel,inner evacuated MLI region, and inner science vesselwith PCM has potential uses for deep atmosphericstudies of not only Venus but also for Jupiter and theother, somewhat cooler, gas giant planets of Saturn,Uranus, and Neptune.

AcknowledgmentsThis work described in this paper was carried out

by the Jet Propulsion Laboratory, California Institute ofTechnology, under a contract with the NationalAeronautics and Space Administration.

References1. Gilmore, M.S., el al. "Investigation of the

Application of Aerobot Technology at Venus:Scientific Goals and Objectives of the ProposedBalloon Experiment at Venus (BEV) and VenusFlyer Robot (VFR)—Final Report", BrownUniversity' Department of Geological Sciences,1 June 1996.

2. Nock, K.T., J.A. Jones, & G. Rodriguez."Planetary Aerobots: A Program for RoboticBalloon Exploration", 34th AIAA AerospaceSciences Meeting and Exhibit. 15-18 January1996, Reno, Nevada, paper number AIAA 96-0355.

3. Moskalanko, G.M. "Mekhanika Poleta vAtmosfere Venery", Mashinosteroenie Publishers.Moscow, 1978.

4. Moskalanko, G.M. "Two Component WorkingMaterial for a Floating Probe in the Atmosphere ofVenus", Proceedings, High Temperature Elec-tronics and Instrumentation Conference,December 1981.

5. Moskalanko, G.M. "Dirizhabl' dlya Venery".NaukaZhizn, No. 9, September 1981, pp. 85-87.

6. Nock, K.T. et al. "Venus Geoscience AerobotStudy (VEGAS) Report", unpublished JPL report,28 July, 1997.

7. Nock, K.T., et al. "Balloon Altitude ControlExperiment (ALICE) Project", 11th AIAA Lighter-Than-Air Technology Conference, 16-18 May1995. Clearwater, Florida.

8. Klaasen, K. "VASSIS Camera SNR Analysis as aFunction of Wavelength, Altitude, and SunAngle", unpublished JPL report, 2 July 1996.

9. Salama, M. "The VASSIS Gondola StructuralDesign", JPL Internal IOM 352G:96:146:MS, 6November 1996.

10. Salama, M. "Revised VASSIS Gondola Design",JPL Internal IOM 352G:96:150:MS, 30 December1996.

11. Yavrovian, A. et al., "High Temperature Materialfor Venus Balloon Envelopes", 11th AIAA Lighter-Than-Air Technology Conference, 16-18 May1995, Clearwater, Florida.



12. Wu, J.J. & J.A. Jones. "Performance Model forReversible Fluid Balloons", 11th AIAA Lighter-Than-Air Technology Conference, 16-18 May1995, Clearwater, Florida.

Table I.Venus aerobot mission flight requirements.

ParameterMission duration

Maximum atmospheric entry decelerationEstimated descent, ascent, and float periods

Flight altitude

Requirement30-90 davs250-500 g

Descent: 3.5 hr, Ascent: 2Float: 0-1 hr

5hr.

surface < z < 60 km

Table 2. Venusian environment parameters corresponding to flight requirements.

ParameterAtmospheric temperature range

Atmospheric pressure rangeAtmospheric corrosives

Solar fluxLength of day as seen by aerobot

Length of night as seen by aerobot

Value-10 °C < T < 460 °C

0.25 atm. < P < 92 atm.Sulfuric acid

40 W/m2 (diurnal average)~ 96 hours~ 96 hours .

Table 3. VASSIS temperature control mass summary.

ComponentPhase-change Material (PCM)

PCM Heat Exchanger (Inside Gondola)MLI

Internal Gravity-fed Heat PipesExternal Heat Exchanger (Beryllium Fins)

GettersTotal

Mass (kg)4.01.50.20.30.90.27.1

Table 4. VASSIS gondola mass summary.

ComponentTemperature ControlOuter Titanium ShellInner Titanium Shell

Tension-band SupportsMiscellaneous

Science Instruments and CommunicationTotal

Mass (kg)7.18.61.00.50.57.3

25.0



oo

500 400 300TEMP (°C)

200 100 50 100

METHYLENECHLORIDE (

'(CH2CI2)

1.0 1.5 2.0 2.5 3.01000/T(K-1)

3.5 4.0

tilccVIV)IDcca

Figure 1. Van't Hoff plot for Venus atmosphere and candidate PCFs.

60

48km93* C Water1.3atm Starts to

Condense42km

INFLATEBALLOON

BYNATURALHEATING

REASCEND •TO A

CLOUDS T

, DESCEND\ TO

Reservoir &Heat Exchanger Op«n Valxs to

VaporizeTraco«d Water

Figure 2. Venus aerobot mission profile with surface descent capabilities.



ENCLOSEDINSTRUMENTHOUSING

MULTI-LAYERINSULATION

PRESSURE VESSEL

TENSION BANDSUPPORT (500-G LOAD)

VACUUM GETTERMATERIAL

HEAT PIPE/HEATEXCHANGER TOVENUS AMBIENT

DIODE HEAT PIPE

PHASE CHANGE MATERIALWITH INTERNAL HEATEXCHANGER

CAMERA WITH ROOM-TEMPERATUREMETAL/GLASS SEAL

Figure 3. VASSIS gondola temperature control schematic.

Figure 4. PBO Tension-band support system.



Figure 5. Heat exchanger configuration.

3600 7200 10800 14400 18000 21600 25200 28800

Flight Time [sec]

Figure 6. Aerobot altitude profile.



4

60

50

40

30

20

10-

-20 -10 0 10

Vertical Velocity [mis]20

Figure 7. Aerobot descent velocity profile.

Heat pipe conduction is minimal

Conduction and radiation through window

Radiation between spheres

10 20 30 40 50 60 70 80 90 100

Heat Load [W]

Figure 8. VASSIS gondola heat loads.



1.0 2.0 3.0 4.0

PCM Solid Mass [kg]5.0

Figure 9. PCM solid mass during flight.


Documents

[American Institute of Aeronautics and Astronautics 36th AIAA Aerospace Sciences Meeting and Exhibit - Reno,NV,U.S.A. (12 January 1998 - 15 January 1998)] 36th AIAA Aerospace Sciences