5ga3

NASA Technical Memorandum 103642

Design Considerations for LunarBase Photovoltaic Power Systems

J. Mark Hickman, Henry B. Curtis, and Geoffrey A. LandisLewis Research CenterCleveland, Ohio

Prepared for the21st Photovoltaic Specialists Conferencesponsored by the Institute of Electrical and Electronics EngineersKissimmee, Florida, May 21-25, 1990

NASA

https://ntrs.nasa.gov/search.jsp?R=19910004946 2018-08-05T16:57:45+00:00Z

Design Considerations For Lunar Base Photovoltaic Power Systems

J. Mark Hickman, Henry B. Curtis, and Geoffrey A. LandisNational Aeronautics and Space Administration

Lewis Research CenterCleveland, Ohio 44135

Abstract

A survey was made of factors that may affect the designof photovoltaic arrays for a lunar base. These factors, whichinclude the lunar environment and system design criteria,are examined. A photovoltaic power system design with atriangular array geometry is discussed and compared to anuclear reactor power system and a power system utilizingboth nuclear and solar power sources.

Introduction

As part of the Space Exploration Initiative, NASA isinvestigating photovoltaic power systems for the lunarsurface. Power systems considered are for short durationstays without storage (14 days) and prolonged periods withenergy storage so that power can be supplied during thelunar night. The purpose of this paper is to discuss thevarious issues and constraints which affect the design ofphotovoltaic power systems on the moon.

Lunar Base Power Requirements

The power requirements for a lunar base are determinedby the crew size, evolutionary stage, and mission objectivesof the base (ref. 1). It is widely accepted that a lunar base willgrow in capacity and function, and thereby in powerrequirements, over time. To support this growth, additionalcrew members will be required. A minimum power level ofapproximately 3 kW of electrical power (kWe) is requiredto support each crew member (ref. 2). As the mission

objectives evolve over time, additional power generationunits may be necessary.

The baseline power source options are photovoltaic(PV) arrays or a nuclear system. Photovoltaic arrays havethe advantage of being modular, lightweight, and reliable,but the disadvantage of requiring an energy storage systemif nighttime power is required. PV arrays have a long recordof reliable power production in space and on the moon,which reduces the technical risk. Nuclear power systemshave the advantage of providing continuous power and oflower mass at high power. However, nuclear power systemspresent a potential radiation hazard to base personnel andequipment. Adequately safeguarding the base is a majordesign concern. In general, the use of nuclear power in spaceis a highly sensitive political issue.

To make use of the strengths of each power systemtechnology, a lunar base may use photovoltaic power for theinitial set-up, and then augment this with a nuclear reactor aspower requirements increase. However the base powersystem is configured, crew surface time for deployment andset-up will be severely limited. It is important that powersystem components (e.g., arrays) be designed such that littleor no assembly or intervention by base personnel is required.

If 100 kWe or more is required within the fast fewflights of the developmentof the base, mission planners mayforgo photovoltaic arrays entirely, except as a deployableemergency power generation system. For high powerlevels, the mass of the energy storage system required tosupply power over the 354-hr lunar night is high. A systembeing considered by NASA for early high power generationis a modified SP-100 nuclear reactor with thermoelectric

energy conversion. Such a nuclear power "module" couldbe emplaced within the first few flights providing 100 kWeearly in the base development. Additional thermoelectricmodules could be emplaced to build up base power. Alter-natively, dynamic conversion engines could be used in placeof the thermoelectres to yield 500 to 1000 kWe. Powerlevels in this range will be necessary for in-situ resourceutilization (ISRU), i.e., lunar mining and processing.

Photovoltaic arrays with regenerative fuel cell energystorage (PV/RFC) is a power system candidate in a lunarbase development plan that does not require high powerlevels early. Option A of the Reference Architecture of theNASA Lunar/Mars 90-Day Study Period manifests a PV/RFC system module followed by two additional modules onthe second and third flights to the moon, respectively (ref. 3).Each module would provide 25 kWe during the lunar dayand 12.5 kWe at night to support a four-person crew. Thissame study option then manifests a 100 kWe nuclear powermodule on flight 7, about 3-1/2 years into the base develop-ment Option E of this same study includes PV arrays for thelunar base only as an emergency backup to nuclear reactorpower.

For low power outposts (i.e., less than 50 kWe), awayfrom the main base, such as an astronomy science outpost onthe moon's far side, PV/RFC units are mass competitivewith all other power systems. Should outpost power berequired only during daylight hours a photovoltaic powersystem (without RFC energy storage) would be the systemof choice on a mass basis, especially in the region of 10 to100 kWe.

Lunar Environment

Array Temperature

A solar array on the moon will operate at significantlyhigher temperatures than arrays in near-earth space. Oper-ating temperatures are determined by the energy balance,where the incident energy minus the energy converted intouseful power is radiated thermally according to the fourthpower of temperature. The lunar soil is a good thermalinsulator, and thus the solar array will be able to radiate tospace only from one side. The operating temperature on themoon can thus be estimated from operating temperatures inhigh orbit by assuming that the solid angle available forradiation iscut in two. The maximum operating temperatureon the moon is therefore increased by about 19 percent.Since typical operating temperatures for geosynchronousorbit arrays are —305K, this yields a maximum operatingtemperature of 90 °C (decreasing slightly if the cell effi-ciency increases). This is very close to the temperatures

reached by the lunar surface at local noon (ref. 4). Averagedaytime temperature will be somewhat lower.

These numbers are roughly consistent with those mea-sured by instrument packages left on the moon duringApollo. For example, the Apollo 11 PSEP reached amaximum temperature of 88 °C at lunar noon (ref. 5).Similarly, the Apollo 12 Surface Magnetometer reached amaximum extemal temperature of about 78 °C (ref. 6).

The large areas required for the solar array make itunlikely that cooling techniques will be usable. Since solarcell performance decreases with increasing temperature, thesolar cell material selected should not be highly sensitive totemperature. The temperature dependence is primarily afunction of the bandgap of the material with lower tempera-ture sensitivity for wide-bandgap materials, such as GaAs oramorphous silicon. If the bandgap can be increased, as bygoing to a ternary III-V compound such as A1GaAs, thetemperature sensitivity is decreased yet further, although atsome cost in decreased efficiency at standard temperature.Cascade (or "tandem") cells also have high temperaturesensitivity, typically equal to the sum of the sensitivities ofthe individual component cells, and are thus less desirablefor lunar use, although of higher baseline performance atstandard temperature.

The temperature variation of power (1/P aPMT) forgallium arsenide cells is about 0.25 percent/°C (refs. 7and 8). For cell operation at 90 °C, the power would bederated by about 17 percent due to temperature. Amorphoussilicon would be comparable or slightly better. For silicon,the temperature variation is about 0.33 percent/°C, leadingto about 23 percent loss, with CuInSez expected to be aboutthe same.

For the single crystal solar cell technologies, GaAs andSi, the temperature extremes are not expected to presentlifetime problems if adequate design safeguards againstthermal cycling are taken. For thin-film technologies, long-term operation at high temperatures and vacuum thermalcycling stability have not yet been demonstrated, and reli-ability will have to be verified before such arrays can be usedon the moon.

Radiation Environment

The moon has no permanent general magnetic fields;hence, there are no trapped radiation belts. The major sourceof natural particle radiation for an array on the lunar surfaceis solar flares which consist mainly of protons. Protonsdamage cells by displacing atoms within the lattice causingdefects. These defects change the electronic properties ofthe material shortening cell life. Unlike the continuous Van

2

1015

10 14

.zP

101710 20 30 40 so 60

Coverglass Thickness (mils)

Figure 1.—Annual equivalent 1 MeV electron fluence due to solarflares.

Allen belt radiation, solar flares occur sporadically withvarying magnitudes. The effect of solar flare protons isusually handled statistically with an equivalent 1-MeVelectron annual fluence of 1.Ix10 14 e/cm2 for silicon cellswith a 3 mil (75 µm) coverglass (ref. 9). Data for othercoverglass thickness are shown in figure 1. During the lunarnight, when the moon is between the sun and the arrays, thearrays will be protected from solar flare protons. Thus theflux shown in figure 1 will effectively be reduced by a factorof two.

Lunar Dust

Dust on the array surface will reduce light incident tothe array and increase the array operating temperature.Likewise, dust on radiator surfaces--fuel cell radiators, forexample--will reduce the radiator effecti veness. Dustcanbetransported to the array and radiator surfaces by astronautsor rovers kicking up dust during EVA, by dust blown ontothe array by rockets ascending and descending, and possi-bly by other mechanisms involving electrostatic transport.To a large extent, this problem can be ameliorated bylocating the solar arrays away from high-traffic areas of thebase, and notallowing astronaut activity in the array vicinity.Since small dust particles will likely be electrically charged,any dust on the array will adhere to the surface by electro-static attraction. If it is not possible to eliminate dust fromthe surface, the adhesion could be reduced by a transparentconductive surface layer to ground the electrostatic charge.

Photovoltaic System Design

Cell Technology

Current technology spacecraft solar cells are madefrom silicon (Si) and gallium arsenide (GaAs) (refs. 10

and 11). The bestpresent flight technology uses thin (62 µm)silicon cells. Efficiencies of 19 percent AMO (Air MassZero) have been demonstrated; however, production cellsare more typically around 15 percent efficient. GaAs cellswith an 18 percent AMO efficiency are in production, andproduction readiness has been demonstrated for 20 percentefficientGaAs cells. Recent GaAs cells have been manufac-tured on germanium substrates to improve its handlingcharacteristics (ref. 12). The germanium can then be etcheddown to a 50 pm thickness to reduce the weight. Analtematemethod of producing such ultra lightweight GaAs cells is touse a technique which separates the cells from a reusablesubstrate, such as the CLEFT process (ref. 13). An array ofsuch thin GaAs cells using existing array structures couldhave a specific power of about 300 W/kg.

Cascade solar cells make more efficient use of the solarspectrum by stacking subeells of different materials designedto absorb a different wavelength range. This technology hasproduced the highest efficiency solar cells to date, withdemonstrated efficiencies under space (AMO) sunlight ofover 30 percent. However, the technology is still in theresearch stage and is unlikely to be production ready fornear-term use.

Thin-film technologies include CdTe, CuInSe2, ter-nary compounds, and amorphous silicon, plus cascade cellsmade from these materials. Current technology for thesematerials is comparatively low in efficiency (5 to 9 percentAMO), but the cells can be made extremely thin (1 to 2 µm)and thus potentially have specific powers of well over1000 W/kg (ref. 14). Cascade thin-film solar cells, such asCdZnTe on CuInSe2, have potential for both high efficiencyand low weight. To date only amorphous silicon has beenproduced on thin, lightweight polymer substrates, whichhave efficiencies less than those achieved on rigid sub-strates. Polymer substrates have not been extensivelystudied as most thin-film research has been directed towardterrestrial applications.

Storage

>

The energy storage requirements for nighttime powersupply dominate the power system mass. Currently usedpower storage systems, such as NiH 2 batteries, are inad-equate for the large power requirements fora lunar base. Thebaseline reference for energy storage at the lunar base callsfor hydrogen/oxygen regenerative fuel cells (RFC). TheseH/O RFC's are expected to provide 500 Whr/kg by the year2000, using gaseous reactants. By cryogenically cooling theH/O reactants, specific energies of 1000 to 1500 W hr/kg areanticipated (ref. 15). Figure 2 shows an artist's conceptionof a 50-kWe energy storage system employing cryogenicreactant H/O RFC's. Other energy storage systems such assuperconducting energy storage coils, massive flywheels,

r AIR

C-88-11517

Figure 2. —50kWe solar photovoltaic-regenerative fuel cell powersystem with cryogenic storage for a lunar observatory.

6.1%

Figure 3.—Mass breakdown dof lunar PV array tent with RFCenerrgy storage.

and thermal salts are either insufficiently advanced to beavailable for the lunar base or impracticable for applicationon the moon (ref. 16). Even advanced cryogenic RFC's,however, can constitute 70 to 90 percent of the mass of a PV/RFC power system (fig. 3). Therefore, even tremendousadvances in cell technology will not significantly affect thetotal mass of a solar power system. Consideration must begiven to other figures-of-merit, such as cost, technologyreadiness, lifetime, reliability, maintainability, and safety.

To minimize storage requirements, the power usedduring the night should be minimized. Some applicationssuch as resource utilization (for example, recovery of oxy-gen or hydrogen from lunar soil for use as rocket propellant)could be scheduled to require power primarily during thedaytime. Other usage, however, such as lighting and lifesupport, will require continuous power. One option forreducing the nighttime life support requirements is to storethe waste gasses at night for processing during the daytime,rather than to reprocess during the night. This has the

potential for reducing the minimum required night power tobelow the 3 kW per person baseline. To account for the factthat night power requirements may be different from dayrequirements, we define the power fraction f as the ratio ofthe required night power to the required day power. (Thisvalue is also sometimes referred to as the energy storage dutycycle.)

Array Orientation

One major design feature of the lunar PV array is itsorientation to the sun. Both planar and concentrating arraysarepossible. A concentratorarray requires constant trackingto within about a degree of arc which in turn requiresadditional structure and mechanisms. For the purpose of thisstudy, the complexity of a tracking concentrator arrayeliminates it from consideration. A planar array in ahorizontal configuration will have times during the lunarday where little or no energy is being generated due to poorsun angles.

C-90-05218 j

Figure 4.-Self-deploying tent array.

An array geometry which lessens this problem is atriangular or "tent" configuration (fig. 4). This arrangementof two panels sloping upwards toward each other is moreefficientnear lunar dawn and dusk than a horizontal configu-ration. By setting a requirement that the arrays must provide100 percent of the daytime load power from sunrise tosunset, the mass of the storage system that would otherwisebe required to supply energy during the lunar morning andevening is obviated. The angle of array tilt required toprovide this power profile is discussed in the next section.Figure 4 shows a schematic of a mechanism that coulddeploy an array with very little human intervention. Fig-ure 5 shows an artist's conception of how a power system fora moon base might appear shortly after landing.

RunCZ 'Alk sin a = 2(cos(x + 1)/(TCk). (5)

where we have defined k = (1 + f/rl). To minimize thestorage, we require that the array power at sunrise equal thedaytime load P&y , i.e., immediately at sunrise no power isdrawn from the storage system. This then gives us anequation for the array tilt angle as

Figure 5.—Artist's conception of photovoltaic power systemsemployed on a lunar, showing the east/west "tent" arrayorientation.

Triangular array tilt angle. - Consider an array con-sisting of two identical panels, each tilted an angle a fromthe horizontal, respectively toward sunrise and sunset. If therated array power at normal incidence of the panels com-bined is A, and 0 is the sun angle with 0 = 0 defined as solarnoon, the power for the tilted array is:

P = A cos a cos 0, for 101 < n/2 - a, (la)

P = A(cos a cos 0-sin (x sin 0)/2,for -7c/2 <_ 0 5 -n/2 + a, (lb)

P = A(cos a cos d+sin (x sin 0)/2,for n12-a <_ 0 <_ n/2, and (lc)

P = 0, for 101 > n/2. (ld)

Thus, the average power over the daytime is:

Pave = A[cos a + 1 ]/Tc (2)

which, as should be expected, has a maximum value of 2/nfor a = 0, a horizontal array. (For comparison, a trackingarray has Pave/A=1.) The power at sunrise equals thepowerat sunset,

Psunrise = (sin (x)/2. (3)

Consider energy storage with an efficiency Tl (energyou t/e,nergy in) and power fraction f. Then the average powergenerated during the day, Pgen, must be larger than the day-time load by a factor k:

Pgen = 0 + fM)Pday = kPday, (4)

The solution to this equation is:

a = cos- t [(k2 - 4/IC2) / (k2 + 4/7E2)] • (6)

As an example, suppose night and day power require-ments are equal, and the energy storage efficiency is100 percent. Then the sunrise power must be exactly halfthe average daytime power, and the angle a is:

a = cos-1 [('c2-

1)/(1[2+ 1 )] = 35.3 0 . (7)

From equation 2, the array considered provides 58 per-cent of the power per unit area of a tracking array. Fora morerealistic example, suppose the required night power is halfthe daytime power and the round-trip storage efficiency is60 percent. Then Vq = 0.833, and the array angle a= 38.4°.This is 57 percent of the power per unit area of a trackingarray. As can be seen, the required angle increases as f/9decreases.

This method yields the array tilt angle such that theaverage power integrated over the lunar day is sufficient fordaytime load and nighttime storage requirements. Caremust be taken, however, in cases where the nighttime powerrequirement is a low percentage of the daytime power(low f). In these cases, the tilt of the arrays from thehorizontal is so large that the power variation during the daymay drop below the load requirement, requiring use ofenergy storage during the daytime. This would requireadditional array area and fuel cell radiators designed to workat the higher daytime temperatures. A triangular array witha round trip storage efficiency of 60 percent (TI = 0.60) anda power fraction of only 5 percent yields a tilt angle of 60.9°(fig. 6). Tent angles above 60° allows the generated powerto dip below the load power level. Thetiltangle a willequal60° when k = 243/n.

Different constraints apply if no storage is required, asfor a base occupied during the daytime only. In this case, itis desirable to make the power profile as close to uniform aspossible. This is accomplished with a tilt angle of 60°. Anarray with a=60°, called an equilateral tentarray, will havefour power generation minimums (at 0 = 0°, 60°, 120°, and180°, i.e., lunar dawn, 118 hr, 236 hr, and sunset, respec-tively). The minimum is equal to the load requirement.

9

s8

lgoA- 0%Ni&Power

--_-

60% Round-Trap Storage EHniency5% Nigh Power

160--"-s---- 10% N-91. Power

....... .. ............................................----^---- 20% Nigk Power

120 s' s'-^ r_'^^.r^w'•y1i ^a-r'•-x•.77.

100

An

0 60 120 ISO

Sun Angle from Dawn to D-k, deg—

Figure 6.—Effect of low nighttime power fraction on powerprofile

9W

2D%Nigk Power 60% Round -Trip Slonbe Ffficiency50% Nigk Power

350- 80%Nigk Power m-'a--,p,,w.

____W

250 ..................... Z ......... ... ...........N .................

200

150 ......... r5 ............. ..................... . ............:a, .X...........

100 ..................................................................................................

60 120 Igo

sun Angle from Dawn to Dusk, degmrn

Figure 7.—Effect of high nighttime power fraction on powerprofile.

60 120 180

Sun Angle from Dawn to Dusk, degrees

Power Management and Distribution

Thepower management and distribution (PMAD) sys-tem for the lunar base will be required to supply power tocrew habitats, science stations, ISRU facilities, and launch-ing and landing facilities. Each of these activity zones mustbe several kilometers distant from each other and from thePV arrays-, the activity from one zone must not interfere withthe activity or operation of mother. The science laboratorieswithin the habitation unit or in special attached lab moduleswill require a standard operating voltage and amperage.Much like in the Space Station Freedom and with terrestrialutilities, the power conditioning must be able to servicemany users with different power requirements.

The long transmission distances (on the order of 1 kmfrom the central habitation zone to any of the other zones)and the accommodation of users will drive up the mass o f thePMAD system. Transmission distances from nuclear reac-tors would most likely be on the order of a kilometer or moreto reduce radiation effects. This would require the formationof a "zone of exclusion" around the reactor wherein humanactivity would be severely restricted.

Specific masses of PMAD systems range from theSpace Station Freedom PMAD system at several hundred to1 kg/kWe or less for advanced systems with dedicated loads.It was assumed in this study that the lunar base PMADspecific mass would be about 20 kg/kWe. This is based onthe assumption of a more advanced PMAD system than forthe Space Station with consideration of user requirementsand transmission distances.

Comparison of Photovoltaic to Nuclear Reactor andMultiple Source Power Systems

When comparing masses of potential lunar base powersystems, photovoltaic power systems are generally found to

^ 1,1I13mi1S,PVA11RFC(500Whr/k9)

--cr-- W/3 mil Si PVA & RFC(1500 Whr/kg)

---- SP-1001E R.-tm/PVA/DIPS

+ SP-1001F. Rector Power System

d

W?

a0sb

Figure 8.—Power profile for 60 degree tent angle. 5

RFCs to provide 50% night Power.

For high power fractions (>50 percent), the powergenerated at lunar noon is several times the load levelrequirement (fig. 7). The larger the peak power, the moremassive the power management system becomes. It wouldbe advantageous to keep peak power close to the load levelwithout dropping below it. This is especially true for a PVpower system designed to provide power only during thelunar day. This is also best accomplished at a tent angle of60° (fig. 8).

9

^ 2

ar

s

20 40 60 80 100

Power (<ve1, kWe

Figure 9—Comparison of systems for continuous powergeneration.

6

Multiple So— SP-100'IE RcwWr/PVA/D1PS

SP 1007E R­ Po Sy—

W/3 mil Si PVA -- Rqu Urral Tent

u

y

n10a8

r ^ .K-

00 50 100 150 200

Power L-4 kW.

Figure 10.--Comparison of systems for daytime powergeneration.

be heavier than nuclear power systems at high power levels(ref. 17). This occurs because the energy storage subsystemrequired by PV power systems to provide power over the354-hr lunar night is extremely massive, constituting up to80 to 90 percent of the PV power system mass.

In figure 9, a nuclear reactor power system is comparedwith two versions of the "array tent" PV power system, oneusing cryogenic reactant RFC storage (1500 Whr/kg), theother using gaseous reactant RFC storage (500 Whr/kg).Each PV system uses multijunction solar cells on a 3 milsilicon substrate. A fourth power system, shown in thefigure employs multiple power generation sources.

During NASA's 90-Day Study process, concerns wereraised that a single source power system would be vulner-able to systemic power system failure. For example, if dustis a problem for habitat arrays, then it will be a problem forarrays on rovers and on remote scientific instruments. Ifthermal cycling reduces the lifetime of the refractory metalsin one nuclear power module, then other similarly designedmodules may have the same problem.

One solution may be to design the lunar power systemusing multiple sources. Autonomous sources could gener-ate power, independently feeding into a power grid as withterrestrial power plants. Alternatively, a single source couldserve as the primary source with other sources available foremergency backup power. The multiple source powersystem used forcomparison in figures 9 and 10 is of the lattertype. This system uses a SP-100 thermoelectric reactorpower module as the primary source. In the event of powerloss--whether permanently through reactor failure or cool-ant loss, or temporarily through a transmission line breaknear the habitat--a deployable PV array would be used fordaytime power and a dynamic isotope power system (DIPS)would be used at night to supply continuous survival powerfor base personnel for an extended period, say until a newreactor module can be emplaced or repairs affected. Theemergency PV array is a horizontal GaAs on 3 mil Ge array

sized to provide 25 Me. The DIPS is comprised of five2.5 kWe DIPS units. Both the DIPS and the PV array wouldhave independent lines and conditioning units.

Figure 9 shows that this multiple source system islighter than the PV/RFC systems above 40kWe. This is dueprimarily to the massive RFC systems which in this instanceare only providing 50 percent night power. FigurelOshowsthat if storage is not necessary, PV power systems are lessmassive than nuclear systems. Note, however, that nuclearsystems can provide power through the lunar night, whereasPV systems without storage can only provide power duringthe lunar day.

Standby PV arrays. - Photovoltaic arrays used as anemergency power source would need to be designed withcertain characteristics. Since they are used during a poweremergency, they would need to be deployed quickly andwithout requiring power (at least from the primary source).When the emergency is over, they will need to be retractedto mimize the potential damage of solar flares, dust, andother environmental hazards. Both of these operationsshould be achievable with a minimum of human assistance.The deployable PV array should have minimal weight, a lowstorage volume, and a long shelf life. These characteristicswould be satisfied by a lightweight, thin-film roll-out blan-ket, for example.

Summary

Several features and constraints of photovoltaic powersystems for the lunar surface have been discussed. The mainfindings are:

1. The solar array is a small percentage of the overallPV power system mass.

2. Energy storage for the lunar night is the main massdriver. Minimizing nighttime power usage will signifi-cantly lower mass.

3. A "tent" array configured in an east/west orientationhas advantages over a fixed tilt or horizontal array due topower generation at dawn and dusk.

Future studies of lunar surface PV systems shouldinclude a detailed analysis of the power management anddistribution system (PMAD); a detailed thermal analysis ofthe PV array; long term effects of lunar environmentalfactors such as dust and the cycling to very low temperaturesdue to the 354-hr dark period; development of low massenergy storage systems; and further development of lowmass, deployable PV arrays.

7

References

1. Landis, G.A. , et al.: Photovoltaic Power for a LunarBase. Acta Astronaut., vol. 22, 1990, pp. 197-203.

2. Stever, M. et al.: Human Exploration of Space: AReview of NASA's 90-Day Study and Alternatives.(National Academy of Sciences, NASA Contract:NASW-4003) NASA CR- 186394,1990.

3. Reference Architecture Description for Lunar/Mars90-Day Study Period Developed for the Planet SurfaceOffice. Lunar and Mars Exploration Office, LockheedEngineering and Sciences Company, NASA ContractNAS-17900,1990,

4. Robie , R.A. ; Hemingway, B.S.; and Wilson, W.H.:Specific Heat of Lunar Surface Materials for 90 to350 Degrees Kelvin. Science, vol. 167, no. 3918,Jan. 30,1990, pp. 749-750.

5. Latham, G.V. , et al.: Passive Seismic Experiment.Apollo 11-Preliminary Science Report, NASASP-214, 1969, p. 144.

6. Dyal, P. ; Parkin, C.W. ; and Sonett: Lunar SurfaceMagnetometer Experiment. Apollo 12-PreliminaryScience Report, NASA SP-235, 1970, pp. 55-73.

7. Fan, J.C.C.: Theoretical Temperature Dependence ofSolar Cell Parameters. Sol. Cells, vol. 17, Apr-May1986, pp. 309-315.

8. Weinber, I., et al.: Radiation and Temperature Effectsin Gallium Arsenide, Indium Phosphide, and SiliconSolar Cells. NASA TM-89870,1987.

9. Tada, H.Y., et al.: Solar Cell Radiation Handbook.NASA CR-169662, 1982.

10.Ralph, E.L.: Photovoltaic Space Power History andPerspective. Space Power, vol. 8, no. 1-2, 1989,pp. 3-10.

11.Flood, D.J.; and Brandhorst, H.: Space Solar Cells.Current Topics in Photovoltaics, Coutts, T.J. andMeakin, J.D., eds., Academic Press, 1987, pp. 144-202.

12. Yeh, Y.C.M., et al.: High Volume Production ofRugged, High Efficiency GaAs/Ge Solar Cells. IEEEPhotovoltaic Specialists Conference, 20th, Las Vegas,NV, Sept. 26-30,1988, IEEE, New York, 1988,pp. 451.456.

13. Gale, RP. et al.: High Efficiency Thin-Film A1GaAs-GaAs Double Heterostructure Solar Cells. IEEE Pho-tovoltaic Specialists Conference, 20th, Las Vegas, NV,Sept. 26-30,1988, IEEE New York,1988, pp. 446-450.

14.Landis, G.A.; Bailey, S.G.; and Flood, D.J.: Advancesin Thin-Film Solar Cells for Lightweight PhotovoltaicPower. Space Power, vol. 8, no. 1-2,1989, pp. 31-50(Also NASA TM-102127,1989).

15. Kohout, L.L.: Cryogenic Reactant Storage for LunarBase Regenerative Fuel Cells. NASA TM-101980,1989.

16.Landis, G.A.: Solar Power for the Lunar Night. SpaceManufacturing 7; Proceedings of the Ninth Princenton/AIAA/SSI Conference, Princeton, N.J. AIAA, Wash-ington, D.C., 1989, pp. 290-296.

17. Hickman, J.M.; and Bloomfield, H.S.: Comparison ofSolar Photovoltaic and Nuclear Reactor Power Systemsfor a Human-Tended Lunar Observatory IECEC-89;Proceedings of the Twenty-Fourth Intersociety EnergyConversion Engineering Conference, IEEE, New York,1989, Vol. 1, pp. 1-5 (Also NAS A TM-102015,1989).

8

National Aeronautics and Report Documentation PageSpace Admin stration

1. Report No. 2. Government Accession No. 3. Recipient's Catalog No.

NASA TM-103642

4. Title and Subtitle 5. Report Date

Design Considerations for Lunar Base Photovoltaic Power Systems

6. Performing Organization Code

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

J. Mark Hickman, Henry B. Curtis, and Geoffrey A. Landis E-5823

10. Work Unit No.

326-81-109. 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 CodeWashini-,ton, D.C. 20546-0001

15, Supplementary Notes

Prepared for the 21st Photovoltaic Specialists Conference sponsored by the Institute of Electrical and ElectronicsEngineers, Kissimmee, Florida, May 21-25, 1990.

16 Abstract

A survey was made of factors that may affect the design of photovoltaic arrays for a lunar base. These factors,which include the lunar environment and system design criteria, are examined. A photovoltaic power systemdesign with a triangular array geometry is discussed and compared to a nuclear reactor power system and apower system utilizing both nuclear and solar power sources.

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

Photovoltaic Unclassified - UnlimitedPower Subject Category 91Lunar base

19. Security Classif. (of this report) 20. Security Classif. (of this page) 21. No. of pages 22. Price'Unclassified Unclassified 10 A02

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