Space isotopiC power systems

With the technology sound and growing, and units already built for missions ranging from 120 days to 5 years, the designer can and should plan appropriate space application of isotopic systems

BY CAPT. R. T. CARPENTER, USAF U.S. Atomic Energy Commission

A new space power system technology technology, and aerospace nuclear Concurrently, the terrestrial appli­ -isotopic power-has developed to safety technology contributed by the cations-the Snap-7 programs-sus­ the point where it can and should be program and used as a foundation for tained the isotopic power development considered by the space-vehicle de­ follow-on space isotopic power-system program and promoted the fission­ signer for use in many types of mis­ developments. product separations and processing sions. Because of this sound technical capability that exists within the Com­ The' Atomic Energy Commission's basis, the Commission's space-oriented mission today! The interest among isotopic space power program dates isotopic power development program terrestrial power users-the Navy, back to several years before Sputnik has made a steady, although some­ Weather Bureau, and Coast Guard­ I, but the program suffered a severe times slow, comeback through a series was sufficiently strong to sup:port this setback in 1959 when the Snap-1A of events since 1959, so that today a . fuels production program, whe:r:eas the generator development program was program technically comparable to interest in Snap-1A had been inade­ cancelled.' This pioneer program was Snap-1A could once more be under­ quate. At the same time, significant not completed because it may have taken with a high probability of suc­ quantities of the long-lived alpha­ been tt;lO ambitious for its day. The cessful completion. This series of emitter fuel, plutonium-238, were be­ need for isotopic power had .not yet events can help demonstrate the status ing produced so that it could be allo­ become apparent to space program of today's space isotopic power pro­ cated to low-powered space systems. planners; its place and full signifi­ gram. Details of the various systems Relatively little direct radiation haz­ cance in the nuclear space power pro­ have been described many times and ards are associated with alpha-emitter gram were not clearly established; its will not be repeated here. For refer­ fuels compared to beta-emitter fission applicable thermoelectric energy con­ ence purposes, the characteristics of products. version technology was still very new; several. space isotopic power systems Because of these background efforts, large quantities of isotopic fuel ma­ are given in the table appearing on it was possible to fabricate, fuel, test, ter'ials were not readily available; and page 70. and get approval to use the first plu­ the operational safety of large quanti­ While efforts were being made to tonium-238 fueled Snap-3 generator ties of radioactive material in space interest space power-system users in on the Navy's Transit-4A navigation vehicles was a brand new unknown in isotopic systems, the first practical satellite in June 1961, with a lead time a space program already full of un­ radioisotope-fueled thermoelectric gen­ of only five months. The launches of knowns. erator-Snap-3-was being subjected the 2.7-w, Pu-238-fueled Snap-3s on That the demise of the Snap-1A pro­ to exhaustive electrical tests; shock, Transit-4A and Transit-4B (in No­ gram was not due to a lack of tech­ vibration, acceleration and thermal­ vember 1961) were not just "firsts in nical soundness is evident when one vacuum environmental tests; and fil'e, space."5 The experience gained in in­ looks at the thermoelectric generator explosion, impact and re-entry burnup tegrating these units into satellites, fabrication technology, isotopic-fuels nuclear safety tests!" the flight-test data obtained from

CAPT. R. T. CARPENTER is Deputy Chief of the AEC's Isotopic Power Branch in the Division of Reactor Development. He has been manager of space and terrestrial power development projects with the Commission's isotopic power program since April 1959. Capt. Carpenter's previous assignments include responsibilities as R&D project engineer at the Chemical Corps Biological Labs and at Wright Air Development Center. He holds a B.S. in chemical engineering and an M.S. in nuclear engineering. 68 orbit, and the interagency procedures aspects. The first electrically heated have been developed under the and policies established prior to their Snap-9A generator was operated six Snap-9A program without an inter­ launch produced in a matter of months months after go-ahead from the Navy vening flight test. a state of the art for isotopic space and award of the contract to Martin Snap 11, a 25-w RTG being devel­ power which probably would have Co. by AEC. The first flight-accept­ oped for use on NASA's Surveyor soft taken several years with a much able generator was fueled with Pu-238 lunar-landing missions, has also con­ larger unit, such as Snap-lAo nine months after go-ahead. It is tributed significantly to isotppic space The generator on the Transit-4A scheduled for use this year. power systems technology.s After a satellite continues to produce enough During the past year, a substantial design study and a preliminary safety power to transmit the low-powered amount of data has been obtained analysis had been completed, NASA Doppler navigation signals to earth from ground tests of the electrically established a requirement for the after being in orbit almost two years. heated and fueled generators. Impor­ Snap-11 generator development pro­ Because of a failure in the satellite's tant results were obtained from long­ gram late in 1961. During the past telemetry system a few weeks after term thermal-vacuum tests of the year, a detailed design was completed launch, quantitative data on the per­ units under simulated space operating that would meet all the interface re­ formance of the generator are no conditions. The generators were found quirements of the Surveyor spacecraft. longer available." Qualitative perform­ to be very stable power sources when These included the electrical, physical, ance of the unit is still being moni­ subject to sunlight and shadow con­ nuclear radiation, and thermal inter­ tored by the Transit tracking stations. ditions for a 600-mi. polar orbit. The face specifications." The electrical out­ Telemetry data from Transit-4B in­ high heat capacity of the compact gen­ put can be easily matched to the pay­ dicated the radioisotope thermoelec­ erator allowed a gradually changing load through a DC-to-DC voltage con­ tric generator (RTG) performed pre­ surface temperature and power out­ verter similar to that used with con­ cisely as expected for about eight put under these conditions, intlicating ventional power supplies. months. In early June 1962, an abrupt the solar input had little effect on the The physical limitations of the ve­ drop in generator output voltage was system. The Pu-238 fuel pr:ovides an hicle naturally dictate the size, weight, oeserved. During the next week, the unalterable source of heat far more and shape of an RTG. For the Sur­ voltage came right back up to the nor­ stable than an electrical heater. Power veyor program, it was decided to ex­ mal operating value of 2.1 Y, dropped degradation was observed during tests tend Snap-11 out from the spacecraft again to millivolts, came back up and of the fueled generators that was too (because of overriding thermal con­ then dropped to and stayed at prac­ small to detect with the variations in siderations) so that an optimized RTG tically zero voltage. This cyclic be­ line voltage, etc., experienced with configuration could be used. The sep­ havior is not characteristic of an RTG electrically heated units. Although ex­ aration distance and provisions for failure. Upon analysis of the data, it tremely small, these changes would be shielding in the design of the curium- was concluded that a capacitor across significant over the five-year design 242 fuel capsule will allow Snap-11 to the input to the DC-to-DC voltage lifetime of the generator. meet the extremely stringent back­ converter had shorted out. The solar The cause of this power degradation ground radiation levels specified for power supplies aboard Transit-4B was quickly diagnosed because similar the sensitive radiation detectors failed soon after the high-altitude nu­ but more easily detected power losses aboard the spacecraft. Thermal inte­ clear test of July 9, 1962, and signals had been observed in the Snap-7 gen­ gration problems were most severe are no longer being received from the erators a few months earlier. The ma­ and caused abandonment, for the pres­ satellite: terials used in the generators were ent, of a design for conducting heat One of the lessons learned from outgassing in a time-temperature de­ to the sensitive payload instruments these early flight tests was the im­ pendent manner, causing a ' greater during the cold lunar night. portance of integrating the nuclear heat loss to bypass the thermoelectrics. A thermal mockup of the Snap-11 unit into the payload as soon as pos­ To counteract this, the generators has been fabricated and is undergoing sible. Because the Snap unit was sub­ have been baked at higher tempera­ tests. Electrically heated prototype stituted for a solar-cell panel at the tures and filled with inert gases at generators will be available for inte­ last moment, only a limited number of higher pressures to reduce the effect gration tests later this year. Because telemetry channels were available to of the outgassing on generator per-' of launch vehicle problems, Snap-11 monitor the generator. Only the out­ formance. is not scheduled to fly before 1965, put voltage and surface temperature The temperature cycling experi­ unless the results of earlier solar­ were monitored. Knowledge of the out­ enced by the generators during these powered Surveyor spacecraft dictate put current at the time of failure ground tests was far more frequent otherwise. probably would have confirmed the and severe than that which would oc­ The significant differences between failure mode: cur in space. This thermal cycling in­ Snap-11 and Snap-9A are due pri­ ~hese sUccessful flight tests of ex­ creased internal impedance of the marily to the different mission lifetime penmental RTGs led to the early ini­ generators and consequently resulted requirements (120 days vice five t' t ' Ia Ion of the Snap-9A program to de- in a mismatch between the RTG and years) and, therefore, the different velop a 25-w RTG to deliver all of the the payload. The fix for this problem fuels employed. The short-lived Cm- POWer for the Navy's operational came from the more advanced ther­ 242 fuel requires a power-flattening ~rototype Transit-5 navigation satel­ moelectric fabrication technique used feature to remove excess heat gener­ lItes. The unit is shown on page 70. for Snap-11 which had been developed ated early in the mission. Thus Snap-9A, the first isotopic power sys­ and tested since the design for Snap-l1 incorporates a temperature­ te~ developed for a specific space ap- Snap-9A was frozen. This new tech­ controlled, liquid metal-actuated heat­ phcationS ,uses t ec h nology gamed . from nique provides a lower internal im­ dump shutter which starts open and nap-1A thermoelectric module de­ pedance for RTG, and makes it less then gradually closes as the isotope velopments and the experience of the subject to change during prelaunch decays in order to maintain a con­ ~aunches of the Pu-238 fueled Snap-3 handling. Snap-9A was reworked to stant temperature on the' hot junction In v h' I' . dr e IC e mtegration, ground-han- incorporate these improvements. Thus, of the thermoelectric generator and a Ing, and aerospace nuclear-safety in effect, two generations of RTGs constant power output. This technique May 1968 69 SPACE ISOTOPIC POWER SYSTEMS Power Output, Weight, Size Isotopic Design Operational Designation Use (W)* ( , b) (in. OD X in. ht.) Fuel Life Date

Snap·1A Air Force satellite 125 175 24 X 34 Cerium·144 1 year Cancelled in 1959 Snap·3 T hermoelectric 3 4 4.75 X 5.5 Poloniu m·210 90 days Demonstrated in 1959 demonstration PU ·238 Fueled Snap·3 Transit 4A & 48 2.7 4.6 4.75 X 5.5 Plutonium·238 5 years Launched in 1961 satellites Snap·9A Transit 5 satellites 25 27 20 X 9.5 Plutoniu m·238 6 years 1963 Snap·11 Surveyor soft lunar 21-25 30 20 X 12 Curium·242 120 days 1965 landing Sna;:> 13 Thermionic 12 .5 4 2. 5 X 4 Curium·242 120 days Demonstration in 1964 demonstration 500·w Generator (Therm· Design study only 500 100-175 Curiu m·242 6 months ionic) 500 175-225 Plutonium·238 1- 5 years 500 250-300 Cerium·144 1 year IMP Generator (Thermo· IMP satellite 25 21 22 X 11 X 10** Plutonium·238 1-j years 1964*** eleotric) (design only) Sr·90 Generator Communications 30 20-25 Strontiu m·90 5-10 years 1965 *** (Thermoelectric) satellites 60 40-50 Strontium·90 5-10 years 1965*** 120 70- 80 Strontium·90 5-10 years 1966*** 300 150-175 Strontium·90 5-10 years 1966*"* * Raw power from generator. Voltage converter efficiency 75-85% not included. *' In. length X in. width X in. height. *** First use in space for planning purposes.

is similar to the one developed under emitter) fuel production program for prove long-term reliability, and the the Snap-1A program. The other dif­ space power needs. shape of the generator and radiator ference is in the safety criteria used Recently, a design effort was under­ fins has been tailored to fit into the for designing the Snap-ll fuel cap­ taken for an isotopic power system existing vehicle design. sule. In the event the unit should re­ for use on the Interplanetary Monitor­ For the most part, the IMP gen­ enter the earth's atmosphere, it will ing Probe (IMP) satellite being de­ erator uses proven materials and tech­ be designed to remain intact instead veloped by NASA's Goddard Space niques, the only changes made being of burning up. The short-lived fuel Flight Center, shown on page 71. This those required for system integration. will decay to practically nothing be­ program also involved vehicle integra­ For example, one of the primary ob­ fore re-entering from the parking tion problems since the RTG had to jectives of the IMP satellite is to map orbit.. The fuel capsule is designed to fit on a spacecraft designed to use a the magnetic field between the earth prevent widespread contamination of solar power supply. The power system, and the moon, so a very sensiti\'e mag­ the lunar surface in the event of a if adopted by NASA, will include two netometer is included aboard the ve­ vehicle malfunction leading to high­ RTGs placed on opposite sides of the hicle. Therefore, some of the materials velocity impact on the moon. spacecraft to maintain proper weight used in the RTG and the electrical Because of the Snap-ll fuel re­ and balance for stabilization. The IMP circuitry of the generator had to be quirement, a Cm-242 production ca­ generators will each produce appr oxi­ changed to reduce the generator's ef­ pability has been established at Oak mately 25 wand will be fueled with fect on the magnetic field. Ridge National Laboratory. The first Pu-238 because of the longer than The purpose of this brief review fuel-load quantities of this extremely one-year mission lifetime. These RTGs has been to review the current state high power-density material will be incorporate design improvements over of the art of space isotopic power produced later this year. This Cm-242 Snap-9A which provide for easier fab­ systems. It can be summarized as fol­ development program is also contrib­ rication. and lower system weights. lows: uting supporting technology applica­ The number of seals required in the To date, low-powered systems (up ble to a curium-244 (long-lived alpha generator has been reduced to im- to 25 w) have been developed and

25-watt radioisotope thermoelectric generator for use on the Navy's opera· tional prototype Transit.5 navigational satellite. The first flight.acceptable generator was fueled with Pu·238 nine months after go­ ahead, and is scheduled for launch this year. 'fi d for operational use. Only the 4. Thermionic power conversion be used only for missions with a life­ qua lIe. f "safest" alpha-emItter uels have been technology-to demonstrate the capa­ time of one year or longer, except resent RTG's display a power- bilities of radioisotope-fueled therm­ where they can be recovered and re­ d P use. . ht ratio of a b out 1 w / lb. Iso- ionic power systems. used, as in a manned space capsule. to-welg. . . fueled statIc thermoelectrIc con- 5. Aerospace nuclear safety tech­ Long-lived alpha emitters can also be tOPIC-. n systems exh'b't I I a h'19 h d egree nology-to provide for safe employ­ used over and over for prelaunch versIO l'f t' of reliability. The I e Imes to be ex- ment of larger and more powerful nu­ checkout of systems designed to use ted are a function of the isotope clear devices in space. beta emitters. As alpha emitters be­ ~e~l used. Units have been designed Now how does the current technol­ come more abundant in the last half a~d built for mission lifetimes rang­ ogy relate to future isotopic power of this decade, they can be considered 'ng from 120 days to five years. The system developments and applications. for use in more or larger power sys­ ~ystems are flexible and can be inte­ Radioisotope fuel properties and avail­ tems.' grated into almost any type of space ability determine, to a large extent, For those unmanned missions which vehicle. Lead times have been ex­ the fuel selected for a given applica­ do not include sensitive radiation de­ tremely short in many cases, but tion.Io Because the supply of alpha­ tectors, it seems advisable to use the launch schedules have been met. Inter­ emitter fuels will be limited for the fission product, beta-emitter fuels­ agency cooperation and support have next few years, their use in high­ cerium-144 (for mission lifetimes up been successfully demonstrated. powered systems, or for missions re- to one year) and strontium-90 (for

Interplanetary Monitoring Probe design study for NASA, illustrated here, employs two radioisotope thermoelectric generators, extended from opposite Sides of the spacecraft to maintain proper balance for its stabilization.

In addition to the prototype gen­ qui ring a large number of satellites, mission lifetimes greater than one erator developments already described, will be limited. Alpha-emitter fuels year). These isotopes are being pro­ sUPPorting R&D efforts are being pur­ are best used for planetary probe mis­ duced in large quantities today, are sued by AEC in the following areas: sions or on other spacecraft where relatively inexpensive to process, and .1. Isotopic fuels technology-to pro­ low radiation levels are mandatory. can be compounded in high-tempera­ vI~e fuel forms with higher melting They can also be used for low-altitude ture fuel forms which can be readily pomts· h and h'Ig h er power d'ensltIes . satellite missions where lifetime and dispersed at high altitudes during re­ h w f IC s t"llI meet aerospace nuclear weight are prime considerations and entry into the earth's atmosphere. sa ety criteria. radiation effects of the artificial belts The disadvantages of the beta­ . 2. Isotop e pro d'uctwn and process- make it impractical to use solar cells. emitter fuels, compared to the alpha mg cap b"l" abil"t a 1 lty_to improve the avail- Th3 short-lived alpha emitters-po­ emitters, are their low power density l fuel Y and lOwer the cost of isotopic lonium-210 and curium-242-are good and the direct radiation emitted as s. for mission lifetimes up to six months X-rays, or bremmstrahlung, which 3. Therm I tech I oe ectric power conversion which justify the use of isotopic sys­ given off as the beta particle's energy no ogy_t . . ll ature 0 provIde hIgher temper- tems over competitive systems. is transformed into heat and gamma weigh~ :~ore efficient and/or lighter The long-lived alpha emitters­ rays. Low power density means the ermoelectric systems. curium-244 and plutonium-238-should system will be larger and somewhat May 1[)63 71 heavier. The direct radiation requires in excess of 500 w will be required, consideration during ground handling there may be some advantages in con_ engineers • scientists and launch operations. sidering the use of isotopic power An AEC program is underway to systems, providing that adequate lead demonstrate safe ground handling and time is allowed for AEC to make the NfWCARffR launch procedures of a beta-emitter necessary fuels available. fueled unit. Procedures of this kind The time to factor an RTG into a were developed under the Snap-1A payload is during the design phase. POSITIONS program, and have been improved to Then one can take advantage of the be more compatible with launch pad unique characteristics of isotopic operations and different launch ve­ power systems, such as making use of at Fast Moving hicles. Generally, if proper considera­ the predictable heat source to posi­ tion is given to self-shielding of pay­ tively control the payload tempera­ Atlantic Research load components and to the use of ture. Space isotopic power systems separation techniques, if necessary or are still considered very much in the Corporation possible, the radiation from these R&D stage, but so is the space pro­ beta-emitter fueled power sources re­ gram. Since there are many variables quires little or no additional shielding in the selection of an isotopic system; ARC slarled operations fourleen years ago to protect electronic instruments dur­ since the development cycle is short and loday employs more Ihan 2,500 persons. Our organizalion may besl be described as an ing operation in space. Use of t hese (two to three years); and since de­ advanced-Iechnology company wilh highly more abundant beta emitters per mits velopment costs are relatively low, diversified inlerests and special emphasis on solid-propellanl rockelry. Conlinued ex­ consideration of isotopic power for there are obvious advantages in de­ pansion has crealed several oulslanding pro­ higher powered applications, for ex­ veloping an optimized power system fessional opportunilies including Ihe following: ample; some scientific satellites, and for a given application and making AERODYNAMICIST operational systems requiring many use of this rapidly advancing tech­ Will delermine aerodynamic loading and per­ formance of mullistage rockels in Ihe satellites in orbit, such as communi­ nology . . 'O 1 velocity range from static 10 hyper­ cation and meteorological satellites. • ' Isotopic power will not answer all sonic; establish analy.lical techniques; recommend model design and tests; Power-to-weight ratios for RTGs the problems facing a space power supervise tests and be responsible for are being improved. For alpha-fueled system designer. However, it stands analysis of test results from models lIb ready to be used for those many space to full-scale flighls. as or MS in systems, 2 to 3 w is achievable with aeronautical engineering. Los Angeles area. 25 to 250 w generators within the missions where it is applicable and TEST PLANS GROUP HEAD next two to three years, by using where other systems cannot or should To lead Test Group in eslablishmenl of tesl thermoelectric materials and fabrica­ not be used. mel hods and techniques for delermining tion techniques already under devel­ spectrum signature of communication, navi­ References gation and radar equipmenl on all frequency opment in the supporting R&D pro­ ranges. Requires good knowledge of basic grams cited above. Beta-fueled sys­ 1. Harvey, D . G. and Morse, J ~ G., "Radio­ principles of operalion of this equipment and nuclide Power for Space Missions," Nucleonics, slale-of-Ihe-art instrumenlation to perform t ems'will weigh 1.5 to 2 wllb for com­ Vol. 19, No.4, pp. 69-72,April 1961. power, frequency and spectrum analyses parable power levels. These gains are 2. Morse, J . G. and Harvey, D. G., "Radio­ measurements. Responsibilities include expected with no loss in reliability, isotope Auxiliary Power Systems," Aerospace monitoring of tests and verification of results. Engineering, Vol. 20, No. 11, p. 8, Nov. 1961. as eleclrical engineering wilh applicable ex- and with some improvement in con­ 3. Wilson, R. J., "Operational and System . perienee. version efficiency of the thermoelectric Testing of Snap-3 Thermoelectric Generator,n TEST ENGINEER Martin Co. Paper RT-61-2-37 presented at the Will plan and design tesl programs for rockel generators. Joint Technical Society, D ept. of D efense, Sym­ propulsion systems; prepare sched ules and Developments in the thermionic posium on Thermoelectric Energy Conversion, coordinale program to meet schedules; R&D programs indicate that, by the Jan. 1961. analyze test results and prepare tesl reports; 4. Davis, H. L., "Isotope Costs and Avail­ plan and coordinate environmental test pro­ time the high power-density alpha ability," Nucleonics, Vol. 21, No.3, pp. 61-65, gram. as or MS in ME or physics. emitters become more available in the March 1963. . 5. Fischell, R. E ., "Nuclear Powered Thermo­ ROCKET TEST SUPERVISOR 1966-67 period, thermionic isotopic electric G enerator for the Transit Satellites," Will coordinate work of instrumentalion staff power systems will also offer improve­ ARS2201-61, presented at the American Rocket with olher segments in tesl area and through­ Society Space Flight Report to the Nation, New out ARC. Responsibilities include design, ments in specific power ratios and con­ York, N.Y., Oct. 9- 15, 1961. selection, modification and operation of veJ;,sion efficiencies. Thermionic sys­ 6. Dick, P. J. and Davis, R. E., "Radioisotope instrument components and instrumentation Power System Operation in the Transit Satel­ systems used in stalic testing of propulsion tems are farther away because of the lite," presented at the AlEE Aero-Space Trans- , units. Will also plan, organize and direct higher operating temperatures in­ portation Conference-1962 Summer General activities of instrumentation technicians in volved and because systems fueled Meeting, Denver, Colo., June 1962. calibralion of equipment and carrying out of 7. ".space Nuclear Power Applications- 1962," lests. as or MS in EE or physics and with beta emitters are not competitive Hearings before the Joint Congressional Com­ experience, preferably in rocket instrumenta­ with thermoelectric systems in weight, mittee on Atomic Energy, U.S. Government lion field. Printing Office, Washington, D.C., page 121, OTHER POSITIONS efficiency, or reliability. However, Sept. 13, 14, and 19, 1962. .. . are available to a Sr. Controls Engineer, thermionic systems will be smaller 8. Streb, A. J., Bustard, T . S., and Wilson, R. J., "Nuclear Auxiliary Power Unit for Lunar Reliabilily Assurance Supervisor, Electrical than thermoelectric systems at com­ Test Engineer, Electronic Systems Engineer. Exploration," Transactions on Nuclear Science, parable power levels. NS-9 Vol. I, Jan. 1962, Las Vegas, Nev., Oct. 1961. Send resume to: Depl. 243 There is no magic limit to the power 9. Carpenter, R. T. and Harvey, D. G., Director, Professional Personnel levels for isotopic power systems. "Integrating Isot-Opic Power Systems," Astro" There is, however, an optimum powel'­ nautics, p. 30, May 1962. 10. Harvey, D. G. and Carpenter, R. T., Atlantic Research to-weight ratio for each power level "Transit and Beyond," lAS Paper 62-61, pre­ for a given generator concept. There­ sented at the 30th Annual Institute of the Aerospace Sciences Meeting, New York, N.Y., ~?x~~~:~~~~~ In fore, it might be desirable to use two Jan. 22, 1962. . (A residential suburb ~ or more optimum units for higher 11. Snyder, N. W., "Power Systems," Astro­ of Washington, D.C.) ~ power requirements. For a few rather nautics, pp. 1lO-114, Nov. 1962. 12. Morse, J. G., uNuclear Power Sources for An equal opportunity employer unique missions, such as manned lunar Communications Satellites," pp. 57- 62, Proceed­ exploration or Mars and Venus or­ ings of the 1962 Annual Conference of the Atomic Industrial Forum, Washington, D.C., biters or landers, where power levels Nov. 1962. •• 72 Astronautics and Aerospace Engineering