The Space Congress® Proceedings 1967 (4th) Space Congress Proceedings

Apr 3rd, 12:00 AM

An Undersea Radioisotope Power Supply

H. C. Carney Aerojet-General Corporation

Follow this and additional works at: https://commons.erau.edu/space-congress-proceedings

Scholarly Commons Citation Carney, H. C., "An Undersea Radioisotope Power Supply" (1967). The Space Congress® Proceedings. 4. https://commons.erau.edu/space-congress-proceedings/proceedings-1967-4th/session-18/4

This Event is brought to you for free and open access by the Conferences at Scholarly Commons. It has been accepted for inclusion in The Space Congress® Proceedings by an authorized administrator of Scholarly Commons. For more information, please contact [email protected]. AN UNDERSEA RADIOISOTOPE POWER SUPPLY

H. C . Carney Aerojet-General Corporation San Ramon, California

St.mnnary

The results and current status of an Aerojet­ cable repeaters, etc to support expanding ocean sponsored program to develop a product line of low­ engineering activities . The first lead shield, power radioisotope thermoelectric generators for radioisotope-fueled URIPS has been fabricated and marine applications are discussed . A 1-watt( e) tested, and was publicly exhibited at the Off­ Undersea Radioisotope Power Supply ( URIPS) has been shore Exploration Conference (OECON ) at Long designed, fabricated, and tested . Ex tensive para­ Beach, California in February of this year . A metric studies were performed to select optimum de­ second lead shield URIPS has been purchased by sign characteristics; typical parameters investi­ the Navy and, following qualification testing, gated wer e : r adioisotope and chemical fuel form, will be delivered to the Naval Civil Engineering thermoel ectr ic material, fuel capsule L/D ratio, Laboratory at Port Hueneme, California this s h iel d materia l and geometry , and operating condi­ sununer. A compact, lightweight uranium-shield tions . The design emphasizes high reliability and URIPS is currently in the final fabrication stage, low cost, but considerable emphasis was placed on and will be subjected to long term endurance test­ adaptability to meet a wide spectrum of user re­ ing at Aeroj et. quirements . URIPS is designed to provide a steady power output for a minimum of five years at ocean Design Considerations depths up to 20 , 000 feet . An RTG consists, basically , of an encapsu­ Introduction lated radioisotope heat source, thermal insulation, biological shielding, a thermoelectric converter, In 1965 , Aerojet-General Corporation initiated a power conditioner, and various structural com­ a company-sponsored program to develop a product ponents. line of low -power radioisotope thermoelectric gener­ a tors (RTG) for marine applications . The initial Thermal energy from the radioi sotope heat objective of this Undersea Radioisotope Power Supply source is converted into low voltage d -c electri­ Program was to design , fabricate, and test a highly cal power by the thermoelectric module . Thermal reliable power supply capable of providing a continu­ insulation is required to minimize parasitic heat ous output of 1 watt(e) for a minimum of 5 years in loss. Radiation originating from the radioactive ocean depths up to 20,000 feet . decay process requires the provision of a biologi­ cal shield. A power conditioning sub-system is Conventional primary chemical batteries, al­ required to transform the low voltage, low current though less costly than an RTG have an endurance output of the thermoelectric module to the higher limit of sev eral y ears; RTG systems have demonstra­ potential required by the load . Other functions ted the capability for 5 years of operating life of the power conditioner are to regulate the out­ and show promise of satisfying 10 to 20 year re­ put voltage and power, by compensating for the quirements . Optimum use of an RTG is in missions fluctuation over the mission lifetime associated requiring a long-term unattended electrical power with the decay of the radioisotope and with the supply . For example , the longer life of the RTG degradation in performance of the thermoelectric more than offsets the lower capital cost of con­ module. The electrical power can be used to ventional bat teries in cases where: charge a secondary battery system , or, alternate­ ly, an energy storage capacitor can be used , if 1 . Costs associated with battery replace- required, to provide a low impedance source for a ment are high, as in remote areas where access is pulsing lamp or transducer load. difficult . The interrelationship between the design 2 . A complete battery-powered system must variables associated with each of the components be periodically replaced because the system is not discussed above must be defined and systematical­ recoverable, resulting in high capital costs . ly evaluated within the framework of established design goals and criteria to arrive at an optimum 3 . Power interruptions or physical dis- RTG design. The major URIPS design goals were low turbance during periodic replacement operations cost, and high reliability . The design criteria, are unacceptable . while of similar importance but which are in some cases subject to overriding constraints. imposed URIPS was developed to satisfy the indicated by the design goals, were: requirement for small , economical radioisotope t hermoelectric generators suitable for powering scientific instrumentation, acoustic transponders,

18-33 60 1. Minimum size and weight. shield thickness required for Co is approximate­ ly twice that for sr90 for this low power system, 2. Optimum design for a power range capa- which results in a considerably heavier shield and bility from 150 milliwatts to 5 watts. more costly system. The titanate fuel form of strontium (SrTi03) was selected because of its ad­ 3. Adaptable to a wide variety of appli- vanced development status, availability in suffi­ cations with minimum design modifications. cient quantity, and relatively low solubility in water, which is advantageous from the hazards 4. Minimum unattended lifetime of 5 years standpoint. and potential capability of 10 to 20 years. Thermoelectric Material 5. Capable of immersion in sea water at depths up to 20,000 feet. A large number of thermoelectric materials were evaluated for DRIPS application (see Table 1). 6. Capable of withstanding all forseeable System electrical efficiency, ~E' which includes environmental conditions during storage, transpor­ both the thermoelectric module and power condition­ tation, handling, and operation without mechanical er efficiency, was approximated to provide a tem­ or electrical damage or degradation in rated per­ perature-dependent analytic function for the TEHE- formance. 3 code by using an effective (psuedo) thermo­ electric figure of merit, ZE, in the basic thermo­ 7. Capable of complying with all safety electric efficiency equation for matched load con­ regulations specified in applicable local, state ditions: and federal ordinances, laws, and codes concerning manufacture, transportation, handling, testing, operation, and disposal. ~E T -T 2T + ~ - H c Design Selection H ZE -2-

An extensive parametric analysis of RTG de­ where : sign characteristics and operating parameters was performed during the conceptual design phase of hot junction temperature (°K) . the DRIPS program, to select optimum design cold junction temperature (°K) features consistent with the previously described 0 1 goals and criteria. The results of the studies effective figure of merit ( c- ) concerned with fuel, shielding, thermoelectric selection, and materials, and thermal insulation The effective figure of merit characteristic of are sununarized be­ overall RTG system optimization each thermoelectric material ( see Table 2) was were performed using low. The parametric analyses determined by computing detailed performance maps by Aerojet an Im-1 7094 computer program developed relating electrical efficiency to thermoelectric Large Milliwatt under the AEC sponsored Advanced operating parameters and design characteristics. Program.l This Thermoelectric Heater Generator These computer programs are also described in (TEHE-3) code optimiz;s the pe;formance ~aluation Reference 1. Typical results, showing the system of an RTG and calculates the corresponding physi­ diameter, weight, and efficiency as a function of and associated costs. cal characteristics ZE for a 1-watt(e) and 150 milliwatt(e) RTG are shown in Figures 1 and 2. Radioisotope Fuel TABLE 1 All radioisotopes currently under develop­ ment by the AEC for heat source applications were CANDIDATE THERMOELECTRIC MATERIALS evaluated. In consideration of the design cri­ FOR DRIPS APPLICATIONS a minimum unattended lifetime of teria requiring Bismuth Telluride, N/P-Type (Asarco) 5 years (and the capability for as long as 20 years), the candidate radioisotopes were restric ­ Lead Telluride, N-Type (MMM -T EGS-3N) with half-lives in excess of 5 years; ted to those Lead-Tin-Telluride, P-Type (MMM-TEGS-3P) these included Co60, sr90, csl37, Pu238, and Cm244 Cobalt-60 and strontium-90 proved to be Lead Telluride (MMM-TEGS-2N/2P) for DRIPS application on the the most competitive MCC 40, N/P-Type (Monsanto) basis of fuel cost and availability. Strontium-90 was selected since it results in the lowest over­ Silicon Germanilllll Alloy, N/P-Type (RCA) weight, and size. Most of the all system cost, Cupron Special (Wilbur B. Driver) Co-60 decay energy is released as gamma photons, and since these are penetrating radiations, much Tophel Special (Wilbur B. Driver) is deposited outside the fuel of the energy Platinel 7674 N (Engelhard Industries) capsule in surrounding material. Consequently, a thermal shield equivalent to approximately one Platinel 5355 P (Engelhard Industries) sur­ inch of uranium must be placed immediately Chromel - P (Hoskins) rounding the fuel to absorb the photon energy . This shield reduces the effective power density of Constantan (Driver-Harris) the heat source (which increases system size and parasitic thermal losses), and reduces design flexibility. In addition, the total biological

18-34 TABLE 2 Biological Shield PERFORMANCE PARAMETER FOR THERMOELECTRIC MODULES Minimum RTG size and weight are always at­ tractive features because such characteristics Effective Figure of Merit, normally simplify user integration problems. Two z x 103 °c-l typical shield materials were evaluated: lead, Thermoelectric Material E which is representative of minimum cost for a Cupron/Tophel Special 0.10 reasonably compact, integral shield; and uranium, which minimizes size MCC-40 0.25 and weight. Three basic shield configurations were investigated: SiGe Alloy 0.30 internal shield; external shield; and split PbTe (3N/3P) 0.40 shield (a partial internal uranium shield and partial external uranium or lead shield), as SiGe-PbTe (2N/2P) 0 .45 illustrated in Figure 3. Lead cannot be used as Cascaded an internal shield because the fuel capsule tem­ peratures PbTe (2N/2P) 0.55 typically exceed the melting point of lead. All shields were sized to result in a PbTe (2N/2P) - Bi Te 0.65 radiation dose 2 3 rate no greater than 200 mr/hr at Cascaded the surface of the RTG, which is consistent with ICC/AEC shipping regulations. The results of 0.90 these investigations, presented in Figure 4, lead to the following conclusions:

The metallic thermoelectric materials are in­ 1. Minimum RTG diameter is obtained with efficient, and result in a relatively large system an external uranium shield. size, weight and cost . The metallic materials are best suited for low milliwatt (less than 150 mw) 2. Weights of the external and internal applications, where current fabrication limitations uranium shield are approximately equal for a restrict the performance capability of semiconductor 1-watt(e) RTG but uranium split shields are materials. Silicon-germanium is basically a high heavier than either an external or internal urani­ temperature material which cannot be usefully em­ um shield. ployed in a low power RTG because of the compara­ tively large parasitic heat losses associated with 3. Partial external lead shielding in- the temperatures required to achieve high thenno­ creases the system weight, reaching a value about electric conversion efficiency. Lead telluride and twice that of uranium for the full external lead bismuth telluride were found to be the most competi­ shield. tive material currently available for URIPS use. Segmented and cascaded systems which perform better In addition, the efficiency of an external shield than the simple non-segmented thermoelements were system is approximately twice that of an internal investigated; however, the segmented technology shield because the parasitic heat losses are mini­ is currently in the developmental stage and not mized by placing the insulation immediately yet suitable for application in commercial low around the heat source. This configuration re­ power systems requiring high reliability and long sults in minimum system cost, and was therefore endurance. Bismuth telluride was selected for use selected as reference for URIPS. Of equal import­ in URIPS in preference to lead telluride for the ance, any one of a number of shield materials (in­ following reasons: cluding an expedient, rather than an integral shield) can be applied as the external shield with 1. RTG cost, size, and weight are less minimum design modifications, since thermal manage­ for a power range up to 2 watts and competitive ment is accomplished, primarily, inside the shield. to 5 watts. Selection of a compact uranium shield or low cost lead shield URIPS depends on the user's cost/ 2. Bismuth telluride operates efficiently effectiveness requirements. at low temperatures, which maximizes the overall system reliability. Fuel Capsule Length/Diameter Ratio

3. Thermoelements with large length-to- The effect of fuel capsule L/D ratio on RTG area (L/A) ratios have been fabricated into com­ performance was investigated to determine the opti­ mercially available compact thermoelectric modules mum values for minimum weight and maximum effic­ which match the electrical load characteristics of iency (which are directly related to minimum low power RTG and provide a relatively high output system cost). Results of the TEHE-3 analysis are voltage and correspondingly high power condition­ summarized in Figure 5 for a 1-watt(e) RTG. A ing efficiency. fuel capsule L/D ratio of 1.2 is optimum; however, weight and efficiency do not vary appreciably 4. The relatively low operating tempera- throughout an L/D range of 1.0 to 1.5. ture permits separate encapsulation of the thermo­ electric module from the thermal insulation system in a high purity, inert gas environment without excessive parasitic heat losses. 18-35 Thennal Insulation

Thermal conductivity is the most important heat losses. The as-received insulation is in physical property in the selection of insulation block form and contains a small percentage of materials for RTG application. A low conductivi­ volatiles, consisting of phenolic binder com­ ty is required to minimize parasitic heat losses pounds and bound moisture. It is not necessary and the thickness of insulation required. The to remove these volatiles to protect the thermo­ thennal conductivity of a number of insulation electric material, since the power module is materials listed in Table 3, are compared in separately encapsulated and maintained in a high Figure 6. Min-K 1301, backfilled with xenon, purity argon atmosphere. However, the insulation represents the best of the non-evacuated fibrous is heat-treated to remove volatile contaminants insulation materials (Min-K 501 is somewhat for the following reasons: better but is limited to a 400-500°F operating temperature). Super Linde Insulation is far 1. The compressive strength is increased. superior to any of the fibrous materials but must be evacuated and, for long duration usage, getter 2. The thermal conductivity is decreased. systems are required to control outgassing and maintain a vacuum of about 10-3 torr. In view of 3. Long-term degradation of thermal pro- the developmental status of vacuum insulations and perties is reduced. of the high associated cost, Min-K 1301 was se­ lected for use in URIPS. 4. The insulation is more easily machined to close tolerances. URIPS Design Description Several URIPS insulation assemblies were The 1-watt(e) uranium-shielded system (URIPS­ subjected to shock and vibration tests. As a re­ Ul) and lead-shielded system (URIPS-Pl) designs sult of these tests, the insulation-capsule sup­ are shown in Figures 7 and 8, respectively. The port structure is qualified to meet the transpor­ cylindrical source capsule consists of a hot­ tation shock and vibration requirements for common pressed strontium titanate pellet encapsulated in carrier as specified by MIL-STD-810A. In addition, successfully sustained a Hastelloy C. The bellows-encapsulated Bi2Te3 the insulation structure thennoelectric converter is pressed against one transverse axis shock pulse of 50g, which was ap­ end of the heat source with contact pressure plied to simulate a shipboard handling event. applied by springs in the converter. The source capsule is radially and axially supported by the Thermoelectric Converter Min-K 1301 thermal insulation which surrounds the heat source and thermoelectric converter. The in­ The thermoelectric converter consists of an sulation space is backfilled with xenon during encapsulated assembly of the bismuth telluride final assembly to minimize parasitic heat loss and power module. The module is encapsulated in a long-term degradation of the thermal insulation. bellows container filled with high pur ity argon, to isolate the module from contaminants evolved by In URIPS-Ul, the uranium acts as both the the thermal insulation. In addition to improving biological shield and pressure vessel, but a steel the lifetime and reducing performance degradation pressure vessel is required in URIPS-Pl. Both of the thermoelectric module, separate packaging systems are encased in a 70/30 copper-nickel facilitates pre-assembly performance a nd qualifi­ (0.5 Fe) alloy housing for protection against the cation testing of the converter. corrosive sea water envirorunent. The URIPS converter is comprised of two Radioisotope Heat Source Asarco TH1010 thermagotron power generation modules thermally and electrically-connected in parallel. The 46 watt heat source consists of 7200 These miniaturized modules incorporate close-packed, curies of strontium-90 titanate. The hot pressed sintered bismuth telluride thermoelements. A photo­ fuel is sealed into a Type 304L stainless steel graph of the TH1010 module and converter assembly is liner and encapsulated in Hastelloy C. The capsule shown in Figure 10, and module design and perfor­ is designed to withstand an external pressure of mance characteristics are summarized in Table 4. 10,000 psi with negligible permanent deformation The performance characteristics of 22 TH1010 modules of the walls or end caps in sea water for 300 were experimentally evaluated to verify fabrication years (in excess of ten half-lives of sr90), based reproducibility. Four converters have been fabri­ on an average corrosion rate for Hastelloy C of cated and subjected to performance and qualifica­ 1 x lo-4 in./year. Pressure tests showed that the tion tests. empty capsule will withstand an external pressure of 15,000 psi before rupture throughout the same Housing Assembly decay period. Figure 9 shows the basic heat source components. The housing assembly for both the URIPS-Pl and URIPS-Ul is comprised of the main housing, Thermal Insulation housing cap, flange, electrical connector, and lifting lugs. The housing cap is bolted to the The radioisotope heat source and thermo­ flange and a neoprene 0-ring used for the primary electric converter are surrounded by Min-K 1301 seal. The lifting lugs are welded to the main thermal insulation. The insulation, which has a housing, and the body of the electrical receptacle relatively high compressive strength, mechanically and main housing are welded to the flange. supports the fuel capsule and minimizes parasitic

18-36 TABLE 3 INSULATION MATERIALS LIST Max. Cont. Service Gas Density Temperature 3 Interstitial Pressure Data No. Product Name lbs/ft OF Gas ~ Source

lA Improved Min-K-2000 Johns -Manville 14 2000 Air 760 J/M Survey Data lB 14 2000 He 760 lC 14 2000 Air SS lD 14 2000 He SS lE 14 2000 Air or He 8 lF 20 2000 Air 760 lG 20 2000 He 760 lH 20 2000 Air SS lI 20 2000 He SS lJ 20 2000 Air or He 8 2 Min-K 1301 20 14SO Air 760 2A 20 14SO Xenon 760 3A Micro -Quartz Type I 2000 Air 760 J/M Literature 3B 3.S 2000 Air 760 3C 3.S 2000 He . 760 J/M Lit. & G/D Corr. 3D 3.S 2000 Air High Vacuum 4A Micro-Quartz Type II 4.S 27SO Air 760 J /M Survey Data 4B 6.2 27SO Air 760 4C 8.0 27SO Air 760 SA Thermoflex RF 600 2000 Air 760 J/M Literature SB RF 300 3 2000 Air 760 6A Refrasil Batt B-100 H.I.Thompson 3 2000 Air 760 G/D Ft.Worth Rept.

6B II A-100 3.S 2000 Air 760 Hitco Literature 6C B-100-1 6 2000 Air 760 7A S.I.62 Compressed Linde 28 Air High Linde Literature (1 atm) Vacuum 7B Hi. Temp. S.I. ~18 ~1800 Air G/D & Linde Data 8A Foamsil Pittsburgh-Corn. 11 2200 Air 760 Pitt. -Corn. Lit. 8B 10 2200 Air 760 ASD-TDR 62-21S X226-32A Foam American Latex <400 Air 760 G/D Rept.MRG-202 10 Zirconia-A Fiber H.I. Thompson 8 3000 Air 760 Hitco Rept. AF-33 (6 S7) 3902 llA ADL-17 A. D. Little 12 Air 760 Bell Survey Data llB ADL-17 12 Air o.s 12A Fiber Frax Blanket Carborundum 6 2300 Air 760 Carb. Survey Data 12B Fiber Frax Paper 9.S 2300 Air 760 13 Micro -Quartz Type II 6.2 27SO Air 8 G/D Cale. Data & Platinum Foils 14 "San toe el" A Mons an to 1300 Air 760 Monsanto Lit. lS Kaowool Paper B & W 12 - 13 2300 Air 760 B&W Literature 16 Foamglas Pittsburgh-Corn. 800 Air 760 P-C Literature · 17 "Tipersul" Block DuPont 2000 Air 760 DuPont Lit.

18 PKT DuP ont 32 1800 Air 760 19 WRP-X Refractory Prod . 18 2200 Air 760 RPC Lit. 20 Min -K SOl Johns-Manville 10 soo Air 760 J/M Literature

21 Dyna-Quartz 10 27SO Air 760 22 MT-10 Eagle-Picher 800 Air 760 EP Data 23 Min-KF -100-10 Johns -Manville 10 1200 Air 760 J/M Literature 24 Eccosphere& SI Emerson & Cuming, 11 2SOO Air 760 E&C Literature Inc. 2S PV Supcrtemp Tilock Eagle-Picher 1900 Air 760 EP Data 26 Salami Solar Aircraft: 40 1800 Air 760 Solar Data 27 Xenon Gas 18-37 TABLE 4 TABLE 5 CHARACTERISTICS OF ASARSO TH1010 POWER MODULES CORROSION DATA FOR 70/30 COPPER-NICKEL ALLOY Thermoelectric Material Sintered Bismuth Telluride Depth Exposure Density, g/ cc Feet Days Environment MDD MPY N-type 7.70 5640 123 w 7.36 1.11 P-type 6.76 5640 123 M 4.95 0.80 0 Melting Points, c 2500 123 w 3.89 0.63 N-type 2340 197 w 4.39 0. 71 Liquidus 592 2340 197 M 1.29 0.21 Solidus 587 w Exposed in sea water P-type 601 3 ± M Partially embedded in bottom sediment Number of Couples 50 MDD Milligrams per square decimeter per day Electrical connection Series MPY Mils penetration per year Module Dimensions, in. Length o. 750 Width o. 750 Power Conditioning Unit Thickness 0.210 It is difficult to obtain the optimum im­ Element Dimensions, in. pedance match to the load for maximum efficiency with semiconductor-type thermoelements for low Square cross-section 0.060 power output thermoelectric generators because of Length 0.200 materials fabrication limitations. As a consequence, a power conditioning subsystem is required to trans­ Nickel barrier thickness, in. 0.0005 form the low voltage, low current output obtained Copper link thickness, in. 0.005 from an electrically-matched thermoelectric module to the higher potential required by the load . Epoxy perimeter thickness, in. 0.0545 Other functions of the power conditioner are to Insulation thickness, in. 0.005 regulate the output voltage and power by compensat­ ing for fluctuations that occur over the mission Internal resistance (R.T.), ohms 2.9 0.3 ± lifetime due to radioactive decay and performance Performance (TH = 210°c, TC - 30°C) degradation. The electrical circuit used in URIPS to condition the output power from the thermo­ electric converter is shown in Figure 11 . The Open circuit voltage, volts 3.0 - 3.2 power conditioner is completely sealed in plastic. Short circuit current, amps 0.80 - 0.88 Space-rated, high reliability electronic components with certified reliability history, and operation Internal resistance, ohms 3.5 - 3.8 under derated conditions assures a high reliability Thermal efficiency, % 5.0 - 6.0 for this subsystem. Maximum power, watts 0.6 - 0.7 Design Summary

Overall DRIPS design characteristics are sum­ The housing assembly, which includes all com­ marized in Table 6. The first Aerojet Undersea ponents with surfaces exposed to sea water, is Radioisotope Power Supply is shown in Figure 12 fabricated from 70/30 copper-nickel alloy (0.5 Fe). being readied for tests. Development efforts are This material, which has excellent resis tance to continuing to increase the overall system reliabili­ corrosion, pitting, and fouling in sea wa ter, is ty by the application of redundancy concepts in the commonly used for the most severe service where re­ thermoelectric converter. liability and dependability are paramount. The nominal corrosion rate is 0.1 to 1.5 mils per year with a typical rate of penetration in pits of 1 to 5 mils per year for a sea water velocity of 0 to 3 ft/sec. 2 The results of corrosion tests per­ formed on copper-nickel alloy (CDA No. 715) in the Pacific Ocean, 75 to 80 miles west of Port Hueneme, California,3 are shown in Table 5.

18-38 TABLE 6 URIPS DESIGN SUMMARY

Radioisotope fuel SrTi0 3 Power output 1-watt(e), 24-volts Minimum design life 5 years Thermal power (BOL) 46 watts Thermal efficiency (EOL) 2.5% Power conditioning efficiency 80% Shield material Depleted uranium Lead Diameter of system 9.7 in. 13.7 in. Height of system 15.2 in. 19.7 in. Weight 450 lb 800 lb Housing Material 70/30 Copper-Nickel 70/30 Copper-Nickel Thickness 0.25 in. 0.25 in. Thermal insulation material Min-K 1301 Fuel capsule material Ha stelloy C Thermoelectric module T/E material Bi Te N/P 2 3 Hot junction temperature 400°F Cold junction temperature 75°F T/E output voltage 1.6 volts No. of couples in series 50 Radiation dose rate Less than 200 mr/hr surface and 10 mr/hr at one meter. Security All aspects of URIPS are unclassified Licensing Packaging conforms to ICC/AEC shipping regulations

References

1. Aerojet-General Nucleonics, Large Milliwatt Generator Program - Parametric Study Data, Report No. AGN-8182, Vols. I and II, March 1966, San Ramon, California.

2. Corrosion Handbook, H. H. Uhlig, John Wiley & Sons, Inc., N.Y., 1948.

3. Fred M. Reinhart, "Effect of Deep Ocean En­ vironments on the Corrosion of Selected Alloys", Tech. Note N-781, NCEL, October, 1965.

18-39 1.0watt - SrTi0 - External Full U-Shield - MIN-K-1301 - L/D 1.2 - TC 75 3

300

~ c.. I- ::i: CJ w ~

. • 200

5

: 1._.:_· 4

~0 :::J' 0 ~ 3 >-u z wu u: 2 u.. w

8 9 10 11 DIAMETER {in.)

Figure 1. Effect of Thermoelectric Material nn 1-Watt RTG Perfonnance

18-40 1- :r: (.') w ~

..o O'

>­u z lJ.J u u.. u.. w

6 7 8 9 10 DIAMETER (in.)

Figure 2. Effect of Thermoelectric Material on 150-Milliwatt RTG Performance

18-41 II" SHIELD INSULATION SHIELD """ INSULATION SHIELD INSULATION - SHIELD """ - 1-l T/E 111 R/I ,, IT/E: 00 ~ I [0 ~ G t....:> \...

~ ~

INTERNAL SHIELD EXTERNAL SHIELD SPLIT SHIELD

Figure 3. Typical RTG Shield Configurations 0 FULL EXTERNAL SHIELD THERMOELECTRIC MATERIAL - Bi Te 2 3 THERMAL INSULATION - MIN-K-1301 A FULL INTERNAL SHIELD FUEL CAPSULE L/D - l. 2 FULL SHIELD - 200mr/hr AT SURFACE COLD JUNCTION TEMP - 75°F MAX HOT JUNCTION TEMP - 4S0°F

::0- ~ t- I (.'.) w ~ 200

12 c. 11 -0:: w t- w 10 ~ ~ 0 9

8

>­u z . "". :: ::: ; :::: .: ... LU . ' ·: :::· 1: .. ' ...... ::..:.:.:.: :;:: u ·---· -· -·- ...... u.. u.. 0 -·-·...,...... , ...... _.__.__.__.~· ....__,_...._ ____·~·-···------···-:-!:...._;;-:;_;: UJ 1.0 0.8 0.6 0.4 0.2 0 EXTERNAL SHIELD (URANIUM OR LEAD) !~~~--·~~~-·~~~--·~~~-·~~~---~--~ 0 0.2 0.4 0.6 0.8 1.0 INTERNAL URANIUM SHIELD

SHIELD FACTOR

Figure4. Optimum Design Parameters for 1-Watt RTG as a Function of Shield Configuration

18-43 l .Owatt - SrTi0 - External Full U-Shield - MIN-K-1301 - Z O. 90 - TC 75 - Bi Te 3 2 3 r--i~~_,..~...,...... ,.-~~.,.....,..-r--__,._,....,..._,.....,....~-T-~.,...... ,_l"l'-'l__,,._....,... ______....;...,._.400

;$!. 0 3 ::;- 0 w ->-u .J- z w ~ - ~ u.. u.. ,_J w 2J--l I- I:

8 9 10 11 DIAMETER (in.)

Figure 5. Effect of Fuel Capsule L/D on 1-Watt RTG Performance Parameters

18-44 o. 18

'LL 0 l ~ o. 16 I ~ e::::> >- 0.14 !:::: ~ uI- :::> z0 o. 12 0 u ....J <( ~ 0::: w o. 10 I I-

0.08

0.06

0.04

0.02

0 -500 0 500 1000 1500 2000 2500 MEAN TEMPERATURE (°F)

Figure 6. Thermal Conductivity of Insulation Materials

18-45 . 1 WATT~) URIPS-U1, DEPLETED URANIUM SHIELD

HOUSING CAP

PRESSURE VESSEL HEAD

POWER CONDITIONING UNIT

ELECTRICAL LEAD ELECTRICAL FEED-THROUGH RECEPTACLE ...... ,~~--SHIELD PLUG

15.2 ~~-----...._--THERMOELECTRIC CONVERTER

~~~--t+--- RADIOISOTOPE HEAT SOURCE

INSULATION

BIOLOGICAL SHIELD

LIFTING LUG

- --- 9.70 DIA. ---

Figure 7. 1-Watt(e) URIPS-Ul, Depleted Uranium Shield

18-46 • 1 WATT(e) URIPS·Pl, LEAD SMIELD

HOUSING CAP

PRESSURE VESSEL HEAD

POWER CONDITIONING UNIT

LIFTING LUG

ELECTRICAL ELECTRICAL LEAD FEED-THROUGH RECEPTACLE

MAIN HOUSING

19.7 THERMOELECTRIC CONVERTER

RADIOISOTOPE HEAT SOURCE

PRESSURE VESSEL

BIOLOGICAL SHIELD

Figure 8. 1 Watt(e) UR.IFS-Pl, Lead Shield

18-47 18-48 - ----

M - Q) s:: •r-1 ~

~ l"'""f Q) :l i::i:. l"'""f - Q) Q) .j,J - Cl) --- C/J rn --,___ Q) - l"'""f s:: •r-1 - ca .j,J --- Cl) ~-- ~ ~ Cl) ~ Sn - •r-1 ---~..._--- ~ ---.__ t~ :::-_ta------

18-49 18-50 Figure 9D. Welded Strontium Titanate Fuel Capsule

18-51 18-52 T1

Nf2 CRl R3

Q1 CR ZL Nsl 3

R2 VB

I-' 00 I Cl c:...:>

~~I~+ - TE MODULE

Figure 11. Power Conditioning Circuitry URIPS TEST CONSOLE 1 WA TT(e) URIPS·Pl LEAD SMIELD .

Figure 12. Aerojet Undersea Radioisotope Power Supply (URIPS-Pl-1001)

18-54