Multi-Evaporator Hybrid Two-Phase Loop Cooling System for Small Satellites

Transcription

1 SSC7-X-11 Multi-Evaporator Hybrid Two-Phase oop Cooling System for Small Satellites David C. Bugby ATK Space 55 Powder Mill Road, Beltsville, MD 275 ABSTRACT This paper describes a small satellite thermal control architecture based on the multi-evaporator hybrid loop heat pipe (ME-HHP), a two-phase loop cooling system that combines capillary pumped loop (CP) and loop heat pipe (HP) functionalities. The cooling system incorporates multiple heat load sharing evaporators for cooling/heating remote (or nearby) components, dual counter-flow freeze-tolerant condensers for reduced attitude dependence, and miniaturized Teflon wick evaporators for minimum control power. With concomitant functionalities like heat load sharing and thermal diode action, this system can better meet the needs of future extended eclipse/limited power small satellite missions compared to the traditional "cold-biasing plus heater power" approach. Extensive ME-HHP ground testing has been performed to demonstrate the capabilities of the system for future small satellite missions. This paper will review the design/testing of this and five additional ground-based ME-HHP cooling systems. INTRODUCTION The traditional spacecraft thermal design approach of sizing radiators for the "hot case" and using heater power for the "cold case" will not suffice for future smallsat missions with large temperature extremes and stringent power/mass limitations. Thus, NASA initiated the ST-8 research effort several years ago to develop a new thermal management system (TMS) for small satellites. 1 One of the initial ST-8 studies, which is the subject of this paper, resulted in the first-ever ground test of a miniaturized multi-evaporator hybrid loop heat pipe (ME-HHP) cooling system. 2 In the three years since that study concluded, five additional ME-HHP based cooling systems have been successfully ground tested. This paper will describe the design/testing of each cooling system and outline an implementation of a centralized ME-HHP thermal bus for a small satellite. BACKGROUND The ST-8 program goal was a TMS that would enable component placement flexibility, minimize power/ mass/volume, improve power resource efficiency, and be scalable up/down from kg, 2 W. The features necessary to meet those objectives included a multievaporator bus, heat load sharing (HS), miniaturized components, thermal diode action, multiple condensers, set-point controllability, and high conductance. The TMS that could best provide those features was a multievaporator two-phase loop. The available two-phase loop architectures included the capillary pumped loop (CP), loop heat pipe (HP), and hybrid loop heat pipe (HHP). Although CPs/HPs had a TR head start, as several CPs and numerous HPs are now on-orbit, the HHP had features that made it the best choice. Two-Phase oop Architectures Due to its remote reservoir design, as shown in Figure 1, the HHP architecture is more like that of a CP than it is like that of an HP. Thus, the HHP has excellent temperature controllability/expandability. The HP, though, with its adjacent-to-evaporator reservoir design, sacrifices controllability/expandability for robustness. A secondary wick between the HP evaporator and reservoir allows it to manage high "back conduction" heat leaks in high pumping metal wicks. CPs must depend on inlet liquid subcooling to manage back conduction and thus are limited to low conductivity, low pumping polymer wicks. The HHP overcomes the CP wick limitation by using a sweepage evaporator to create an auxiliary flow that sweeps vapor/heat from the primary evaporator cores. The HHP thus provides CP expandability/controllability and HP robustness. IQUID HP HEAT INPUT 4-Port Evaporator Flow SET-POINT RESEROIR SHUNT CONDENSER SWEEPAGE SEC. EAP HEAT REJECTION CP APOR HHP Figure 1: CP, HP, and HHP Architectures Bugby 1 21 st Annual AIAA/USU

2 SSC7-X-11 Expanding to Multiple Evaporators SYSTEM DESIGN / TESTING HHPs and CPs can be expanded in parallel virtually without limit. Figure 2 illustrates the simple in parallel expansion of a single-evaporator HHP into a dualevaporator ME-HHP. HPs, however, cannot be expanded without limit due to the geometric growth of reservoir volume with the number of evaporators. Thus, ME-HPs are limited to 2-3 evaporators. The HHP can also be expanded (or plumbed) in series. Figure 3 illustrates the two methods for plumbing an ME-HHP. Series plumbing uses the liquid sweepage flow from an upstream evaporator as the liquid inlet to the adjacent downstream evaporator. The advantage of a seriesplumbed ME-HHP is that the minimum sweepage flow into any evaporator is the mass flow rate exiting the sweepage evaporator, although the subcooling nominally decreases downstream. The advantage of a parallel-plumbed ME-HHP is that the subcooling entering all evaporators is equal, although the sweepage mass flow rate may vary if the evaporator load varies. IQUID HEAT INPUT 4-Port Evaporator HEAT INPUT 4-Port Evaporator Flow SET-POINT SWEEPAGE APOR KEY FEATURES for negligible back conduction and low control power evaporator/reservoir enables core sweepage to manage back conduction biased heat exchanger (CBHX) for Back pressure regulator (BPR) for heat load sharing, vapor line clearing located flow regulator (CFR) for switching between multiple radiators flow condensers for freeze tolerance (if no spot heating) This portion of the paper outlines the design/testing of the ST-8 ME-HHP small satellite cooling system and five additional ground test ME-HHP cooling systems. ST-8 ME-HHP Small Satellite Cooling System To meet the ST-8 goals, a quad-evaporator, parallelplumbed ME-HHP was designed, fabricated, and tested. Figure 4 illustrates the architecture. Key loop features included: (a) sweepage evaporator for back conduction management; (b) miniaturized Teflon wick evaporators for low back conduction and control power; (c) dual counter-flow condensers for multiple sinks with freeze tolerance; (d) back-pressure regulator for heat load sharing and vapor line clearing; (e) co-located flow regulator for condenser switching; (f) cold-biased heat exchanger (CBHX) for fine temperature control; and (g) dual diode heat pipes for CBHX cold-biasing. Figure 5 illustrates the test loop, which utilized a lab chiller for condenser cooling and CBHX cold-biasing. Diode Heat Pipes (provide coldbiasing) apor Side Core iquid Side iquid ine Core apor ine EAP 1 (PTFE WICK ) EAP 2 (PTFE WICK ) Thermal Storage Unit (TSU) EAP 3 (PTFE WICK ) EAP 4 (PTFE WICK ) Sat. Temp. Heater Heater Multiple Radiators with Counter-Flow Condensers RESEROIR SHUNT CONDENSER SEC. EAP HEAT REJECTION Figure 2: Parallel Expansion: HHP to ME-HHP Parallel Plumbing E1 E2 PARAE R E3 SE C SERIES E4 Secondary Back Pressure Cold-Biased Co-ocated Flow Regulator Evaporator/ Evaporator Regulator (BPR) Heat Exchanger (CFR) and (CBHX) Figure 4: ST-8 ME-HHP Architecture Counter-Flow Condenser 2 EAP 1 EAP 2 Shunt EAP 3 EAP 4 BPR (Heat oad Sharing) Cold-Biased Sight Glasses Heat Exchanger (CBHX) Single-Phase Secondary apor Evaporator Two-Phase iquid Counter-Flow Chiller ines Condenser 1 Series Plumbing E1 E2 E3 E4 Figure 5: ST-8 ME-HHP Test oop R SE C Figure 3: ME-HHP Parallel vs. Series Plumbing With ammonia as the working fluid, a series of 21 tests was carried out with the ST-8 ME-HHP test loop resulting in: (1) quad-evaporator transport of 8-28 W; (2) single-evaporator transport of 2-1 W; (3) power Bugby 2 21 st Annual AIAA/USU

3 SSC7-X-11 cycling from 5-2 W; (4) maximum heat flux of 3 W/cm 2 ; (5) conductance of 5-8 W/K per evaporator; (6) heat load sharing greater than 95%; (7) successful condenser switching; (8) freeze-tolerant condenser (liquid exited at saturation); (9) temperature control of +/-. K with a variable set-point; (1) rapid 3 minute start-up; (11) low secondary evaporator control power of 4W; (12) loop isolation/diode action; and (13) Teflon evaporator K temperature cycling. Selected test results are provided in Figures 6-1, which respectively illustrate the power cycling, heat load sharing, reservoir set-point control, heat transport limit, and loop isolation test results : :3 16: 16:3 17: 17:3 18: 18:3 inlet (1) (4) 2nd pump (6) BPR outlet (14) FR outlet (43) iquid line (45) E2 body (52) E3 body (57) vapor sweepage line (73) ambient (8) 2nd pump [W] E2 Power [W] E3 power [W] Total Evap Power [W] Figure 9: ST-8 Heat Transport imit Heat oad [W] Heat oad [W] Heat oad [W] :3 19: 19:3 2: 2:3 21: 21:3 22: inlet (1) (4) 2nd pump (6) BPR outlet (14) FR outlet (43) iquid line (45) E2 body (52) E3 body (57) liquid sweepage line vapor sweepage line (73) ambient (8) 2nd pump [W] (68) E2 Power [W] E3 power [W] Total Evap Power [W] Figure 6: ST-8 Power Cycling : 11: 12: 13: 14: : 16: 17: 18: 19: 2: (1) (4) 2nd pump BPR (14) inlet (6) outlet FR outlet (43) iquid line (45) E1 body (48) E2 body (53) E3 body (57) E4 body (62) liquid sweepage line (68) vapor sweepage line (73) ambient (8) Q-meter (82) Q-meter (83) 2nd pump [W] E1 Power [W] E2 Power [W] E3 power [W] Total Evap Power [W] Figure 7: ST-8 Heat oad Sharing Primary Evaporators 4 3 FR outlet 35 Heat oad [W] -4 7: 8: 9: 1: 11: 12: 13: 14: : 16: 17: inlet (1) (4) 2nd pump (6) BPR outlet (14) iquid line (45) E1 body (47) E2 body (52) E3 body (58) E4 body (62) ambient (8) Cond #2 () Total Evap Power [W] Figure 1: ST-8 oop Isolation ST-8 Status. One of the major accomplishments during this 6-month ST-8 study was the development of a miniaturized Teflon wick 4-port evaporator. Figure 11 illustrates this novel design, which features a.64 cm outer diameter wick. Moreover, all 21 tests conducted were highly successful. Despite this success, NASA did not select the ME-HHP small satellite cooling system for the ST-8 flight experiment. A dual-evaporator ME- HP with TEC reservoir cold biasing was selected instead. 4 However, that selection elicits concerns such as: (a) the risk of expanding beyond two evaporators; and (b) the impact of TEC failure on loop temperature controllability. Although these two issues are not concerns for the ME-HHP, flight verification still is. Thus, a flight test to verify the ME-HHP architecture in zero-g, preferably with 4-6 evaporators, is needed before future flight implementations are likely. 3 2 Heat oad [W] : 6:3 7: 7:3 8: 8:3 inlet (1) (4) 2nd pump (6) BPR outlet (14) FR outlet (43) iquid line (45) E2 body (52) E3 body (57) liquid sweepage line (68) vapor sweepage line (73) ambient (8) 2nd pump [W] E2 Power [W] E3 power [W] Total Evap Power [W] Figure 8: ST-8 Set-Point Control Figure 11: Miniaturized Teflon Wick Evaporator Bugby 3 21 st Annual AIAA/USU

4 SSC7-X-11 Additional ME-HHP Cooling Systems Five additional ME-HHP cooling systems have been successfully ground-tested since the conclusion of the ST-8 study three years ago. These systems include the following applications: (1) moderate flux laser; (2) high flux laser; (3) large spacecraft; (4) rack electronics; and (5) intermittent power instrument. The five applications are discussed below. A summary table with key data on each test system is provided later in the paper. Moderate Flux aser. The problem addressed in this program was the cooling of a moderate flux (3 W/cm 2 ) laser crystal with a sub-ambient operating temperature. The solution was an ammonia ME-HHP with four evaporators mounted in an Al heat sink and a liquid cooled shield (CS). The CS is a self-cooling twophase loop plumbing feature originally developed to enable the cryogenic CP. 3 Figure 12 illustrates the architecture and Figure 13 illustrates the test loop. In lab testing with a test heater, the cooling system met the requirements for heat flux, operating temperature, and heat sink uniformity. In testing at the customer site, the cooling system was successfully integrated with a working laser. Crystal waste heat was quantified by turning the diode array off and applying power to a Kapton heat sink heater and then matching the liquid inlet temperature, creating an in-situ Q-meter. High Flux aser. The problem addressed during this program was the cooling of multiple low profile heat sources in a high power/high flux laser system. The solution was a dual-evaporator ammonia ME-HHP with: (a) innovative inlet/outlet ports to enable the cooling of multiple low profile heat sources; (b) series plumbing; and (c) a mechanical pump (no sweepage evaporator). Figure 14 illustrates the architecture and Figure illustrates the test loop. In lab testing with a test heater, the system successfully operated in the "flow-through" mode with a total heat load of 88 W on the evaporators heated from one side. The system was designed for two-sided heating of the evaporators (.64 cm wick OD) and a total heat load of 16 W. Other key results are as follows: (1) system operated below ambient due to high mass flow rate (sub-ambient operation is possible even at W due to the mechanical pump); (2) target heat flux of 5 W/cm 2 was achieved with single-sided heat input (system was designed for two-sided heat input); and (3) successful operation with -45 W and -224 W power cycling. 1. Mechanical Pump 2. Filter 3. Calorimeter 4. Evaporator 1 5. Evaporator 2 6. Evap 1-2 iquid ine 7. apor ine 8. Condenser alve 11. Chiller/Shunt 12. Chiller Path Chiller Path Chiller Path 1. Chiller Path DP Transducer 17. Fill Tube Flow Regulator (4) iquid Cooled Shield (CS) 1 1 iquid Header 13 Four Parallel Condenser ines Figure 14: High Flux aser Cooling Architecture apor ine Evaporator Mounting Plate (Four parallel four-port evaporators) Condenser apor Header Back Pressure Regulator (BPR) Secondary Evaporator Figure 12: aser Crystal Cooling Architecture CS iquid Header Condenser 4 Primary Evaporators in Al Heat Sink Alum. Shunt Flow Reg. apor Header Secondary Evaporator Figure 13: aser Crystal Cooling Test oop Figure : High Flux aser Cooling Test oop arge Spacecraft. The cooling of very high power (up to 1 kw) next-generation military spacecraft is the problem addressed by the Dual-Use Science and Technology (DUS&T) program, conducted jointly with AFR/PRP. The solution involves a mechanical pump assisted ammonia ME-HHP with the ability to smoothly transition from capillary to mechanical pumping. The program was conducted in two phases. Bugby 4 21 st Annual AIAA/USU

5 SSC7-X-11 During the first phase, a 3-evaporator 2 kw riskreduction test bed was designed, fabricated, and tested. The results from the 2 kw system provided guidance to design a 1 kw test bed for the second phase, which is configured as follows: (1) 6 low impedance evaporators; (2) series plumbing with parallel as an option; (3) 4-port evaporators with 3-port evaporators as a valving-enabled option; (4) mechanical pumping with capillary-only pumping as an option; and (5) large spacecraft features that include evaporator-condenser separation of 5 m and evaporator elevation differences of 3 m. Figures 16 and 17, respectively, show a block diagram and a photo of the 1 kw test bed. Results for the 1 kw system will be published shortly. Mechanical Pump Flowmeter 3 2 Heater Blocks Filter 2 Coolant Block ine apor ine 93 apor ine 93 iquid ine 93 Rack Electronics. In conjunction with TA&T, this Navy SBIR program addressed the problem of rack electronics cooling on Navy ships/submarines. In particular, since air-cooling is nearing its limits, and is an acoustic noise source for submarines, an alternative was sought. The solution was two-pronged: (1) boxmounted water ME-HHP cooling loops; and (2) rackmounted single-phase water-cooling loop with "wedge" interfaces. Figure 18 illustrates the solution. Figure 19 illustrates the ME-HHP test loop. Figure 2 illustrates the wedge interface system. The results from this effort are as follows: (1) 1 W heat load on each evaporator (3 W total) with 5 W on the secondary evaporator; (2) loop saturation temperature of 328 K; and (3) endto-end conductance was 9 W/K (35 W, 348 K heater, 313K chiller). Future work will involve installation and testing of (derivative) single-evaporator HPs with ceramic wicks and air-cooled condensers. This alternate approach will hasten system implementation given infrastructure resistance to implementing rack cooling. 132 ine 93 iquid ine bonded saddle next to pump body) 6 E1SN1 E4SN E2 SN6 E3 SN12 E5SN8 E6SN Sight Glass (2) Pressure Transducer (4) Condenser Plate (3) Figure 16: Schematic of 1 kw Test Bed Figure 18: Rack Electronics Cooling Solution Flat Plate 3 Evaporators 2m Above Condenser Figure 19: Water ME-HHP Condenser WEDGE Elevation Control Hinges/Flex ines WEDGE MATES TO RECEIER Figure 17: Photo of the 1 kw Test Bed Bugby SERER IN RACK WEDGE CONDENSER 3 Evaporators 1m Below Condenser WEDGE RECEIER SCREW TIGHTENED TO FINAIZE CONNECTION Figure 2: Wedge Condenser System 5 21st Annual AIAA/USU

6 SSC7-X-11 Intermittent Power Instrument. This system involved the cooling of an instrument with multiple distributed heat sources, a -4 W variable load, and nearambient operation. The solution was a cascaded dualloop (instrument-side ME-HHP, radiator-side HHP) system with many novel but necessary "thermal toolbox" elements including the following: (1) liquid cooled shield (CS) for sub-ambient HHP operation; (2) thermoelectric cooler (TEC) for HHP reservoir cold-biasing; (3) ME-HHP reservoir differential thermal expansion thermal switch (DTE-TSW) shunt for minimum control power; (4) ME-HHP thermal storage unit (TSU) condenser with a sub-ambient phase change material (PCM); and (5) multiple HHP condensers for attitude independence. Figure 21 illustrates a system schematic. Figure 22 illustrates an annotated Pro/E model of the system. Both loops used ammonia as the working fluid. The cooling system was able to very precisely control the temperature of a distributed, intermittent power instrument. INSTRUMENT EAP EAP EAP EAP AP HDR SWP HDR Primary Evaporator Instrument-Side TSU/Condenser IQ HDR APOR INE 2ND EAP HTR1 SWEEPAGE INES HTR3 TSW RSR HTR2 SWEEPAGE INES INSTR-RAD INTERFACE CONDENSER TSU HEAT PIPES SODERED (OC. FOR HP) EAP HTR4 STRAP TEC RSR HTR6 INSTRUMENT-SIDE IQUID INE IQUID INE STRAP 2ND EAP HTR5 CS RADIATOR-SIDE APOR INE CS INE "OUT" CS INE "IN" Figure 21: Dual-oop Cooling System, Secondary Evaporator, TEC, and Strap 6" iquid-cooled Shield (CS) 6" Flow Regulators (2X) (to maximize condenser utilization) for Condenser #1 Condenser #1 (lines on other side) Instrument-Side Primary Evaporator Plate (4X) for Condenser #2 Condenser #2 (lines on other side) CONDENSER 2 CONDENSER 1 RADIATOR 2 RADIATOR 1 Instrument-Side alves to Enable 3 vs. 4 Port ME-HHP Plumbing and Enhanced Transient Operation (5X Regular alves, 2X Check alves [C], Test Only) Instrument-Side Thermal Switch (TSW) Instrument-Side Instrument-Side Secondary Evaporator Instrument-Side Back Pressure Regulator (BPR) Instrument-Side alves to Enable Parallel or Series ME-HHP Flow Configurations (12 X, Test Only) C Instrument-Side Primary Evaporator (4X) C 2" 2" Instrument-Side Heat Pipes (to spread heat on TSU bottom) Patent US 6,889,754 Figure 22: Pro/E Model of Dual-oop Cooling System Summary of ME-HHP Cooling Systems. A summary table of the characteristics of the ST-8 ME-HHP cooling system and the five additional ME-HHP cooling systems is provided in Table 1. The table lists the number of evaporators, wick OD, saddle width, evaporator length, wick material, maximum loop heat load, maximum evaporator heat load, maximum heat flux, transport length, adverse elevation, evaporator body material, working fluid, and advanced features. Table 1: Summary of ME-HHP Features SYSTEM IMPICATIONS TMS Features/Rationale/Benefits The TMS features necessary to meet ST-8 goals were previously identified as: (1) multi-evaporator bus; (2) heat load sharing; (3) miniaturized components; (4) thermal diode action; (5) multiple condensers; (6) setpoint controllability; and (7) high conductance. isted below are reasons why the aforementioned features are needed and the expected benefits for small satellites. Multi-evaporator buses can decouple the structural/ thermal design process, so that component placement is more flexible, simplifying design and saving mass. Heat load sharing is needed to keep environmentally exposed instruments warm when they are not turned on, reducing the need for smallsat heater power. Miniaturized components are necessary to reduce weight and expand smallsat packaging options. Thermal diode action is necessary enable smallsat payloads to be isolated from extreme environments, which expands the missions that can be undertaken. Multiple condensers (radiators) reduce a satellite's need to adjust attitude for thermal control, resulting in increased time for science. Set-point controllability reduces payload temperature variations, which minimizes temperature cycling and lengthens payload life. High conductance enables centralized component configurations, which minimizes electrical harness lengths, simplifies the structure, and saves mass. TMS Implementation Given the clear benefits listed above, how to implement an ME-HHP into a small satellite needs a brief discussion. Consider first a traditional small satellite configuration in which high power components are placed on body-mounted radiators, space/earth viewing instruments are mounted externally, and additional components are mounted internally. Radiators are sized/coated to handle the hot case environment and heater power is used to keep components from getting Bugby 6 21 st Annual AIAA/USU

7 SSC7-X-11 too cold in the cold case. Figure 23 illustrates a layout using the traditional thermal design approach. Figure 24 illustrates a centralized ME-HHP using the Figure 23 architecture/components. In Figure 24, all components except external viewing ones are coupled to a central ME-HHP bus, earth/space viewing components are kept warm when OFF by HS, and hot radiator soakback is small due to ME-HHP diode action. ACKNOWEDGMENTS The author would like to gratefully acknowledge the important contributions of the following individuals to the work described herein: Matt Beres, Pete Cologer, Jessica Kester, Dmitry Khrustalev, Steve Krein, Ed Kroliczek, Chuck Stouffer, Dave Wolf, Kim Wrenn, and James Yun. Radiator 1 Sized/Coated for Hot Case Including Hot Radiator 2 MI Covers Non-Radiator Surfaces Space-iewing Components Need Heater Power When OFF Internal Components Coupled Conductively and/or Radiatively to the Walls High Power Components Mounted On/Near Exterior (may need heater power when OFF) Earth-iewing Components Need Heater Power When OFF Radiator 2 Sized/Coated for Hot Case Including Hot Radiator 1 Figure 23. Traditional Small Satellite Design ayout Radiator 1 Sized/Coated for Hot Case... Can Ignore Impact of Hot Radiator 2 Total MI Coverage Except Protruding Components, ines, Standoffs Space-iewing Components Kept Warm When OFF by HS All Components Except External iewing Ones Centrally ocated Coupled to ME-HHP Thermal Bus CBHX Radiator 2 Sized/Coated for Hot Case... Can Ignore Impact of Hot Radiator 1 REFERENCES 1. NASA New Millennium Program Space Technology 8, NRA 3-OSS-2, February Bugby, D., E. Kroliczek, and J. Yun, "Development and Testing of a Miniaturized Multi-Evaporator Hybrid oop Heat Pipe," Space Technology Applications International Forum (STAIF-5), Albuquerque, NM, January,. 3. Baumann, J., B. Cullimore, D. Bugby, and E. Kroliczek, "Development of the Cryogenic Capillary Pumped oop," 33rd IECEC, IECEC-98- I197, Colorado Springs, CO, August Ku, J.,. Ottenstein, D. Douglas, M. Pauken, G. Birur, "Miniature oop Heat Pipe with Multiple Evaporators for Thermal Control of Small Spacecraft," Paper No. 183, 3th GOMACTech Conference, as egas, N, April. Earth-iewing Components Kept Warm When OFF by HS 1 DHP to CBHX DHP to CBHX 2 Figure 24. Centralized ME-HHP Satellite Design CONCUSIONS This paper has described the design, fabrication, and testing of a multi-evaporator thermal bus architecture for small satellite thermal control. The basis for the system is the multi-evaporator hybrid loop heat pipe (ME-HHP), a two-phase loop cooling system with CP/HP underpinnings, but with key advantages over each. This system was designed/built/tested as part of the NASA ST-8 Phase A study from Jan-Jun 24. At that time, it was the first-ever ground test of a miniaturized ME-HHP cooling system. The design and testing of five subsequent ME-HHP based cooling systems -- in the areas of laser, spacecraft, electronics, and instrument cooling -- were also described. Although the architecture has been clearly proven for ground applications, to fully validate it for future smallsat missions, an ME-HHP flight experiment is needed. Bugby 7 21 st Annual AIAA/USU