Heat Pipe Architecture Offers Flexible Thermal Control
A hybrid loop heat pipe cooling system marries the controllability/expandability of capillary pumped loop schemes with the operational robustness of a loop heat pipe.
It has been 40 years since George Grover, a Los Alamos engineer working on space nuclear-power-system cooling, verified his idea of a surface tension-driven heat transport system or “heat pipe.” Since then, heat pipes—sealed tubes that enclose an internal porous structure saturated with liquid—have become commonplace. Copper-water heat pipes are used extensively in notebook computers, and aluminum-ammonia heat pipes are used extensively on spacecraft. Some advantages that heat pipes have over heat conduction devices (bars, straps, plates and so on) include high conductance, low weight and potential compliance.
Although heat pipes have no moving parts and are therefore highly reliable, they have key limitations. Because the porous structure or “wick” extends along the entire length of the pipe, lengthening the heat pipe—for applications with greater source/sink separation—reduces its heat transport capability. In addition, reducing the wick pore size (to improve its against-gravity pumping capability) has the same effect. So for applications with high heat loads and long transport lengths, the wick pore size needs to be large, but that also creates system orientation (levelness) constraints when heat pipes are tested/utilized on the ground.
The two-phase loop, developed about 15 years after the heat pipe, overcomes the key heat transport limitations of the heat pipe. In a two-phase loop, the wick is isolated to the portion of the system where heat is added (the “evaporator”) and the rest of the loop is smooth-walled tubing. With this design, increased transport lengths can be handled by slight increases in the tubing diameter. In addition, and perhaps more importantly, a smaller pore-size wick actually increases the heat transport capability of the two-phase loop. Figure 1 illustrates diagrams of a heat pipe and a two-phase loop and provides the mathematical relationships that show how heat transport capability (Q) varies with effective wick pore radius (rP).
Two-Phase Loop Architectures
There are three available two-phase loop architectures: loop heat pipe (LHP), capillary pumped loop (CPL) and hybrid loop heat pipe (HLHP). Each transports heat by evaporating liquid at the heat source and condensing vapor at the heat sink. When heat is applied to the evaporator, capillary pressure developed across the primary wick drives the flow. Most two-phase loop evaporators contain annular wicks with liquid entering on the inside of the annulus and heat input/evaporation occurring on the outside of the annulus.
All two-phase loops have an actively or passively “cold-biased” fluid reservoir that controls the loop operating temperature and handles fluid level fluctuations caused by varying sink conditions/operating temperatures. Figure 2 illustrates the features of a generic two-phase loop with an annular evaporator. For robust operation, evaporator core vapor production caused by heat leak across the annular wick must be actively managed or mitigated.
NASA, DoD and commercial firms have used single evaporator LHPs and CPLs to solve various ground/space thermal problems, but no multi-evaporator LHP or CPL systems are currently in use. Single and multi-evaporator HLHPs have been thoroughly laboratory tested and several key programs are planning to utilize the
ME-HLHP architecture to meet demanding thermal requirements. Figure 3 illustrates flow diagrams of the single evaporator LHP, CPL and HLHP architectures. Brief descriptions of each are provided below.
LHP Architecture: In an LHP, the reservoir is adjacent to the evaporator. A secondary wick between the evaporator and the reservoir keeps the primary wick wetted at all times. Thus, the LHP starts up automatically, vents core vapor to the reservoir where it is condensed, and auto-regulates the reservoir temperature according to an energy balance involving core vapor condensation heating, return liquid cooling and heating/cooling by the surrounding environment. LHPs can utilize small pore-size metallic wicks (1-1.5 micron Ni or Ti), which provide high against-gravity pumping capability. The LHP has temperature controllability limitations because the reservoir temperature depends on the aforementioned energy balance, not a thermostatic heater. When expanding to multiple evaporators, LHP reservoir size grows exponentially, so ME-LHPs cannot practically exceed more than three evaporators.
CPL Architecture: In a CPL, the reservoir is remote from the evaporator. Thus, the CPL operating temperature is highly controllable because the reservoir can be mounted on/near the heat sink and its temperature can be controlled with a thermostatic heater. Because of the cold-biased remote reservoir, CPL evaporators may not always be liquid-filled prior to start-up. For reliable start-up, the CPL requires loop preconditioning that involves reservoir heating to flood the evaporator. CPLs do not have an active means of managing core vapor production, thus they must mitigate it by using low thermal conductivity polymeric wicks, whose pore sizes are relatively large, and by ensuring that the temperature of the liquid entering the evaporator stays several degrees below the loop operating (saturation) temperature. When expanding to multiple evaporators, CPL reservoir size grows linearly, so ME-CPLs can theoretically be expanded without limit.
HLHP Architecture: The HLHP utilizes a CPL-style remote reservoir for a high degree of temperature controllability. For active core vapor management, a secondary “sweepage” evaporator is located adjacent to the reservoir. This sweepage evaporator/reservoir combination is actually an LHP evaporator. Application of heater power to the sweepage evaporator adds excess mass flow into the loop, which acts to sweep vapor from the primary evaporator core—or cores in the case of an ME-HLHP— thus the mechanism is sometimes referred to as “core sweepage.” With this feature set, the HLHP combines LHP operational robustness with CPL temperature controllability/expandability.
Multi-Evaporators, Multi-Condensers
For applications requiring multiple evaporators, two-phase loops require a backpressure regulator (BPR), which is a tubular, wick-bearing device mounted at the vapor inlet to the cold sink. In operation, liquid can flow with little impedance through the saturated BPR wick. Vapor can flow through the BPR only when the vapor backpressure exceeds the BPR wick capillary limit—less than the evaporator wick capillary limit. With the foregoing functionality, the BPR ensures that any two-phase loop can achieve maximum heat load sharing (HLS). HLS is a process that occurs when cold evaporators function as condensers. HLS is useful on power-limited spacecraft to minimize survival heater power for duty-cycled, space-viewing instruments.
For applications requiring multiple condensers, two-phase loops also require a flow regulator, which is a tubular, wick-bearing device mounted at the liquid return side of the cold sink. In a two-phase loop with multiple condensers, where one or more may be hot but at least one is cold, the flow regulator prevents vapor from the hot condensers from penetrating into the liquid return leaving the cold condenser. For a spacecraft with opposed radiators—one on the sun side and the other in the shade—the flow regulator enables autonomous switching between radiators without active intervention.
NASA’s Demonstration Project
In February 2003, the NASA/Jet Propulsion Laboratory (JPL) New Millennium Program (NMP) issued a NASA Research Announcement (NRA) for the Space Technology 8 (ST-8) Flight Demonstration Project. One of the overarching research objectives was to develop a new thermal management system (TMS) for future small (150 kg, 200W) spacecraft. The project was to culminate in a 2007 flight experiment to validate the technology. NASA’s stated goals were to provide the following TMS capabilities: multi-evaporator for flexibility in component placement; heat load sharing for low survival heater power; miniaturized components for low weight/volume; heat flow controllability/set-point control; and modeling tools for scaling designs up/down in size.
Swales was one of ten awardees that received Phase A funding on ST-8. Swales’ concept was a four-evaporator, dual-condenser ME-HLHP called the Small Spacecraft Integrated Thermal Management System (SSITMS). The full SSITMS concept is illustrated in Figure 4a. Because SSITMS is a dual-condenser TMS, passive cold-biasing of the reservoir was accomplished using a pair of diode heat pipes connected to the radiators. During the six-month Phase A program, a breadboard of the SSITMS concept was designed, fabricated and extensively tested. Figure 4b illustrates the SSITMS four-evaporator, dual-condenser breadboard.
The SSITMS ME-HLHP met all of NASA’s stated goals and included the development of miniaturized Teflon primary wick evaporators to minimize heat leak across the primary wick, which also minimizes the sweepage power required for core vapor management. Despite the program success, it was not selected by NASA to continue to the flight demonstration phase. The project selected was a two-evaporator, two-condenser LHP with variable emittance radiators proposed by NASA Goddard Space Flight Center.
More Applications Using ME-HLHP
Five additional ME-HLHP applications that Swales Aerospace is currently working on are described below.
High-Power Spacecraft Cooling: As part of an effort jointly funded by Swales and the Air Force Research Laboratory (AFRL) at Wright-Patterson Air Force Base (WPAFB), initiated through the Dual Use Science & Technology (DUS&T) Program, a multi-evaporator, two-phase loop architecture for cooling future high-power spacecraft is under development. Heat loads of up to 10 kW will be tested in the lab. The multi-evaporator architecture is still being finalized, but one feature that differs from SSITMS is the potential use of a liquid pump to assist core sweepage.
Laser Amplifier Cooling: In conjunction with Coherent Technologies, Inc. (CTI), Swales is developing an ME-HLHP-based two-phase loop cooling system for a diode-pumped laser system. The goal is to provide cooling for a laser diode array and NdYag crystal with heat fluxes of 30-50 W/cm2. A quad-evaporator ME-HLHP architecture similar to that shown in Figure 5 is being utilized.
High-Power Laser Cooling: In conjunction with CTI, an ME-HLHP-based two-phase loop cooling system for a high-power laser waveguide system is under development. This ME-HLHP cooling system will likely use mechanical pump assist and flat evaporators to provide cooling for heat fluxes of 50-100 W/cm2.
Rack Electronics Cooling: In conjunction with Technology Assessment & Transfer (TA&T) and Aavid Thermalloy, an ME-HLHP-based two-phase cooling system, using water as the working fluid, is being developed—with Naval Sea Systems Command funding—to remove heat from electronics onboard U.S. Navy submarines/ships. One of the novel features of this project is the development of a ceramic flat plate evaporator for CTE matching to low-CTE heat sources. Figure 5 illustrates the concept.
Centralized Thermal Bus: Swales is pursuing the development and implementation of a centralized thermal bus (CTB) for spacecraft. This concept focuses on improving the payload-carrying capabilities of spacecraft by minimizing the weight inefficiencies of current thermal management practices. Instead of mounting hot/high power components on/near radiators, which is the current practice, components are instead mounted in a close-packed centralized arrangement, well in-board of the radiators.
A multi-evaporator heat transport system—such as the ME-HLHP—is used to maintain thermal control of the close-packed components. Remote instruments requiring a space-view are kept warm when not operating by the heat load sharing (HLS) capabilities of the CTB. This approach saves weight for three reasons. First, radiators can be made much lighter as they only have to support their own weight. Second, a more lightweight spacecraft structure can be used due to the more centralized location of mass. And third, electrical harness length and mass are significantly reduced.
An IR&D program utilizing ST-8 ME-HLHP technology will be used to demonstrate the feasibility of the concept. A centralized thermal bus-endowed “Futuresat” will be capable of heat load sharing for low survival heater power, diode action/maximum multi-layer insulation for isolation from hot/cold environments, thermal storage to load-level transients, close-packed electronics for minimal harnessing mass, deployable/steerable/retractable radiators for variable view/area heat rejection and others.
Swales Aerospace
Beltsville, MD.
(301) 595-5500.