Jeanne Dion 5/06/2005
ME 388R.2
Stirling Radioisotope Generators:
An Efficient Alternative to Power Future Space Missions
By Jeanne Dion
The University of Texas at Austin
ME 388R Nuclear Power Engineering
May 6th, 2005
Abstract: The Stirling Radioisotope Generator (SRG) is being developed as a high efficient source of electrical power for future NASA missions. Using the heat from the radioactive decay of Plutonium Dioxide fuel, the SRG employs a dynamic Stirling Cycle to generate 110 W of electrical power at impressive system thermal efficiencies exceeding 20%. SRGs provide substantial improvement of electricity production and inherently better fuel utilization than its predecessor, Radioisotope Thermoelectric Generators (RTGs). Each Free-Piston Stirling Convertor, designed by the Stirling Technology Company (STC), generates no friction between moving parts and are maintenance free. This paper presents a comprehensive view of the SRG system including its history, motivation for development, and comparison to the existing RTGs. Major system components and subsystems are discussed along with corresponding reliability considerations.
Table of Contents
Introduction
Background
History
Development
Comparison to Current Technology
Stirling Cycle
Stirling Thermodynamics
Free-Piston Stirling Convertors
SRG Convertors Components
Flexure Bearings and Clearance Seals
Heater Head
Linear Alternator
Advanced Controllers
SRG System
Step 2 GPHS Modules
Advanced Vibration Reduction System
SRG Reliability
Uncertainties
Conclusion
Nomenclature
Figures
References
Introduction
Stirling Radioisotope Generators are potential highly efficient power system that could replace Radioisotope Thermoelectric Generators in future NASA missions. The Since the 1990’s, the Department of Energy (DOE), in support of Project Prometheus, has funded the development and evaluation of Stirling Radioisotope Generators. Research and development is being conducted under contract with the DOE at the NASA Glenn Research Center (GRC), Lockheed Martin (LM), and the Stirling Technology Company (STC). The SRG is a modular design intended for use on a variety of NASA missions, particularly for the exploration of deep space and neighboring planetary terrain. Initial plans project the first mission to use an SRG to launch around 2010 on an unmanned mission to Mars (DOE 2005).
Radioisotope power systems are needed for missions where solar power is impossible. Since 1961, over 40 Radioisotope Thermoelectric Generators (RTG) have been used to provide all or partial electrical power for spacecraft [DOE 2002]. The current thermoelectric power systems offer high reliability due to the absence of moving parts. Although current technology using thermocouples to convert heat from radioactive decay into electricity are proven and reliable, they demonstrate low thermal efficiencies of around 7%. With system thermal efficiencies exceeding 20%, the Stirling Radioisotope Generator (SRG) is being considered as a highly efficient alternative to power future NASA space missions. Perhaps the most important feature of the SRG is the use of flexure bearings and clearance seals that allow for highly efficient and friction-free operation. The SRG’s ability to perform over 3 times more efficiently over long missions makes its the potential replacement of RTGs very likely in the future. The SRG system design utilizes similar modular Step 2 General Purpose Hear Source (GPHS) units as the current traditional RTGs. Free-Piston Stirling Convertors use the heat source to drive a dynamic Stirling cycle. SRGs provide substantial improvement of electricity production and inherently better fuel utilization than RTGs. Although it is being developed as modular heat source capable of powering a variety of missions, the SRG’s first mission is projected to be on a Mars rover [Thieme 2002]. If successful, the SRG will be the first dynamic power conversion system used for space applications and will represent a huge technical achievement.
This paper presents a review of the current SRG power system design beginning with the background of SRG power systems highlighting the motivation of its development. A brief description of current radioisotope space generators will be given to establish a comparative datum to the SRG system. For a comprehensive understanding of the SRG system, an explanation of the Stirling thermodynamic cycle and its implementation as a Free-Piston Stirling Convertor is provided. Next, a summary of the STC’s Stirling convertor and its components is presented including descriptions of the flexure bearings, clearance seals, heater head, linear alternator, Step 2 GPHS modules, and advanced controllers. A reliability assessment will emphasis and discuss potential functional and structural failures of the SRG. Lastly the future of the SRG will be presented within concluding remarks.
Background
History. The current work to adapt a Free-Piston Stirling Convertor for spacecraft power systems is founded on previous work done at the NASA Glenn Research Center (GRC) to develop a Stirling automobile engine in the mid 1970s [Schreiber 2002]. The Automobile Stirling Engine Project, funded by the DOE, laid the foundation for the development of the SRG. The established history of Stirling power convertors has utilized GRC’s expertise in the following technical areas: materials, structures, structural dynamics, magnets, cycle analysis, and computational dynamics [Schreiber 2002].
Each SRG system consists of two GPHS modules and two Free-Piston Stirling Convertors. A cross sectional view of half an SRG system is shown in Figure 1. At the beginning of mission (BOM), the GPHS modules each provide 250 W of thermal power from the radioactive decay of plutonium-238 fuel pellets.
Development. Lockheed Martin (LM) is the SRG’s System Integration Contractor. Plans are underway to complete the 110 We Stirling Radioisotope Generator (SRG110) engineering unit in 2006 [Thieme, 2004]. The DOE contracted the Stirling Technology Company of Kennewick, WA to provide the Stirling convertors for the SRG system. STC developed Technology Demonstration Convertors (TDC), which are designed to provide 55 We each when operated at 650 C and 120 C between the hot and cold ends, respectively [Schreiber 2002]. Each SRG system will include two Stirling convertors which are based from the TDC design; see Figure 2. The TDC demonstrates the long life and supports the SRG concept design. STC is currently working with GRC and LM to adapt the TDCs into a flight convertor (FP) [Thieme, 2004]. The new FP redesign will allow for hermetic sealing and reduce weight [Qui 2002].
Table X. Comparison of TDC and FP Convertors
A major technical challenge exists in minimizing the unwanted system vibrations. With a TDC dynamic system, this can be challenging since the physical dynamics are dependent of complex thermodynamic analysis.
In actual power systems, multiple SRG units may be included to increase modularity and redundancy. Previous vibration testing only addressed system dynamics of a single SRG unit and has not discussed multiple SRG configurations.
Comparison to Current Power Systems. Static power systems, like the Radioisotope Thermoelectric Generator (RTG), have been used successfully for over three decades to provide electricity on spacecraft such as Pioneer, Viking, Voyager, Galileo, and Ulysses [DOE 2004a]. Eighteen General Purpose Heat Source (GPHS) modules, as shown in Figure 3, each contain 600 g of Plutonium-238 Dioxide (PuO2) fuel and provide 4500 W of net thermal power at the beginning of mission (BOM). Silicon germanium thermocouples directly convert heat from the radioactive decay of PuO2 into electricity [Space-1 2002]. Because there are no moving parts, RTG systems have an inherently long life and high reliability. The RTG, with its impeccable safety record, has proven to be a reliable and maintenance-free source of electric power for long space missions where solar power systems are not practical. Without radioisotope power systems, deep space exploration and planetary surface missions would not be feasible.
The solid-state thermoelectric direct conversion in RTGs exhibit low thermal efficiencies of about 7% [DOE 1996]. Since the price of the PuO2 fuel is the dominate cost factor in the RTG, reducing the fuel inventory is desirable. This decrease in radioisotope inventory can be achieved with higher thermal efficiencies [Schreiber 2002]. With a system efficiency exceeding 20%, the SRG is an attractive alternative to RTG systems. The SRG system is a modular design that uses new and improved GPHS modules, similar to the ones used in past RTGs, and thermodynamically independent Free-Piston Stirling Convertors. The SRG engineering unit is estimated to be comparable in size and weight to the RTG. The most impressive advantage SRGs exhibit is their ability to convert thermal energy dynamically into electricity with frictionless, non-contacting Free-Piston configuration; see Figure.4 The use of flexure bearings and clearance seals allow the Stirling convertor to seal the working fluid space without the use of lubrication.
Table XX. Comparison of RTG to SRG
Radioisotope Generatro / Mission Life(yrs) / Thermal Efficiency / #GPHS modules / Power Output (We)
RTG / 5 / 7 / 18 / 315
SRG / 14 / 25 / 2 / 110
Stirling Cycle
Stirling Thermodynamics. In 1816, a Scottish minister named Robert Stirling patented an “air engine” [American 2002]. This “air engine” only needed a temperature differential to output work efficiently. Using the expansion and compression of gas, the “air engine” operates on a Stirling Cycle with an external heat source and heat sink. With different gases as the working fluid, Reverend Stirling’s “air engine” became known as the Stirling Engine. Stirling Cycles are a closed system that uses external combustion. In 1873, it was discovered that, with a refrigerant and exchanger, the process may be reversed to make a refrigerant process and take work as an input [Morrison 1999].
As a working fluid’s temperature increases and decreases so does the pressure. If a volume of gas could be reheated and cooled quickly, the resulting pressure differential can be used to perform work. Since rapidly cooling a contained volume of fluid is not practical, a means for displacing the fluid between a hot and cold end is necessary to sustain a pressure wave. Such is the underlying principle that drives the displacer-type Stirling Engines.
The displacer piston shown in Figure 5 shuttles the fluid back and forth to the hot and cold ends respectively and creates a pressure wave. A displacer-type Stirling Engine consists of a displace piston, working fluid space, power piston, bounce space as shown in Figure 5. The displacer piston does not compress the gas; a sufficiently large gap exists between the inner diameter of the housing and the outer diameter of the displacer piston to allow the fluid to flow freely from the hot and cold ends [Tuttle 1998]. Thus, the sole purpose of the displacer is, as its name indicates, to displace the fluid (stages 2 to 3 and 4 to 1 in Figure 6 show the displacer moving air at a constant volume). Isothermal compression occurs from stages 1 to 2, as most of the fluid is in the cold end; see Figure 6. The resulting low pressure at stage 2 causes the displacer to drop, thus shuttling the fluid to the hot end. As the working fluid heats up the pressure rises in the constant volume heat addition process from stages 2 to 3. The high pressure forces the power piston out and produces a work output from 3 to 4 during isothermal expansion. A constant volume heat rejection occurs from stages 4 to 1.
Free-Piston Stirling Convertor. The displacer and power piston respond to the pressure differential in the working space as mass-spring-damper systems [STC 2002]. Both the power and displacer pistons are mounted with flexure bearings, which act as springs allowing each to move in the axial direction. The system is dominated by the power piston’s mass. When properly tuned as a dynamic mass-spring-damper system, a free-piston configuration will resonate on its own at the systems natural frequency [Shelter 2002].
SRG Convertor Components
This section contains a description of important components found in an SRG Stirling Convertor.
Flexure Bearings and Clearance Seals. A major advantage of the SRG is its ability to operate maintenance-free over the life of the mission. STC’s convertor design is marketed as a long-lasting, reliable dynamic system that generates no wear in the absence of contacting parts. The proven high reliability of the convertor is made possible by using flexure bearings and clearance seals.
Previously life-limiting sliding seals found in most Stirling Kinematic Engines are replaced with integrated flexure bearings and clearance seals. Figure 7 shows two axially mounted flexure bearings [STC 2002b]. The flexures support the power piston or the displacer piston by allowing them to move axially and constraining them with radial stiffness. Table 1 shows some advantages of flexure bearings.
Table 1. Advantages of Flexure Bearings [STC 2002]
1 / No friction between moving parts2 / Low manufacturing costs
3 / Highly accurate modeling techniques
4 / Mechanically simple and highly predictable
5 / Proven long operating life
Fabricated from sheet metal at very precise tolerances, spiral kerfs allow for elastic deformation; see Figure 8. The center of the flexure bearing is clamped concentrically to the piston at the center and clamped to the Stirling machine along the circumference [White 1996]. A mounted flexure flexing upward and downward is pictured in Figure9a and 9b respectively [STC 2002]. The spiral kerfs allow for free axial movement while their thick arms resist radial movement.
Between the piston and the housing, there is a clearance seal (an extremely small gap less than 0.025 mm) [White 1996]. This clearance seal prevents dynamic leaking of the convertor pressure vessel. Thus, the convertor can be hermetically sealed to form an airtight system.
Heater Head. The heater head conducts thermal energy from the GPHS module and delivers it to the Stirling power cycle. At elevated temperatures the grain size of the Inconel 718 material increases. This makes the creep the dominate failure mode over the 14 year mission life [Shah 2004]. Probabilistic and deterministic Heater life assessment has been conducted at GRC [Thieme, 2005].
Higher thermal efficiencies can be achieved if the SRG’s hot end temperature at the heater head could be raised. Mass optimizations indicate that reducing the cooling systems mass by raising the cold end temperature is ideal; this objective can also be achieved by subsequently increasing the hot end temperature at the heater head. In Figure10 the same efficiency can be obtained for varying rejection, temperature if the hot end temperature is increased [White 2001]. The TDC convertor IN718 heater head is designed to operate at a maximum temperature of 650 C. Hot end temperatures of 850 C can be obtained using advanced superalloys and 1200 C with refractory metals and ceramics.