tions involved in developing the ASRG. In addition, the RPS program is continuing to work on a configuration management plan and other related plans and processes.

A failure mode, effects, and criticality assessment of the ASRG engineering unit identified 51 single-point failures (SPFs). By comparison, the design of the RTG-GPHS (the standard RPS used prior to the MMRTG) has only 17 SPFs. However, a numerical comparison of the number of SPFs does not provide a good understanding of the relative reliability of the two types of devices. The likelihood of the SPFs must also be understood. For example, about 80 percent of the SPFs on the ASRG engineering unit are structural in nature, and the designers believe that the likelihood of these failures has been reduced to very low levels through the use of conservative structural designs. In any case, the issue is not whether an ASRG will be as reliable as historic RTGs; the issue is whether mission managers can be convinced that an ASRG is sufficiently reliable to meet engineering and programmatic requirements for a given mission.

NASA has used fault tree and probabilistic analysis techniques to estimate that system-level reliability is 0.967 for an ASRG at full-power operation over the entire 17-year design life. System electronics (i.e., the electronics required to control and synchronize the ASCs and to convert the electrical output from ac to dc) have been identified as the major contributor to the estimated probability of failure. System-level reliability at half-power operation (that is, the probability that an ASRG will have at least one of its two converters functioning and producing power at the end of the 17-year design life) has been estimated to be 0.984. Extended life tests will provide additional data regarding reliability, but there is not enough time or money to build enough ASRGs and then test them for long enough to determine rigorously what level of reliability they will have over a 17-year lifetime. However, this has been the case for earlier RPSs—and for other critical spacecraft hardware as well. There has never been a numeric reliability requirement specification for an RTG, and NASA does not intend to establish one for the ASRG.


NASA appears currently to be well positioned with regard to key RPS research and development facilities. These facilities are located at GRC and JPL.6 The facilities at greatest immediate risk are those associated with advanced RPS research (e.g., advanced thermoelectric and TPV research facilities). NASA has not yet lost any critical RPS facilities, and the projected budget seems adequate to sustain necessary research and development facilities. However, there are concerns related to other facilities that are necessary for the production of flight systems.

The MMRTG will fly on the Mars Science Laboratory, but this is the only mission that is firmly committed to using the MMRTG. As this work is completed, the industry teams that developed and built the MMRTG are expected to disband, and the industry facilities are expected to be reconfigured for other purposes. It remains to be seen if NASA will sustain work on MMRTGs to keep the MMRTG industrial teams and facilities intact and related infrastructure in place until a final decision is made on what system will power OPF 1. If the ability to manufacture MMRTGs is not sustained at least until (1) the ASRG is demonstrated to be flight ready and (2) NASA commits to using ASRGs (or another comparable RPS) for long-life, deep-space missions, then even with an adequate supply of 238Pu, the United States could lose the ability to manufacture any RPSs, at least for a time.

FINDING. Multi-Mission Radioisotope Thermoelectric Generators. It is important to the national interest to maintain the capability to produce Multi-Mission Radioisotope Thermoelectric Generators, given that proven replacements do not now exist.

RECOMMENDATION. Multi-Mission Radioisotope Thermoelectric Generators. NASA and/or the Department of Energy should maintain the ability to produce Multi-Mission Radioisotope Thermoelectric Generators.


The next major milestones in the advancement of ASRGs are to freeze the design of the ASRG, to conduct system testing that verifies that all credible life-limiting mechanisms have been identified and assessed, and to demonstrate that ASRGs are ready for flight. However, neither the DOE nor NASA have formal guidance or requirements concerning what constitutes flight readiness for RPSs. In general, RPSs (and other systems) on spacecraft for deep-space missions are flight ready when the project manager for that mission says they are flight ready. Given this situation, ongoing efforts to advance ASRG technology and demonstrate that it is flight ready are being guided by experience with past programs and researchers’ best guess about the needs and expectations of project managers for future missions. While this approach has enabled progress, the establishment of formal guidance and processes for flight certification of RPSs in general and ASRGs in particular would facilitate the acceptance of ASRGs as a viable option for deep-space missions and reduce the impact that the limited supply of 238Pu will have on NASA’s ability to complete important space missions.

FINDING. Flight Readiness. NASA does not have a broadly accepted set of requirements and processes for demonstrating that new technology is flight ready and for committing to its use.


This section deals with facilities associated with development and fabrication of RPS technologies and RPS converters. DOE 238Pu production and RPS assembly and testing facilities are addressed in Chapter 2.

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement