Multi-Mission Radioisotope Thermoelectric Generator (MMRTG) and Performance Prediction Model

Abstract

Long term reliable performance is a hallmark of Radioisotope Thermoelectric Generators (RTGs). The Multi-Mission RTG (MMRTG) is the most robust, mission flexible RTG yet developed, capable of operating in both planetary surface environments and deep space vacuum. These systems are capable of mission lengths of more than 14 years. The first MMRTG application is a relatively short mission of 3-4 years providing power for the Mars Science Laboratory (MSL) Rover. Whether the mission is long or short, the ability of this reliable solid-state converter to provide stable and very predictable power is a feature that is highly valued by the NASA community. The system is fueled by GPHS Pu 238 heat source modules, with an 87.5 year half life. The fuel decay results in an unavoidable reduction in the MMRTG power output over the short 3+ year MSL anticipated mission duration. There is an additional reduction effect known as thermoelectric (TE) degradation, and the reduction is dependent on the TE element design and operating conditions. Just as past TE systems used on missions from Viking to Galileo have all been well characterized and well behaved (due to the solid state processes), the TE elements used in the MMRTG are also showing very stable and predictable life-time power curves. Characterizing this lifetime power curve to enable reliable generator performance predictions is the subject of this paper. The performance prediction modeling approach will be described as well as tests performed to measure the absolute level of these TE performance changes with time. A recently developed hybrid model is recommended which combines a heritage model with more recent test results, especially for extrapolated long term predictions. Comparison with the heritage model will be shown and explanations provided for the differences between the heritage and MMRTG TE life prediction.