The reserve power of NASA’s Voyager spacecraft and the intricacies of RTG-based power systems

Launched in 1977, the Voyager 1 and 2 spacecraft have been in continuous service for over 45 years, making their way from Earth to the outer planets of our solar system and beyond. Thanks to radioisotope thermoelectric generators (RTGs), which delivered 470W at launch, they can function just as well in the darkness of space as they do in the confines of our sunlit solar system. But since nothing in the universe is truly infinite, these RTGs also wear out over time, both from the natural decay of their radioactive source and from the degradation of thermocouples.

Despite this gradual drop in power, NASA recently announced that Voyager 2 has a previously seemingly unknown source of backup power that will delay shutdowns of more scientific instruments for a few more years. The change essentially bypasses a voltage regulator circuit and associated backup power system, freeing up the consumed power for scientific instruments that would otherwise have been shut down years earlier.

While this is good news in itself, it’s also noteworthy because Voyager’s 45-plus-year-old multi-hundred-watt (MHW) RTGs are the predecessors of the RTGs that, 17 years later, still power the New Horizons spacecraft , and the Mars Science Laboratory ( Curiosity ) for over 10 years, demonstrating the value of RTGs in long-term exploration missions.

Although the rationale behind an RTG is fairly simple, its design has changed significantly since the US installed a SNAP-3 RTG on the Transit 4B satellite in 1961.

need for power

Apollo astronaut photo of a SNAP-27 RTG on the moon. (Credit NASA)

Even on Earth, it can be difficult to find a reliable source of power that will last for years or even decades, which is why NASA’s Systems for Nuclear Auxiliary Power (SNAP) development program has produced RTGs suitable for both terrestrial and space-based use the SNAP-3 was the first to make it into space. This specific RTG only generated 2.5W, and the satellites also had solar panels and NiCd batteries. But as a space-based RTG test bed, SNAP-3 laid the groundwork for successive NASA missions.

The SNAP-19 provided the power (~30W per RTG) for the Viking 1 and 2 and Pioneer 10 and 11 landers. Five SNAP-27 units provided the power for the Apollo Lunar Surface Experiments Packages (ALSEP). left on the moon by the astronauts Apollo 12, 14, 15, 16 and 17. Each SNAP-27 unit delivered approximately 75 W at 30 VDC of power from its 3.8 kg plutonium-238 fuel rod, which was thermally rated at 1,250 W. After ten years, a SNAP-27 is still producing over 90% of its rated electrical power, allowing any ALSEP to transmit moonquake data and other information recorded by its instruments for as long as the power budget allows.

By the time Apollo Project support operations ceased in 1977, the ALSEPs were left with only transmitters on. Apollo 13’s SNAP-27 unit (attached to the outside of the Lunar Module) has managed to re-enter Earth, where it still lies – intact – at the bottom of the Tonga Trench in the Pacific Ocean.

The relative inefficiency of RTGs was already evident back when the SNAP-10A experiment demonstrated a compact 500 W fission reactor in an ion-powered satellite that significantly outperformed the SNAP RTGs. Despite being much more powerful per unit volume and nuclear fuel, thermocouple-based RTGs have the advantage of having absolutely no moving parts and only passive cooling requirements. This allows them to be literally glued to a spacecraft, satellite or vehicle, with thermal radiation and/or convection providing the cold side for the thermocouple.

These thermocouples use the Seebeck effect, the inverse Peltier effect, to essentially convert the thermal gradient between two dissimilar electrically conductive materials into a generator. Much of the challenge with thermocouple-based RTGs has been finding the most efficient and long-lasting composition. Although Rankine, Brayton and Stirling cycle RTGs have also been experimented with, these have the major disadvantage of moving mechanical parts and requiring seals and lubrication.

Considering the 45+ year lifespan of the Voyager MHW RTGs with their relatively old silicon germanium (SiGe) thermocouples, the disadvantages of adding mechanical components should be obvious. Especially when you consider that the MHW RTG has had two generations of successors so far.

Not your 1970s RTG

While Voyager’s MHW-RTG was developed by NASA specifically for the mission, its successor bears the creative title general heat source (GPHS) RTG, was developed by General Electric’s Space Division and subsequently used on the Ulysses (1990 – 2009), Galileo (1989 – 2003), Cassini-Huygens (1997 – 2017) and New Horizons (2006 – ) missions. Each GPHS-RTG produces about 300W of electrical power from 4,400W of thermal, while still using similar silicon-germanium thermocouples.

An interesting aside here is that even the solar-powered Mars rovers contain a radioisotope unit, albeit in the form of a radioisotope heating unit (RHU), with the Sojourner rover having three of these RHUs and Spirit & Opportunity having eight RHUs each. These RHUs provide a constant source of heat, allowing scarce power from solar panels and batteries to be used for tasks other than running heaters.

The GPHS module provides constant heat for a radioisotope power system. (Source: NASA)

Meanwhile, the currently active Mars rovers Curiosity and its twin Perseverance receive both power and heat from a single one Multi-mission radioisotopic thermoelectric generator (MMRTG) unit. These RTGs use PbTe/TAGS thermoelectric pairs, ie a lead/tellurium alloy for one side and tellurium (Te), silver (Ag), germanium (Ge) and antimony (Sb) for the other side of the pair. The MMRTG is designed to last up to 17 years, but likely exceeds its design specs by a lot, as the MHW RTGs and others have done. The Pu-238 fuel with an MMRTG is contained in GPHS (General Purpose Heat Source) modules which are designed to protect the fuel from damage.

The main cause of failure of the SiGe thermocouples was the migration of the germanium over time, causing sublimation. This was prevented in later designs by coating the SiGe thermocouples with silicon nitride. The PbTe/TAGS thermocouples should provide further stability in this regard, and the MMRTGs in Curiosity and Perseverance have served as real life tests.

A matter of fuel

The Voyager 1 and 2 probes are well out of range for a large service and maintenance session, so NASA had to get creative to optimize power consumption. Although the backup circuit may have been considered necessary in the 1970s in case there were power fluctuations from any of the three RTGs on each spacecraft, there is enough real world monitoring data to support the suggestion that it might be redundant barring extraneous interference.

With nearly 46 years of data from Voyager RTGs, we can now see that thermocouple stability is essential to maintaining constant power output, while the decay of the plutonium-238 fuel source is far easier to model and predict. Well, since with the MMRTG units we should have addressed many of the issues that have caused thermocouple degradation over time. The only missing ingredient is the Pu-238 fuel.

Most of the Pu-238 that the US originally sourced from the Savannah River Site (SRS) before that facility and its specialty reactors shut down in 1988. Thereafter, the US would import Pu-238 from Russia before the latter’s stockpiles would also start to run out, causing the US to run out of one of the best radioactive isotopes for use in RTGs for long-term missions. With a short half-life of 87.7 years and only one alpha decay, Pu-238 is fairly benign to surrounding materials while delivering significant amounts of thermal output.

With only enough Pu-238 left for the two MMRTGs in the current Mars rovers and two more thereafter, the US has now resumed Pu-238 production. Although Pu-238 can be produced in a number of ways, the preferred way seems to be to use stored neptunium-237 and subject it to neutrons in fission reactors or similar neutron sources to produce Pu-238 through neutron capture. According to NASA, about 1.5 kg of Pu-238 per year should be enough to meet the needs of future space missions.

A tiny spaceship in the dark

Voyager 1 is currently 159.14 AU (23.807 billion km) from Earth, and Voyager 2 is only marginally closer to Earth at 133.03 AU. As a project that has its roots in the space race and ended up surviving not only many of its creators but also the geopolitics of the time, it’s perhaps one of the few man-made constants that we can all identify with in some fashion.

As carriers of the golden discs that hold the essence of humanity, extending the lifespan of these starships goes beyond the mere science they can perform in the darkness of space. With each additional year, we can learn a little more and see a little more of what awaits humanity beyond the reach of this more ordinary, remote solar system.