The MHW-RTG model Radioisotope Thermoelectric Generators on the two Voyager spacecraft started out at about 1000C and 300C on the hot and cold sides, but will have dropped somewhat due to slow decay of the radioisotope thermal output.

Since the computer is still working, presumably the electronics is warmer than the two spacecrafts' extremities by self-heating or other sources of heat within the spacecraft.

But roughly how cold are the extremities; parts of the spacecraft that are warmed mostly by the distant Sun and a bit by the cosmic microwave background (CMB)?

Would there be some extremities of each spacecraft that are now permanently in the shadow of the high-gain antenna dishes pointed at Earth and Sun, and so not even warmed directly by sunlight?

If there is some telemetry, that would be great to know. If there is a way to calculate by first principles, that's fine also.

Is it colder than LOX? Than solid oxygen (SOX)? Than LH2? Will gas stick to it and start building up hydrogen ice?

  • $\begingroup$ The hydrazin used for attitude control goes solid below 2 °C, the tank has heaters to prevent that. The probe has several layers of thermal insulation and plutonium heater units as well as electrical heaters. $\endgroup$
    – Uwe
    Sep 20, 2017 at 9:37
  • 10
    $\begingroup$ "Would you eat them with a LOX? Would you eat them wearing SOX?" $\endgroup$
    – Bear
    Sep 20, 2017 at 13:33
  • 3
    $\begingroup$ Warming by the distant Sun does not work well, at the moment Voyager 1 is 139,7 AU away from Sun. The radation power per one square meter at 1 AU is about 1350 W, at Voyager 1 there are only 69 mW/m² left. $\endgroup$
    – Uwe
    Sep 20, 2017 at 20:48

2 Answers 2


First-principles calculations of the equilibrium temperatures of objects illuminated by the sun aren't too difficult—IF they have simple shapes, aren't rotating, and you know very well both the visible-wavelength absorptivities and IR emissivities of all the surface materials involved, and the thermal conductivities of all the internal materials. The Voyagers aren't rotating, but the shapes are definitely not simple, and I'm sure nobody measured the absorption/emission spectra of all the surface materials, especially in the far-IR, where most would be radiating now. A good chunk of the spacecraft is behind that HGA. Only the mag boom, the PRA/PWA antennas, the RTG, and the scan platform and part of its boom aren't shaded. In the shadow there are three primary sources of heat: heat conducted through the HGA to its rear-facing side and then radiated, though with the HGA's high albedo that won't be as much heat as it could be, were it black; heat radiated from the RTG; and heat radiated from the spacecraft bus that is being kept warm by heat dissipated by the avionics, electric heaters powered by the RTG (as are the avionics), and some RHUs there as well. There is plenty of multi-layer insulation (MLI) around that bus, so most of its radiated heat will come from its louvered radiators. This winds up being a very complex thermal equilibrium calculation best done with a CAD model and thermal transfer codes, and someone probably did a simplified version of it (early-1970's technology!) to see how the spacecraft would do at Neptune. I'm willing to bet that when the spacecraft was being designed nobody paid to do that analysis for 140 AU!

But we can draw analogies to some of the more distant moons in the solar system, with reliable data available on Wikipedia, based on Voyager science papers. The moons of Uranus have varying visible-wavelength albedos and vary in mean surface temperature from ~60 K to ~80 K, at just short of 20 AU distance. Triton, at Neptune, 30 AU from the sun and with a high albedo, has a surface temperature of ~38 K. Pluto, at ~35 AU right now and with a lower average albedo, averages ~44 K. Some shaded parts of Voyager, such as far ends of the solid rocket motor (SRM) struts, will be a bit colder than that. They'll see ~π/2 steradians of the (well-insulated!) bottom of the bus, and ~7π/2 steradians of 2.7 K. That puts them colder than LOX and SOX! But it's not cold enough to collect hydrogen, which boils at just above 20 K under 1 atm pressure (https://en.wikipedia.org/wiki/Liquid_hydrogen). In vacuum conditions you'd have to build it up as a solid, at 14 K or colder.

Voyager is running on a skeleton crew, and even the crew they have is only part-time on Voyager. But I know Suzie Dodd, the current Voyager Project Manager, and maybe next week I can get some info from her. She was on the technical team when I was with Voyager, so I know she's not just a dollars-and-full-time-equivalent-employees manager!


Voyager was designed to operate at -35 °C. As the RTG output drops, heaters have been switched off. The spectrometer currently runs below -79 °C. Its temperature is not known accurately, as it's dropped below the operating range of the temperature sensor.

The magnetometer on its long boom is heated by a Pu-238 heater (page 20). The other extremity is the camera platform, which was switched off years ago. Haven't found temperature data for either though (no engineering data in the Voyager data collections).

Radioisotope heating units, small non-power-using heat elements that generate one watt of thermal energy, are located on the magnetometer sensors and the Sun sensors. No radioisotope heating units are used near instruments that detect charged particles. Electric heaters are located throughout the spacecraft to provide additional heat during portions of the mission. Many of the heaters are turned off when their respective valves, instruments or subassemblies are on and dissipating power.

The vidicon tubes in the imaging system were designed to operate at temperatures above +10 °C. The hydrazine tanks need to be kept above +2 °C. I suspect the tape recorder needs to be kept fairly warm as well.

The main body of the spacecraft is shaded by the HGA these days (after the last planetary encounter there was no reason anymore to point the HGA away from Earth), only the RTG, camera platform and magnetometer boom are sunlit.

  • 2
    $\begingroup$ I'm pretty sure there are parts of voyager at much lower temperatures than that. Things on booms, things in the shade, etc. Perhaps the -35 °C figure applies only to things with active semiconductor components, where dopant activation and therefore carrier concentration is a strong function of temperature. I'm asking mostly about extremities, and perhaps those permanently shaded by the dish. But the tiny 1W ${}^{238}$Pu heaters are really interesting to read about!! I'd never heard about those. Can you consider including the last paragraph on page 20 as a block quote? $\endgroup$
    – uhoh
    Sep 20, 2017 at 9:19
  • $\begingroup$ Yeah, these numbers seem to show the hottest parts, rather than the coldest! $\endgroup$
    – Innovine
    Sep 20, 2017 at 10:18
  • $\begingroup$ The minimum temperature for parts in the shade of the HGA will be the temperature of the cosmic microwave background radiation, 2.7 K. $\endgroup$
    – Uwe
    Sep 21, 2017 at 8:48
  • $\begingroup$ The spacecraft body is insulated, and contains heaters plus active components dissipating ~200W, plus the heat flux from the RTGs. 2.7K seems unlikely to me. $\endgroup$
    – Hobbes
    Sep 21, 2017 at 8:52
  • $\begingroup$ 2.7 K is only a lower limit for parts in the shade, without heaters and thermaly isolated to the spacecraft body. But the temperature will converge very slowly to this limit in the future when most plutonium has decayed and heat flux from the RTGs and the heaters will be tiny. $\endgroup$
    – Uwe
    Sep 21, 2017 at 12:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.