As soon as they are assembled MMRTGs continuously produce about 2,000 watts of heat energy. That dips only slightly when they are connected to a load and producing electrical power (about 125 watts initially).

In deep space and on Mars some of the heat from curiosity's MMRTG was/is removed by circulating fluid to keep things from overheating during transit and to keep itself warm during the martian nights on the surface.

Space.com's Going nuclear: NASA's Perseverance Mars rover gets its power source for July 30 launch explains that Perseverance is now "hot" in more ways than one; it is now emitting roughly 2000 watts of heat continuously and has an (fairly) uninterruptible supply of electricity.

While the naked rover remains in an air conditioned laboratory this should be no problem due to convection at 1 bar, but it needs to be moved in a protected way and put into its "clamshell" for deep space flight and a payload fairing for launch.

(I've just asked What is the equivalent of Curiosity's "MSL" in the context of Perseverance? What's the official name of the mission? Is the distinction similar? because I don't know a better term than "clamshell" at the moment)

Question(s): From this point forward how is the heat being removed?

Presumably when in deep space the 2,000 watts is radiated fairly efficiently into space by some radiator exposed to "the cold of deep space" such that it doesn't have to get too hot to do so. But once in its "clamshell" how hot does its radiator have to be to radiate agains the walls of a room at 293 Kelvin rather than the cosmic microwave background of 2.7 Kelvin?

Curiosity's "clamshell" from this answer to Where does MSL end and Curiosity begin?

Curiosity's "clamshell" from https://space.stackexchange.com/a/26507/12102

For more "clamshell" imagery see Why are planetary probe RTGs tilted at a jaunty angle?

NASA https://photojournal.jpl.nasa.gov/catalog/PIA23305

enter image description here


1 Answer 1


2kW is not that much on Earth

You've mentioned radiation and convection in your answer (you forgot conduction). Turns out the properties of Earth's atmosphere make conduction and convection way better than radiation for moving heat around.

For an illustration, consider the size of a portable, 2kW, oil-filled radiator: this one lists the size as 620x150x520mm. With nine fins, that gives a total heat transfer area of 620x150x18 = 1.67m². Without fans, this radiator most likely reaches a maximum of somewhere around 43-45°C (source), as this temperature is less likely to cause burns if touched. This is a rough calculation, but should be close.

Thus, if you want your spacecraft to keep less than ~20°C hotter than the local environment, you should only need about 1.67m² of surface area on Earth. Guessing from the image, it appears Curiosity has about 2x(500x2000mm) = 2m² of that radiator plate in the back. While subsequent layers of the casing may raise the thermal resistance, they have much greater surface areas, leading to the additional temperature drop being less than the previous 20°C. From this page we can see that Curiosity's clamshell is about 4.5m in diameter, and the part Curiosity sits in is about 1.5m tall. Assuming a frustrum, this would give a lateral surface area of about 20m² - much larger than Curiosity's radiator. With that 10x greater surface area, you could expect a temperature difference of maybe 1/10th, or 2°C.

Thus, in order to get rid of that 2kW at a room temperature of 293K, one would expect the radiator to reach around 315K, maybe as high as 325K. That shouldn't be a problem for the rover.

  • 2
    $\begingroup$ This is a really interesting and well-reasoned physics-based answer, thanks! It reminds me of one of my first interactions in Stack Exchange years ago :-) $\endgroup$
    – uhoh
    Jul 24, 2020 at 9:14
  • 3
    $\begingroup$ In addition to this answer, payload fairings typically have conditioned air (or even purge gas) pumped into them continuously up until the moment of launch. See, e.g., chapters 6 and 7 of the Atlas Users Guide: ulalaunch.com/docs/default-source/rockets/… $\endgroup$
    – Tristan
    Jul 24, 2020 at 15:34
  • 1
    $\begingroup$ @Tristan I think there's room and perhaps even a need for an answer that explains that. I brought up the fairing in the question for just this reason. $\endgroup$
    – uhoh
    Jul 24, 2020 at 21:58
  • 3
    $\begingroup$ @uhoh That's fair. If someone wants to put together one using the link I posted (or some other source), I'll upvote it. I'm too exhausted by various events to put together a good answer myself right now. $\endgroup$
    – Tristan
    Jul 24, 2020 at 22:44
  • $\begingroup$ I just spotted "...the Rover rejected most of its own heat through the Rover heat exchanger cold plates directly to the backshell." in a block quote in this answer and I think it may be helpful here. $\endgroup$
    – uhoh
    Mar 7, 2021 at 3:34

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.