I am working on a project for science fair that involves a deployable component that will not be heated on a CubeSat. I am trying to find data for the skin temperature of a satellite (or the ISS) over a single orbit. Averages and min/maxes are easy to find, but an actual temperature curve of an orbit has been elusive. Does anyone know where this data is available or what the search term we are missing is?

--------Edit: Thank you for all the responses so far. The deployable component is a digital display. For the project I am testing three options:

  1. Something like this LCD
  2. This OLED
  3. 8x8 LED matrix

My science fair project is about finding out if these displays can withstand vacuum and the temperatures they will be exposed to. Our satellite is a 2U CubeSat and we expect the screen will be mounted 10 to 15 centimeters from the satellite once deployed. The glass of the display will be facing the satellite and the back of the display will always be facing the earth. The setup is to allow an onboard camera to take pictures of the screen with the earth in the background. My science fair project is about testing the suitability of these screens for the CubeSat mission. It is being launched from the space station and will therefore have approximately 15 orbits/day.

I already know the LCD is not a suitable candidate for deployment, but we are including it in our testing to demonstrate why.

Once de-tumbling is complete we expect pointing to be accurate to about 20 degrees and expect little rotation. On the deployable screen there are no plans currently to heat. The plan will be to operate it only in the window of the components operational temperature.

We are trying to find out two things with this question:

  1. What is a comparable temp range this screen will be exposed to (even roughly estimated) so I can test in the lab something that approximates these conditions (hopefully we can test something wider than worst case).
  2. We want to get some idea of the amount of time these digital displays will spend at a temperature within their operational range each orbit.

Also - as an aside, I have not been able to find an example of other digital displays in space (which makes sense, as who is looking at them...) but if anyone knows of an example it would be great to know about it.

  • 3
    $\begingroup$ This is a great question, Welcome to Space! There will be a wide range of possible answers depending on the surface of the object, so any one single curve will not be representative. Also the heating/cooling timescale is short, so things like thermal conductivity between the surface components and other parts, and the rotation status (tumbling versus nadir-pointing or fixed inertial attitude) are important as well. So I think you can have a really wide range of answers, wider than you need. It will therefore be better if you can describe a little more about the deployable component. $\endgroup$
    – uhoh
    Commented Apr 16, 2019 at 5:01
  • 1
    $\begingroup$ also see chapter 9 in this random link from the internet: soaneemrana.org/onewebmedia/… $\endgroup$
    – uhoh
    Commented Apr 16, 2019 at 5:08
  • 1
    $\begingroup$ This is his father: When he was asking me I naively thought this was something that would have a fairly generic range, and am fast learning how to think in terms of operation in a vacuum - something he is picking up far quicker than I am. Hopefully the context he has now added will help point him at something that will let him identify a min/max bounds he can derive... although we are both far less confident now :S $\endgroup$
    – Matthew
    Commented Apr 17, 2019 at 2:32
  • 1
    $\begingroup$ For anyone else looking we found a great resource in: shop.oreilly.com/product/0636920021612.do $\endgroup$
    – Matthew
    Commented Apr 20, 2019 at 16:32
  • 1
    $\begingroup$ Comment is serious - book was really helpful in adding to the below answers for what we needed for data. $\endgroup$
    – Matthew
    Commented Apr 22, 2019 at 20:45

2 Answers 2


The temperature development of a satellite in LEO depends on a variety of factors. How (quickly) does the satellite rotate, how much is it in eclipse (night), what kind of radiators or internal heat sources exist, etc. When a space mission is being planned, thermal control engineers with dedicated software model the temperature development. What applies for one satellite does not apply for another.

If you want to look at specific examples, NOAA publishes graphs for temperatures for meteorological satellites. For example, you can find the status of NOAA-18 HIRS, in this case for the secondary telescope:

NOAA-18 HIRS Secondary Telescope Temperature

Or the temperature for the NOAA-15 AVHRR motor housing:

NOAA-15 AVHRR motor housing temperature

All but the top panel show orbital averages.

You can also download the raw data through the NOAA CLASS Archive, but file formats are old and not user friendly (for example, AVHRR data are packed as 13-bit words and the oldest HIRS satellite data headers mix ASCII and EBCDIC fields).

If you browse the various components you will see that they behave quite differently. There is no active cooling on those satellites, only passive cooling. Sometimes components of the satellite are actively heated for decontamination. As you can see, temperature may be widely varying. To know specifically why, you may have to talk to the engineers at NOAA, NASA, EUMETSAT, ESA, or whatever agency controls the satellite of interest.

To find the temperature curve for your cubesat component is non-trivial and any temperature curve you find on the web is unlikely to match the one for your component. If you have access to the right time and skills, you can try to model it. Or you can design your component to handle the extremes / worst case, add some thermistors to it for monitoring purposes, and hope for the best.

  • $\begingroup$ The temperatures seem to be not only human-survivable, but sometimes even human-comfortable. Is that a coincidence, or are the satellites designed to be that way because that's where electronic components are already designed to work best? $\endgroup$ Commented Apr 17, 2019 at 13:07
  • 1
    $\begingroup$ @RayButterworth The average temperature in Low Earth Orbit is similar to the average temperature on Earth minus the Greenhouse Effect. Add to this some internal heating and the "natural" temperature is quite comparable to what humans find comfortable. Of course a thermal design that reduces temperature swings also increases component lifetime. $\endgroup$
    – gerrit
    Commented Apr 17, 2019 at 13:24

This is a supplemental spherical cow answer, followed by links to other questions, answers, and comments there that you may find helpful when working on this project together.

Per ULA's plan for LH2/LOX 2nd stage that can maintain propellant for an extended period of time? the temperature of a spherical object in Sunlight with a uniform surface defined by a visible light reflectivity or albedo $a_{vis}$ and a thermal infared emissivity $e_{ir}$ is given by

$$T \sim \left( \frac{(1-a_{vis})}{e_{ir}} \frac{I_{Sun}}{4 \sigma} \right)^{1/4}$$

where the solar constant $I_{Sun}$ is about 1360 Watts/meter² and the Stefan–Boltzmann constant $\sigma$ is about $5.67 \times 10^{-8}$ Watts/meter²/K⁴.

Most things that are not clean, polished, conductive metal surfaces have a pretty high emissivity, and many infrared thermometers have assume a default value of 0.9 or 0.95 for the object you are measuring when you turn them on. Even things that are white, or even transparent in visible light usually have a pretty high thermal infrared emissivity.

              visible light albedo (reflectivity)
 infrared        a=0.1       a=0.5      a=0.9
emissivity       "dark"     "medium"   "light"
----------     --------    --------    --------
    0.9          279         241         161     paints, glass, anodized metal, PCB (e.g. FR-4)
    0.5          324         279         187
    0.1          484         418         279     polished metal

A quick look shows that for materials where the thermal infrared emissivity and visible light albedos are similar, the equilibrium temperature is about 279 K which is about 6 degrees C, above freezing but a little cooler than the air temperature of a commercial server farm.

For an example worked on an everyday object see this answer where I've cooked a hot dog in space.

Early satellites were coated in high-emissivity, yet mirror-like surfaces by applying a special coating to reflective metal. Perhaps the electronics inside these first satellites could survive some cold exposure, but once they got too hot they'd be irreversibly fried.

@RydgeMulford's authoritative answer to Why are deeper folds better for absorption? explains, some more exotic surfaces to manage temperature.

It is also common for some satellites to have louvres or doors to cover or expose materials of one set of emissivity and albedo with that of a different set. For more on that, see:

In comments below this answer I've said:

The top of the linked page in the book Elements of Space Technology by R. X. Meyer it says "If this surface is provided with a high ratio of solar radiation absorptivity to ambient temperature emissivity, control over a wide range of temperatures can be obtained."

That linked page is to Chapter 7 (Spacecraft Thermal Design) in Elements of Space Technology for Aerospace Engineers By Rudolf X. Meyer, Rudolph X. Meyer. It may be read in google books, found on the shelves of many university library, or perhaps obtained by inter-library loan. Flipping through a physical book could be fun and inspiring to both of you. Some people (myself included) feel that flipping through good books, and especially library shelves of them is fundamentally superior to just looking at "the internet".

A piece of transparent glass, or one with a mirror second surface would absorb little visible light but radiate well, so it would tend to run fairly cold in space even in sunlight. You can see images of seconds surface mirrors used as "space radiators" in the question What are these very large, square panels on Inmarsat 5? and read about them in @Puffin's excellent answer there. Also see more images in What is the function of this array of what looks like mirrors on TESS? and in @CarlosN's excellent answer as well.

Something that was dark and highly absorbing in visible light, but had a low IR emissivity would run quite hot, but I can't think of a standard material that has that property.


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.