I recently got into a twitter discussion over the ISS attitude. The point of the discussion is the attitude that the ISS mantains while orbiting around the Earth.

From my reminescences of Oribital Mechanics, you need the highest inertia axis (usually the y, for convention) to point towards the Earth, so that the gravitational differential pull will not affect too negatively the Attitude control (and as a neat consequence you can decouple the x-z control loop from the y one, plus you receive a side always pointing towards Earth for communication/observation) and I remember that it is somehow connected to the uncontrolled (and unexpected) spin of Explorer 1.

The other side argued instead that having a side pointing towards Earth is required to have this communication/observation side (i.e., cause and effect are inverted).

Which side is right? Why is it that the ISS always has the "cupola side" towards Earth?

  • 1
    $\begingroup$ The current attitude (XVV) is not minimal energy and requires active control to maintain. So you can assume there are reasons for it beyond orbital mechanics. The station orbited in different orientations earlier for power reasons (when there were fewer solar panels and power was more critical) $\endgroup$
    – BowlOfRed
    Commented Apr 9, 2015 at 23:23

2 Answers 2


Which side is right?

Neither. Both of you are forgetting torque due to atmospheric drag.

Except for periods of high beta angle, the Space Station is typically maintained in a Torque Equilibrium Attitude. In this attitude, gravity gradient torque and atmospheric drag torque (ideally) cancel one another out. This means the disturbances that need to be controlled by the control moment gyros on the Station are rather small and are more or less periodic. That they are small means the CMGs can easily handle those disturbances without the need of assistance from thrusters. That they are more or less periodic means the CMGs can operate for a long time before needing to be desaturated.

  • $\begingroup$ The main attitude (+XVV) is not the only attitude that is also consistent with torque equilibrium. While that is a constraint, it cannot be the only one. This paper shows other attitudes in use (including other TEA attitudes). pims.grc.nasa.gov/pimsdocs/public/ISS%20Handbook/… $\endgroup$
    – BowlOfRed
    Commented Apr 10, 2015 at 21:36
  • $\begingroup$ @BowlOfRed - You are correct. Ignoring atmospheric drag, there are 24 TEAs in the case of a satellite with three distinct moments of inertia (e.g., the Space Station). There are still 24 solutions in the case of constant atmospheric drag, but some of them may be complex rather than real solutions. In the case of drag that isn't constant over time, location, or orientation, things get complex in a different meaning of the word "complex". $\endgroup$ Commented Apr 10, 2015 at 21:53

In a nutshell, to minimise drag caused by the trace of atmosphere up there. (The panels and radiators can be aligned as necessary in sun or shadow, but the best arrangement for the rest of the hull is to minimise area in the "forward" direction.)

You're quite right that having a long axis pointed to Earth is more stable, but it conflicts with the drag requirement.

  • $\begingroup$ uhm, "high inertia" is a short axis (the mass is away from it) and IIRC the truss axis is the longest and is not pointing towards the Earth. $\endgroup$
    – Federico
    Commented Apr 9, 2015 at 13:28
  • $\begingroup$ Isn't it also to keep the antennas pointed the right way? $\endgroup$
    – GdD
    Commented Apr 9, 2015 at 13:33
  • $\begingroup$ @GdD that's my question: that's a nice consequence or a driving requirement? $\endgroup$
    – Federico
    Commented Apr 9, 2015 at 13:35
  • $\begingroup$ The truss axis is the longest as you say. But if the long axis of the truss was aligned with Earth, the drag on the structure would be higher and it would need more frequent boosts to stop the orbit decaying. $\endgroup$
    – Andy
    Commented Apr 9, 2015 at 14:19
  • 1
    $\begingroup$ and that's fine to me, but my "requirement" is to have the shortest, not the longest, axis towards Earth; i.e. I do not say having a long axis pointed to Earth is more stable $\endgroup$
    – Federico
    Commented Apr 9, 2015 at 16:01

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.