A plenty has been written and said about reaction wheels on this site, but there's still one thing I'm curious about: what's the "nominal/rest" speed of a reaction wheel? Say, we've just finished the procedure of desaturating the reaction wheel of a satellite, it went exceptionally well, the satellite is currently maintaining its desired orientation very nicely. What is the RPM of the reaction wheel?

Is it at stop / near stop, or does it turn at some nominal rate, say, halfway between stop and saturation speed, or is it in some other state yet?

  • $\begingroup$ Not worth an answer, but some hints: You might want to prevent a stopped wheel (grease can stick parts together, you need a high torque to overcome static friction) For similar reasons (wearing) you might want to prevent reversing the rotation. $\endgroup$ – asdfex Mar 8 '17 at 17:03
  • $\begingroup$ I did some digging in reaction wheel spec sheets. The "nominal speed" quoted there is the highest speed allowed under normal running conditions. So that's a different use of the word 'nominal' than is used in this question. The rest speed you're after is not mentioned in those spec sheets. $\endgroup$ – Hobbes Mar 9 '17 at 10:54
  • $\begingroup$ @Hobbes: Probably it will be half of that nominal speed. What was the nominal speed? $\endgroup$ – SF. Mar 9 '17 at 11:13
  • 1
    $\begingroup$ Blue Canyon: no speed listed at all. Sinclair Interplanetary: no speed listed. Honeywell HR0610: 6000 rpm. $\endgroup$ – Hobbes Mar 9 '17 at 11:15

It depends on the system architecture and the needs of the mission. If you're trying to momentum bias the spacecraft then it will be running at a nominal speed. If you really want to conserve energy then you might bring it down all the way.

If you're running four reaction wheels arranged tetrahedrally (as is often done to provide failure tolerance) then you can more easily spin reaction wheels all the way down and back up without loss of control authority because you can usually compensate with the other three wheels if going just outside the deadzone would be too much angular momentum.

The nominal zero solution tends to be used more frequently due to the very precise pointing requirements of many spacecraft, as well as wear issues with stopping and reversing reaction wheels. In this case the nominal zero is put in the middle of the usable angular speed range. This can be difficult for nanosatellites to achieve due to requirements that they be shut down entirely when launched.

Also, strictly speaking a reaction wheel that doesn't operate around zero is called a momentum wheel.

  • 4
    $\begingroup$ The classic engineering answer (which is wholly correct): it depends. $\endgroup$ – Tristan Mar 8 '17 at 15:59
  • $\begingroup$ I'd prefer if the answer discussed some actual (used in practice) solutions and gave some numbers. $\endgroup$ – SF. Mar 8 '17 at 20:17
  • $\begingroup$ Those are both actual solutions. $\endgroup$ – Schlusstein Mar 8 '17 at 20:21
  • 1
    $\begingroup$ Well as I mentioned, for a desaturated momentum wheel that will be about of the maximum rate, and for a reaction wheel proper it's zero. $\endgroup$ – Schlusstein Mar 8 '17 at 22:37
  • 1
    $\begingroup$ The last sentence does not seem correct. i think one of the mentions of 'momentum wheel' was meant to be 'reaction wheel'. $\endgroup$ – OrangePeel52 Mar 9 '17 at 18:54

Not quite what you are asking about, but the International Space Station Control Moment Gyros turn at a nominally constant speed of 6600 rpm. The control function comes about by commanding the inner and outer gimbals to specific locations.

enter image description here


  • 1
    $\begingroup$ ah, a completely different animal though... $\endgroup$ – SF. Mar 8 '17 at 15:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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