Perhaps it goes without saying, but according to Newton’s laws “every action has an equal and opposite reaction”. How do astronauts, especially those inside small spacecraft like the Crew Dragon, not “push” the spacecraft when they bounce and push off walls? In orbit, where even button-sized ion thrusters push spacecraft, how does an astronaut pushing against a wall not cause it to move or spin?

  • 9
    $\begingroup$ When I read the title, I thought this was asking why they don't get out and push to move it, like you do when a car is stuck. :D $\endgroup$
    – Almo
    Commented Jun 3, 2020 at 14:46

4 Answers 4


When an astronaut bumps against the wall of the spacecraft, the spacecraft does gain whatever momentum the astronaut transfers to the wall. However, the astronaut loses momentum-or gains it in the opposite direction. The net result is that the center of mass of astronaut- plus-spacecraft does not move, and the combined momentum does not change.

It is worthy of note that, similarly, the combined angular momentum does not change. However, the orientation of the axes of the combined system can change, so there is not a perfect analogy between translation and rotational motion. See "How cats land on their feet".

  • 3
    $\begingroup$ this is relevant for rotation space.stackexchange.com/questions/21413/… $\endgroup$
    – anna v
    Commented May 31, 2020 at 5:59
  • 20
    $\begingroup$ You could add two perhaps obvious things: (1) Crew Dragon weighs > 10,000 kg, resulting in delta-vs from astronaut bumps which are 1/100 of the astronaut's speed, i.e. some single-digit cm/s; (2) the astronaut will quickly (especially after strong bumps), within seconds, hit an opposite wall and counteract the resulting spacecraft movement, either stopping or reversing it, so that the displacement is never more than a few cm. That could perhaps affect a docking maneuver, so children: no playing zero-g catch during docking! $\endgroup$ Commented Jun 1, 2020 at 8:53
  • 3
    $\begingroup$ @S.McGrew I'm surely not saying that. This is plain to see in what I wrote. I literally state that does not occur. $\endgroup$
    – JimmyJames
    Commented Jun 1, 2020 at 14:50
  • 3
    $\begingroup$ Let's call the ends of the cylinder 'left' and 'right'. When the astronaut is at the left end, the CoG is left of center. When they are on the right, the CoG is right of center. If the CoG of the closed system doesn't move in space, what must that mean about the position of the (ends of the) cylinder in space? $\endgroup$
    – JimmyJames
    Commented Jun 1, 2020 at 14:59
  • 2
    $\begingroup$ Where the CoG is located in the cylinder-astronaut system is different depending on where the astronaut is. Can we agree on that? The position of the CoG in space doesn't change, I know we agree on that. How can you put those two things together without the cylinder moving? Whether you can see the astronaut is irrelevant. $\endgroup$
    – JimmyJames
    Commented Jun 1, 2020 at 15:14

How do astronauts, especially those inside small spacecraft like the Crew Dragon, not “push” the spacecraft when they bounce and push off walls?

You're right that when an astronaut collides with the walls of the spacecraft, some of their momentum is transferred to the spacecraft and in turn their momentum either reduces or gets reversed in direction. However, as S. McGrew states in his answer, the centre of mass of the system (astronaut + spacecraft) doesn't move in the long term. This is a result of the law of conservation of linear momentum.

Apart from this, the mass of the astronaut is comparatively lesser than the mass of the spacecraft. So the combined centre of mass remains almost close to the centre of mass of the spacecraft irrespective of the position of the astronaut inside the pressurized compartment. So the combined centre of mass and hence the centre of mass of the spacecraft moves very little due to this collision.

In orbit, where even button-sized ion thrusters push spacecraft, how does an astronaut pushing against a wall not cause it to move or spin?

Even the world's most powerful rocket engine cannot move a spacecraft if its nozzle is mounted "inside" the spacecraft such that all the exhaust particles have no way to go out.

Diagram showing engine mounted outside and inside the spacecraft

In short this is because, any useful momentum gained by the action of the engine is cancelled by the collision of exhaust particles on the opposite side of the nozzle and hence causing no motion of the centre of mass in a long duration.

Rotation of a spacecraft is an interesting thing! In addition to reaction control thrusters (usual rocket engines), spacecrafts are equipped with reaction wheels and/or control moment gyroscopes. When the flywheel is made to rotate with high angular velocity with a motor, it gains some angular momentum in one direction. And since the spacecraft is a closed system, its angular momentum must remain constant, and hence the spacecraft gains an angular velocity in the direction opposite to the spin of the flywheel to maintain the constancy of the angular momentum of the system.

In your case, an astronaut can cause the spacecraft to spin by running along the circular perimeter. Again masses (or more precisely moments of inertia) of astronaut and the spacecraft matter and the effects are usually small.

  • 22
    $\begingroup$ That illustration is fantastic. $\endgroup$
    – Jason C
    Commented May 31, 2020 at 13:30
  • 1
    $\begingroup$ The "not useful" part applies in particular to all the gas molecules rattling around in the combustion chamber and the engine bell just before they bounce off a last time and take off to infinity, leaving their sign-reversed momentum with the capsule. $\endgroup$ Commented May 31, 2020 at 17:45
  • 18
    $\begingroup$ The "not useful" illustration should be added to every KSP guide. $\endgroup$
    – JiK
    Commented Jun 1, 2020 at 0:30
  • 1
    $\begingroup$ Sounds like an excellent way to make s'mores during the boring parts of the trip, and keep the crew warm too ! $\endgroup$
    – Criggie
    Commented Jun 2, 2020 at 5:21
  • 1
    $\begingroup$ @JustJohan: Yes. I used Microsoft PowerPoint. $\endgroup$
    – Vishnu
    Commented Dec 29, 2020 at 9:02

An astronaut pushing against the wall of a spacecraft does cause it to move due to Newton's Third Law and conservation of momentum, as you have noted. However, the movement of the spacecraft is not that noticeable for a couple of reasons. One is that the spacecraft has much more mass than the astronaut, so any change in its velocity will be much smaller the the astronaut's velocity change (in proportion to the ratio of their masses). Another is that the astronaut is inside the spacecraft, so if they push off one wall, they will eventually collide with the opposite wall, effectively cancelling out any change in the spacecraft's velocity, as the total momentum of the spacecraft + astronaut cannot change. Lastly, while it does make a difference whether or not both the astronaut and wall are moving or just the astronaut is (as the motion is non-inertial), this difference can be difficult to notice from watching a video, compared with experiencing it in-person.

  • 9
    $\begingroup$ You can even remove a step. In order to collide with the spacecraft wall, they had to push off from their seat, so colliding with the wall just restores the previous neutral state. If they'd pushed off and went out an open hatch, the spacecraft would be moving (slowly, due to its larger mass) in the opposite direction. Compare to jumping off a small boat on Earth. $\endgroup$
    – jamesqf
    Commented May 31, 2020 at 15:37

The astronaut most certainly can cause the craft to spin, although only exceedingly slowly. If the astronaut runs around the inside of his spacecraft (AFIAK realistically only possible in Skylab) the spacecraft will rotate in the other direction to maintain the same total angular momentum. As soon as the astronaut stops running the spacecraft will also stop. As the spacecraft has far more mass the rotation of the spacecraft will be a tiny fraction of how much the astronaut moves.

Note that this is routinely done by mechanical means. Long duration missions orient the spacecraft with reaction wheels--they move a lot faster than the astronaut could but it's the same thing--and it turns the spacecraft only very slowly.

  • $\begingroup$ I noted that during the recent Crew Dragon approach the ISS switched off their "Russian thrusters which are a bit too... er.... dynamic", as the commentator remarked, and relied on "gyros" only. That must be the attitude control system you mention. I was not aware of that. $\endgroup$ Commented Jun 1, 2020 at 9:02
  • $\begingroup$ Interestingly enough, I think that in your Skylab example, we might call the running astronaut an organic reaction wheel! $\endgroup$
    – Cort Ammon
    Commented Jun 2, 2020 at 21:07
  • 1
    $\begingroup$ @Peter-ReinstateMonica Reaction wheels are rather like large gyroscopes, but the purpose is very different. A gyroscope's job in life is to point in a constant direction, they spin constantly in order to accomplish that. The spun mass is as small as possible--taken to the limit it's just light. A reaction wheel has substantial mass and only spins when the craft wants to point somewhere else. $\endgroup$ Commented Jun 3, 2020 at 2:17

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.