When watching Starlink launches I noticed the MECO/stage separation usually occurs at about 2m30s with the speed of about 8000 km/h and altitude about 65 km. After the separation the first stage with its engines shut reaches apogee of altitude at about 130 km just with its inertia while still having about 7000 km/h at this point, and then it starts to descent.

Couldn't that energy be used to push the second stage a little longer after shutting down the first stage engines, say from those 65 km of attitude by additional 35 seconds till reaching 100 km of altitude (when the booster still has 7500 km/h), and only then proceed with the separation and MVac ignition? I feel a lot of energy is wasted in the process.

With the above values I'm referring to the following video:

Thanks in advance.

  • $\begingroup$ If there was an efficient way to move kinetic energy from stage 1 to stage 2, it would be a great idea. Unfortunately we don't yet know of one. Laser propulsion? Explosive separation? Elastic collision between stages? Railgun? Maybe one day it will be real :) $\endgroup$
    – jpa
    Commented May 13, 2023 at 8:27
  • $\begingroup$ springs, maybe? $\endgroup$
    – njzk2
    Commented May 13, 2023 at 18:33
  • $\begingroup$ @njzk2 Springs apparently have an energy density of 0.1 Wh/kg, or 360 J/kg; similarly Wikipedia. Hydrogen + oxygen carry a whopping 15MJ/kg. Simply carrying more fuel instead of springs would get you thousands of times more energy. This is actually intuitive: How far would a compressed car suspension spring propel your car, compared to a gallon of gasoline? $\endgroup$ Commented May 14, 2023 at 16:45
  • $\begingroup$ @jpa Explosions or thrusters that use the first stage as an opposing surface, creating some kind of ground effect (more force compared to venting the fumes into space) could harvest some of the momentum and energy of the booster stage. Almost like a "gas spring". But Explosions are destructive unless you have a heavier structure, anathema to rockets; and for less violent reactions the time is very limited because the stages separate quickly. $\endgroup$ Commented May 14, 2023 at 16:51
  • $\begingroup$ @Peter-ReinstateMonica I realize that, I was mostly saying that in jest. Thanks for the numbers, thought, it puts things in perspective :) $\endgroup$
    – njzk2
    Commented May 14, 2023 at 16:58

3 Answers 3


After the first stage engines are shut down, the only way for the second stage to extract energy from the first stage is to push away from it, which is how the separation already works. Attempting to push harder (to slow down the first stage more and thus speed up the second stage) ends up being a trade between the mass of the separation mechanism and the mass of additional propellant, and rocket fuel will win nearly any specific power competition you put it in.

Comments on this answer and other answers have adequately covered that simply coasting doesn't help you, but I want to emphasize that at best staying attached to the unignited first stage does the second stage no good. Realistically, it likely carries a number of penalties:

  • Most significantly, for most Falcon 9 missions, the first stage maneuvers for boostback shortly after stage separation. Obviously you don't want to boostback the second stage, so you should detach it after MECO.

  • There's a possible drag penalty from the first stage at non-zero angles of attack.

  • If the stack needs to do any attitude control, this is more difficult with the first stage attached.

  • Added gravity losses from the necessarily-longer time to orbit

  • Oberth effect penalty for not using the second stage engine at the higher kinetic energy it has while it's lower

  • 6
    $\begingroup$ @Endrju But just like your car in neutral does not speed up at all, neither does the rocket with engines off. The engines aren't there just to keep the rocket going, their sole purpose is to make it go faster and faster until it reaches orbital velocity. The moment the first stage shuts down, the rocket stops accelerating (and actually will slowly lose speed to drag, just like your car in neutral), the only way to resume accelerating is to start the second stage ASAP. $\endgroup$
    – TooTea
    Commented May 12, 2023 at 12:54
  • 4
    $\begingroup$ @Endrju if the first stage is shut down, it and the second stage are just coasting until they separate and the second stage starts up. The first stage isn't pushing any more, there's no benefit to keeping them attached. $\endgroup$ Commented May 12, 2023 at 13:18
  • 3
    $\begingroup$ @Endrju we know what you meant. This answer is telling you it's nonsense and explaining why: the stages have to interact for the "wasted" stage 1 energy to be used by stage 2. $\endgroup$
    – Erin Anne
    Commented May 12, 2023 at 17:09
  • 1
    $\begingroup$ @Endrju: If you jump out of a coasting car and come to a much more rapid and violent halt than the car, it's not because you lost the benefit of the car's energy. It's because you lost the benefit of the car's wheels. Wheels aren't an issue for a rocket. $\endgroup$ Commented May 14, 2023 at 2:45
  • 2
    $\begingroup$ ...contrived? mentioning that the first stage has different places to go than stage two is contrived??? what in tarnation $\endgroup$
    – Erin Anne
    Commented May 14, 2023 at 22:35

Look at it a different way. If the two stages separated in a non-violent manner - i.e., one didn't "push" away from the other - then they would continue side by side (or end to end) as gravity acts exactly the same on both of them, slowing them down at exactly the same rate. In that situation, if the Stage 2 engines don't work then Stage 2 and the payload will follow the same path - apogee at 130km, etc. - as Stage 1, assuming Stage 1 doesn't do any recovery burns during that time.

The first stage only adds momentum (and speed) to the second stage if the first stage is itself still accelerating. What that does mean is the first stage can burn longer if it is not trying to return to a landing zone/ship, which is why expendable launches of Falcon 9 (or Falcon Heavy) can either carry more mass to orbit or achieve a higher orbit than the equivalent recoverable launches, even though the Stage 2 (fuel and burn time) may be exactly the same.


First, getting to orbit is about going sideways fast, not going high.

Second, if the upper stage didn't ignite, it would also coast to the same altitude, and probably lose slightly less speed due to ballistic coefficient.

Third, losing speed will almost always result in an increase in the total gravity drag.

  • $\begingroup$ Well, at the end of the day getting to space is very much about going high. Because space is there. How you get high on a restricted budget requires some skills though, one of which is dodging gravity ;-). $\endgroup$ Commented May 13, 2023 at 5:19
  • 1
    $\begingroup$ @Peter-ReinstateMonica I think JCRM is paraphrasing this explanation by Randall Monroe, or at least making the same point: space isn't actually that far away, so getting a bit further doesn't help you very much unless you're happy to come straight back down; to achieve orbit, you need to get faster. "Gaining altitude but slowing down a bit" sounds like it would be useful, but is actually defeating the main aim of the rocket, which is to accelerate. $\endgroup$
    – IMSoP
    Commented May 14, 2023 at 8:31
  • 1
    $\begingroup$ @IMSoP Well, if they paraphrased Munroe, they missed the key point: The distinction between getting and staying. Besides, the last part of the answer was not better: I cannot make sense of "Losing speed will almost always result in an increase in gravity drag". The "drag" (gravitational force?) at a given point is obviously independent of the speed. The "almost always" makes me think they were not quite sure and inserted this hedge. The sentence also does not end with a period -- did JCRM want to say more and hit enter accidentally? $\endgroup$ Commented May 14, 2023 at 9:45

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.