There are several significant advantages to air launching a satellite-carrying rocket:

  • save fuel/propellant as a horizontal take-off is more energy efficient
  • fully mobile platform allowing the launch to be conducted anywhere
  • less affected by weather allowing it to be deployed almost any time

What are the reasons preventing all such rockets from being launched from the air?

  • 6
    $\begingroup$ Nothing explains this misconception better than this XKCD What-if what-if.xkcd.com/58 Yay Space.... Dang $\endgroup$ Oct 8, 2014 at 9:05
  • 3
    $\begingroup$ QO, a very simple answer is: "Rockets are tremendously too heavy to do that." Secondly, the speed of a plane is nothing, compared to the speed rockets need. Also, it's worth noting that "wiki" page is basically crap :) ("Wiki" is a great thing, and wikis are also awesome for shopping lists etc, but it's well worth bearing in mind "wiki" pages are quite often nonsense.) $\endgroup$
    – Fattie
    Oct 8, 2014 at 9:17
  • 3
    $\begingroup$ @MikeyMouse It's true achieving orbital speed is the biggest hurdle. But the 1.5 km/s needed to get above the atmosphere isn't trivial. The energy needed for ascent is often just portrayed as the difference in potential energy. This is wrong since gravity loss is a major expense. $\endgroup$
    – HopDavid
    Jun 5, 2015 at 22:21
  • $\begingroup$ @MikeyMouse Munroe's answer to Brian is questionable. If the sole propellent source were earth, Randall's argument would be valid - a 16 km/s delta V budget is indeed impractical. But there are other possible sources of propellent besides earth. $\endgroup$
    – HopDavid
    Jun 5, 2015 at 22:26

5 Answers 5


This post highlights some misconceptions, so let's do this with big letters.

Launch is about going fast, not high

A helium balloon will get you to the edge of space. It takes a large and expensive rocket to get into orbit. Ok, one at least 20 m long.

Very very fast

"But the plane can give you 600 mph" you say. Well, that's nice, we have 268 m/s of the required 7 km/s. Only 6,700 m/s to go.

Actually, what I mean is attaining a high KE

1 kg going 268 m/s has $0.5*1*268^2=36$ kJ of KE. That leaves $0.5*1*(7,000-268)^2 = 22.7$ MJ to go; all that plane stuff has achieved about 7.5% of the needed energy transfer. [Edited with relative numbers, thanks @NPSF3000]

How about a nuclear explosion in a barrel, then?

Now you're talking. Like a nuclear pop gun blowing a disc of steel up to 70,000 m/s. Kinda. There's a whole wikipedia article on space guns.

Save some money, use a rocket you have sitting around at home

The way to really cheap launches is to use one of those spare enormous orbital vehicles you have lying about the place. Particularly the really big ones.

  • 7
    $\begingroup$ Maybe it's the terminology that is failing people. We talk about getting to orbit as though it was a place, and people assume it's like a place up high. "Accelerating to orbital speed" is less concise but maybe less confusing. $\endgroup$
    – Phil H
    Oct 7, 2014 at 8:54
  • 1
    $\begingroup$ Succint, entertaining, and informative. $\endgroup$
    – Sammitch
    Oct 7, 2014 at 20:04
  • 1
    $\begingroup$ @Baldrickk as a rough rule of thumb, 10% of the energy is spent on going up through the atmosphere and 90% on going sideways. If you could launch from the edge of space, you'd need those 10% of savings to outweigh the added complexity, costs and risks of getting the launch vehicle there. $\endgroup$
    – Peteris
    Oct 7, 2014 at 20:22
  • 1
    $\begingroup$ Correct me if I'm wrong, but doesn't that 300m/s obtained from the aircraft mean that the rocket only has to accelerate to 6700m/s instead of 7000m/s? This, thanks to the sqr means that roughly 8.5%* of the energy needed to power the rocket is saved, which in turns means the rocket can be lighter and save more? *(1- (0.5∗6,700^2) /( 0.5∗7,000^2)) $\endgroup$
    – NPSF3000
    Oct 8, 2014 at 5:37
  • 2
    $\begingroup$ @NPSF3000 I think the last 300 m/s would be 8.5% of the energy required, but the plane would only help with the first 300 m/s, not the last 300 m/s - the first 300 m/s are the easiest, whereas the last 300m/s would be the hardest to achieve $\endgroup$ Aug 5, 2015 at 13:40

Here are a few disadvantages to air launch:

  • Most launch vehicles are too heavy to be carried by any extant aircraft, e.g. Atlas V 401 masses 335 tonnes, compare to the Airbus A380's maximum load of 89 tonnes.

  • For launch vehicles with cryogenic propellants, loading and topoff would be extremely challenging.

  • It is generally not safe for personnel to be closer than a couple of miles to a rocket launch; that would be a problem for the aircraft's crew.

That said, the principle isn't impossible for smaller vehicles, as demonstrated by the Orbital Sciences Pegasus and the proposed Stratolaunch.

  • 6
    $\begingroup$ @Anentropic no, pretty much all of the mass of a ground-based launcher is for getting the required orbital velocity, and getting off the ground happens as a byproduct of that. If you somehow had a launchpad 40.000 feet high, then you'd get some savings in simplicity described above, but you would still need almost the same size multi-stage rocket to get it to orbit. $\endgroup$
    – Peteris
    Oct 6, 2014 at 11:46
  • 2
    $\begingroup$ @Anentropic As they say "getting to space is easy, staying in space is hard". To stay in orbit you need to accelerate to about 8km/s. Thats really really fast and makes getting there seem easy $\endgroup$ Oct 6, 2014 at 15:08
  • 4
    $\begingroup$ I believe, although wasn't able to find a citation on a quick search, that despite all the paper cost savings Pegasus's cost/kg to orbit was significantly higher than hitching a ride as a secondary payload on larger competing platforms, with the result that OSC's 3rd party launches were almost entirely limited to satellites in unconventional orbits which were prohibitive to reach from a standard launch profile. $\endgroup$ Oct 6, 2014 at 19:35
  • 1
    $\begingroup$ Another airplane-launched rocket that everyone seems to forget is the ASM-135, which was launched from an F-15 fighter flying straight up. Although according to Wikipedia, it's max speed was " > 24,000 kmh" (6.6 km/s), so it might not have been quite capable of getting into orbit. Then again, it's probably classified. $\endgroup$
    – Kirkaiya
    Oct 6, 2014 at 22:53
  • 2
    $\begingroup$ The AN-225 can carry 253T; still short but at least it's a significant fraction of the weight. $\endgroup$
    – Antzi
    Jun 30, 2017 at 7:49

Air launch does not actually provide that much benefit.

The benefits are basically, that starting at 40 or 50,000 feet allows the following:

  • The nozzle can be closer to vacuum optimized on the first stage which is good.
  • The launcher can fly to the equator, for a 0 degree inclination launch which is useful.

But as noted, the mass limits on the booster are huge, and rockets are much bigger than any aircraft in existence can lift. Look at the size of Stratolaunch's proposed airplane. It will literally be the biggest airplane in existence and can only lift a smaller than Falcon 9 booster due to mass limits.

The reality is that launching from the ground gets the rocket up to the 50,000 foot mark within the first minute or two. The issue with space flight is not getting high enough, rather it is going fast enough to get into orbit. So rockets typically fly mostly straight up to get out of the 'optimum' amount of atmosphere before turning to accelerate into an orbital path.

Thus airlaunch seems like a good idea, but its limits and minor benefits generally do not pay.

Pegasus was using a fairly powerful launch aircraft (L-1011) but only had a miniscule payload. Stratolauncher will use a huge launch aircraft, possibly at the limits of how big one can be, and will still have a fairly small payload.


As noted on the wikipedia page on air launch to orbit, This article has multiple issues.

One issue is that it compares a vehicle launched from an aircraft versus a similar vehicle launched vertically from the ground. That's not a good comparison. A better comparison would be to look at the aircraft that carries the rocket as a substitute for the first stage. In this comparison, air launch falls far short. A decent first stage will bring the rocket close to the top of the sensible atmosphere and provide a reasonable fraction of the total delta V. A good target is 50 to 80 kilometers altitude and 2 to 3 km/second velocity. Most air launch proposals involve a large subsonic aircraft that might reach the top of the troposphere, or 9 to 15 kilometers altitude and a velocity of 200 to 300 meters/second. Air launch makes for a very lousy first stage.

Another issue is limitations that the air launch places on the rocket. For ground-based launches, various economies of scale and the cube-square law in general mean that bigger is better. Cost per kilogram of payload to orbit decreases as launch vehicle size increases. There is a point at which diseconomies of scale and nonlinearities (e.g., pogo) start to kick in, meaning that there is some point after which bigger is not better, but that point is well beyond the capability of any planned or envisioned air launch system.


Air Launch has its limitations listed in previous answers but also have a number of benefits. The delta V gain is not the most important. More important may be the air launch immunity to moderate weather related problems. You know how often the space launches delay because “shear winds” constrains. The other may be most important benefit is that the air launch is less problematic with range safety.

The idea simply to hang a large liquid propellant rocket under the super-heavy aircraft encountered unexpected difficulties. That scraped the joint project of Stratolaunch and SpaceX in the end of 2012.

The vertically launched rockets dominate the market also by historic reasons. The modern launch vehicles are directly descended from Intercontinental Ballistic Missiles i.e. Atlas, Titan 2 and R7. Some decommissioned missiles are used for satellite launches (Satan, MX).

The evolution of space transportation systems was slowed by multiple reasons:

  • $\begingroup$ Another benefit is that you can launch from virtually any location that best suits your payload. The launch aircraft can fly to the equator if that's important. $\endgroup$ Feb 17, 2018 at 22:39

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.