Balloon tanks, first used on the MX-774 sounding rocket (the direct, though distant, ancestor of the Atlas family of missiles and launchers), are structurally stabilised by internal gas pressure, allowing the rocket’s skin to carry the entire structural and aerodynamic load of launch and dispensing with the large, heavy stiffening ribs and longerons required by non-pressure-stabilised tanks. This, obviously, makes for a far lighter rocket, allowing a much larger payload to be carried to orbit (or, alternatively, allowing a much greater Δv to be imparted to a given payload). The only significant drawback is that the rocket’s propellant tankage must be kept pressurised at all times to prevent the rocket from collapsing under its own weight, but this requirement is easy to fulfill at any launch site equipped to launch liquid-fuelled rockets of any sort.

Yet, despite the tremendous weight savings, very few rocket designs have ever used balloon tanks; so far as I am aware, the only ones to have flown are the MX-774, the older members of the Atlas family (from the Atlas A missile prototype through the Atlas III space launcher, inclusive), and the Centaur upper stage (fitted to all Atlases since the mid-1960s, and still used today on the Atlas V; also formerly used as an upper stage for the Titan IIIE and Titan IV, proposed for use as an upper stage for the Saturn and Shuttle launchers, and planned for future use on the proposed Vulcan launch vehicle), and, of these, only the Centaur remains in production.

Why are balloon tanks seemingly so unpopular?

  • $\begingroup$ safety, and ease $\endgroup$
    – Topcode
    Apr 5, 2020 at 3:48
  • $\begingroup$ Related (but not duplicate): space.stackexchange.com/q/18332/18085 $\endgroup$ Apr 5, 2020 at 13:19
  • 3
    $\begingroup$ There are many problems. First is vertical integration, then it's vertical integration with fuel. Are you sure you want to be the person who mounts the payload to a rocket full of fuel and with a nontrivial possibility to implode? $\endgroup$ Apr 6, 2020 at 1:00
  • 1
    $\begingroup$ @user3528438: What would be the difficulty of constructing the tank with an exoskeleton which could be largely detached from the craft once it was pressurized for the last time? The launch tower could be designed to support much of the weight of upper stages prior to lift-off, to allow for the possibility of a last-minute scrub and depressurization. $\endgroup$
    – supercat
    Oct 12, 2023 at 17:11

2 Answers 2


Balloon tanks aren't used because they require many precautions to work with. The pressure stabilizes the tanks, so the moment the pressure drops, the tank collapses. Not very convenient to work with. Also, it's costly, because you have to have machines that keep it pressurized. The people working around the tank have to also exercise more precautions, because a pressurized balloon tank can burst and hurt them, while an unpressurized, non-balloon tank can't. Yes, you can put supports inside of the tank that stretch the tank to prevent collapse, but obviously those have to be removed before any serious work is done on the tank.

Also, not to mention that ultra thin panels are difficult and expensive to produce and to weld together. You can see part of the Centaur manufacture on smartereveryday's ULA tour (


  • $\begingroup$ One would either need a machine to keep the tank pressurized, or an exoskeleton to support the weight of the craft outside the tank at any time when the tank might not be pressurized, but using the former until just prior to lift off doesn't seem too hard. $\endgroup$
    – supercat
    Oct 12, 2023 at 17:12

Most of the industry as a whole do not use Balloon tanks because they are hard. The struggles of the Space X Starship is a prime example. Hence, the ULA IP for balloon tanks is a significant advantage.


  • 3
    $\begingroup$ Can you give any information on why it's hard? This answer is not very informative as written. $\endgroup$ Apr 6, 2020 at 15:41
  • 3
    $\begingroup$ The SpaceX Starship does not use balloon tanks. It can support its own weight when not pressurized, although the lower tank when unpressurized cannot support the weight of a full upper tank. $\endgroup$ Jun 24, 2020 at 15:02

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.