Recently, a SpaceX test flight was aborted by blasting the whole device.

Wouldn't it make sense to have parts of such rocket be equipped with parachutes when test flights are performed, so at least parts of the aircraft can be recycled?

I comprehend that the whole monolithic device cannot be rescued, but parts of it could be split and safely brought back to earth separately. Perhaps even the huge tank could be split and saved.

Just some layman thoughts …

  • 10
    $\begingroup$ Extra parachutes and decouplers aren't weightless. The idea is to make the rocket reluable enough so that more money is saved by having it work reliably than to save individual parts. To save rockets with people, they do use an escape system with parachutes. $\endgroup$ Apr 25, 2023 at 15:34
  • 9
    $\begingroup$ this is a partial duplicate of Why was SpaceX's Starship blown up since answers there cover that it was never intended to be recovered, even in a fully successful test. We probably need a canonical question for "parachutes aren't trivial" $\endgroup$
    – Erin Anne
    Apr 25, 2023 at 21:03
  • 8
    $\begingroup$ What are the chances that a big, heavy, out-of-control rocket, even if parachutes could be successfully deployed, would land somewhere safe, away from people/vehicles/structures, let alone on terrain that wouldn't damage it? $\endgroup$
    – gidds
    Apr 25, 2023 at 21:41
  • 7
    $\begingroup$ The expensive parts are generally the ones that are going wrong. These days, guidance electronics are cheap (there's no point in using a \$10,000 recovery system to save a \$1000 computer). The expensive stuff that you'd most like to re-use are the engines, but engine failures are usually what causes the problem in the first place. $\endgroup$
    – Mark
    Apr 25, 2023 at 22:14
  • 5
    $\begingroup$ @ErinAnne: For future reference in case that "Parachutes aren't trivial" question, I'll add that they're difficult to test, since we have difficulty modelling them - so sometimes, the "plan" is "Don't try to make a space-grade parachute - just eject the things you want from it and let them parachute down separately from the launch vehicle.". $\endgroup$ Apr 25, 2023 at 23:11

5 Answers 5


The ability to successfully deploy a parachute still requires some level of stability or control, it's simply not an option for a badly out-of-control rocket. In cases where something goes wrong, the rocket may spin wildly out of control, or even start to disassemble midair. A parachute won't deploy reliably in such a case, and even if it did, it may not stop a rocket that's flipped over and is accelerating toward the ground, and it won't do anything to stop large pieces that fall off. Aborting the mission is a last resort, so you'd only deploy the parachute in cases where something is going wrong anyway, which often means that the rocket is not in stable flight anymore. Making sure the rocket explodes into very tiny pieces is the safest option.

  • 15
    $\begingroup$ Also relevant that even if going to plan much of the flight is either traveling too fast or too high for a basic parachute to do much without being some sort of exotic multi stage thing. $\endgroup$ Apr 25, 2023 at 13:27
  • 2
    $\begingroup$ Most of this answer, sadly, also applies to the question of why having parachutes failed to save Columbia's crew. $\endgroup$
    – dan04
    Apr 25, 2023 at 22:00
  • 5
    $\begingroup$ @stevec It can't. When the Space Shuttle re-enters earth's atmosphere, it uses the heat shield to brake against the atmosphere, Columbia failed at this step which is fatal and non-recoverable. There isn't enough air for parachutes to work, and when there is, the shuttle would hit it at near terminal velocity and be smashed into a million pieces. $\endgroup$
    – Nelson
    Apr 26, 2023 at 7:42
  • 4
    $\begingroup$ @stevec The problem isn't necessarily that the object is very heavy, it's more that it's moving very fast. A big enough parachute could theoretically land a payload of very large mass under controlled conditions, but the parachute will get torn to shreds when it's deployed at supersonic speeds. I found this (somewhat old) article stating the force limit on NASA's parachutes is about 200-some tonnes, not even sufficient to descend a constant rate. (nasa.gov/mission_pages/constellation/ares/…) $\endgroup$ Apr 26, 2023 at 18:16
  • 8
    $\begingroup$ @DarrelHoffman Here's the excerpt from the Columbia disaster on crew survivability: "Once the crew module fell apart, the astronauts were violently exposed to windblast and a possible shock wave, which stripped their suits from their bodies. The crews' remains were exposed to hot gas and molten metal as they fell away from the orbiter." No parachute would've survived this. They were in such a harsh environment that it ripped the space suit off their bodies. $\endgroup$
    – Nelson
    Apr 27, 2023 at 1:05

great question, and it is similar thinking that led SpaceX to attempt to land rockets for re-use. In this case though I don't think it would be safe or practical to attempt recovery of a launch vehicle that has lost control during ascent.

The bulk of the rocket mass is fuel, and early in ascent, most of that fuel is still on board the craft. If the rocket goes out of control, you still have hundreds of thousands or even millions of pounds of liquid oxygen and liquid methane (the fuels used by SpaceX for this launch vehicle) that either need to be burned or vented before you get close to the ground. Per Wikipedia, the first stage of this rocket contains about 7.5 million pounds of propellant, and the second stage contains about 2.5 millions pounds of propellant. Thus you would need a plan to somehow remove the risk of fire/explosion/release of the remaining propellants. Hence controlled combustion/explosion, aka, flight termination system.

Assume for discussion that you had a venting system that could safely release the propellants, and that you could stabilize a tumbling craft to deploy a parachute system. Consider then the size of a parachute needed to return a 600,000 pound structure. Some of the largest parachutes ever made were for the Ares program. These were 150 feet in diameter and weighed 2000 pounds each, and were tested with 40,000-50,000 pound loads. So you'd need roughly an order of magnitude more parachute capacity (eg, 10-15 of these 150 foot diameter chutes) to recover the SpaceX Starship structure.

In addition, for a parachute recovery to be practical, you'd also need a quick response ocean-going support team with equipment capability sufficient to retrieve 600,000+ pounds of structure from the water, since it is doubtful the FAA would allow this kind of contingency plan to occur over land, even if sparsely populated. Additionally, it is unclear to me if the saltwater exposure would render some components unsuitable for reuse, which would further reduce the potential benefit of a parachute recovery.


I think the average person doesn't quite realize how much parachute is needed for even a relatively small piece of machinery. When the military drops a humvee out of an airplane, a vehicle weighing around 7 tons, it takes three parachutes each with a 100-foot diameter, which weigh close to 300 lbs all together, and it's not a very soft landing.

The Space Shuttle Solid Rocket Boosters did land under parachute, but solid rockets are vastly simpler and lighter when empty than liquid-fueled rockets are. An empty SRB weighed about 91 tons (the heaviest objects to ever land under parachute), and the three parachutes to land them weighed over a ton each -- and even then they hit the water at highway speeds. The Starship, at 180 tons dry, would need at least six tons of parachute just to land at that speed, and that's fast enough that it would probably destroy the complicated turbines and plumbing that are of particular value in a liquid fueled rocket. (And that's assuming there's somewhere in the structure where you could attach a parachute so it could deploy without just ripping the rocket in half, which is not a trivial problem.)

So, the short answer is: Rocket engineering is very concerned with minimizing weight, so adding a few tons of cloth that you don't even intend to use except in an emergency just isn't an option. And there isn't a simple or straightforward way to just strap some chutes to an existing rocket as a "just for testing" recovery mode. Chute deployment is a very technically challenging process that requires a huge amount of preparation and study -- and mechanically splitting the rocket into pieces first doesn't make that any simpler.

It's also worth noting that one of the main reasons for "flight termination" systems (as opposed to just letting it crash) is that rockets are often full of extremely poisonous fuels like hydrazine, and the termination system is designed to "unzip" the fuel tank, venting it to the atmosphere as high as possible so that the chemicals can burn immediately, far away from people, or at worst get a nice long drop where they can react with atmospheric oxygen and become relatively safe long before they get close to the ground. While the SuperHeavy booster uses a relatively benign methane/LOX mix, that's still an awful lot of extremely explosive gas, which you really don't want at a crash site while you're trying to perform recovery operations.

Could you drain the tanks without wrecking the ship? Maybe, but it's not as simple as you'd think. Consider a soda can: When they're full, you can fully stand on them, but empty, you can crush one by stepping on it with only a fraction of your weight. Rocket fuel tanks work the same way: when it's fully pressurized, the tank is much stiffer than when it's empty, and that strength is used to keep the rocket from bending. Draining the tank can easily cause the rocket to collapse. Rockets that are meant to return and land (like the Falcon 9) often use a secondary gas like helium to keep the tanks partially pressurized even when "empty", in order to retain that structural strength.

Ultimately, recovering rockets by parachute would have significant technical challenges and require so much parachute that it would ruin the rocket's performance, so nobody tries it. That's why the SpaceX rockets are designed to do a fly-back and land on a pad -- it's actually more efficient to carry all the fuel and equipment to land that way than to try to safely splash down.

  • 2
    $\begingroup$ Rocket engineering is very concerned with minimizing weight, so adding a few tons of cloth that you don't even intend to use except in an emergency just isn't an option. - The overly optimistic counter-argument to that is that you could carry that instead of a payload on test flights. (But the parachutes need to be somewhere that can open into the air, unlike a normal payload, so it's definitely not an easy fix.) The OP said in a comment that they hoped parachutes could at least allow recycling of the materials, if not reuse since damage probably happened before needing to abort. $\endgroup$ Apr 27, 2023 at 13:03
  • $\begingroup$ Good answer. You mention the Space Shuttle Solid Rocket Boosters. It's worth noting that, even though they did have parachutes, they also had self-destruct explosives, which were used on STS-51L. So even when you have parachutes, they're just not practical for all the reasons you list. $\endgroup$ Apr 27, 2023 at 16:33
  • 1
    $\begingroup$ @PeterCordes Overly optimistic is right. That would require engineering the chute deployment sequence itself, plus figuring out how to affix the chutes to the rocket in a way that wouldn't just rip it in half (not trivial if the thing isn't designed for it), plus if it lands in water it'll most likely sink anyway unless you add flotation devices. (The Shuttle SRBs landed tail first to capture an air bubble inside instead of deploying big float, but that only works if you have a big hollow space inside, which is easy for solid rockets but not for liquid.) And then throw away the work after. $\endgroup$ Apr 27, 2023 at 18:16
  • $\begingroup$ @AndrewBreza Well, yes, the chutes are there for standard recovery, not emergencies (where ripping open the SRB is the only way to get it to stop firing in a reasonable time). I wasn't intending to address the actual scenario, just talk about what it would cost to stick a chute to a rocket in general. $\endgroup$ Apr 27, 2023 at 18:18

Before SpaceX started the rocket propelled landing system for Falcon 9 they tried to land the empty boosters with parachutes. It proved very challenging, so they decided to go for rocket landed boosters, which proved easier.

Parachutes work well for relatively small things, but anything much larger than a small aircraft and it gets to be VERY difficult to stop. Adding in the extra weight that most of the time wouldn't matter, and it quickly can be shown to not be a great idea.


Two issues that have not been adequately addressed in the existing answers:

  1. A range safety system should cover all phases of flight, and in particular when the rocket is fully-fueled. While we have used parachutes to recover SRBs, keep in mind that these are used after the propellants have been consumed. Considering that propellants can be in excess of 90% of a launch stack's weight, you are scaling the amount of parachute drag needed by an order of magnitude.

  2. The question seems to suggest that safety systems are only needed during test flights, and perhaps could then be omitted during routine flights. There are plenty of instances of aborts that were needed during routine flights.


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.