After seeing all the Starship failures, having something with a membrane keep things in place seems like an obvious solution. Bladder tanks and "balloons" have been discussed here: Why is an inflatable balloon inside a fuel tank not used to prevent fuel from "sloshing around"?

What other novel solutions, apart from pressurization with another gas, or a bladder tank, have been proposed and tried? For example, have rotating tanks, or other designs such as a tank with a compression arm and membrane, or a heavier liquid or something else like floating pebbles or pouches that can provide enough downward pressure, been tried? Or what about some novel tank design with an intricate coral-like structure inside? What are the advantages and disadvantages of some of these designs?

  • 19
    $\begingroup$ The answer to this question is baffling 😎 $\endgroup$ Oct 15, 2021 at 14:34
  • 2
    $\begingroup$ Try having a semitrailer full of a light liquid like milk or gasoline. Sloshing can throw the truck all over the road or even flip it over. There's a solution, and dare I say it? "not rocket science" $\endgroup$ Oct 15, 2021 at 20:08
  • $\begingroup$ @Harper-ReinstateMonica Not just trucks. Slosh is also an issue with railroad cars and ships at sea. And of course with rockets that use liquid propellants. One of the main jobs of the range safety officer is to be prepared to hit the "big red button" that triggers explosive devices planted next to one of the vehicle's propellant tanks. (It's not really a button but we do call it that; it's typically two toggle switches that must be flipped, both of which have protective covers, so four actions are needed to hit the proverbial BRB.) $\endgroup$ Oct 15, 2021 at 20:21

2 Answers 2


Sloshing is not a problem in zero-g conditions. A key challenge with regard to starting the firing of a thruster in a zero-g situation is to get the propellant flowing to the thruster, without any ullage gas also flowing to the thruster. This is the primary reason some spacecraft use bladdered tanks.

Other spacecraft do not have bladders in their propellant tanks. Without a bladder, the liquid propellant and ullage gas form what one spacecraft engineer described to me as an "ethereal foam". (That's not my term, but I do like it.) Those spacecraft instead use what are called "propellant management devices" (PMDs) to ensure that only propellant flows to the thrusters. Google the term "propellant management device" and you will find multiple books, multiple technical articles, and patents galore.

Sloshing becomes an issue once a vehicle undergoes what should be continuous uniform thrust. The challenge is that the thrust is never perfectly uniform. If the vehicle is in the atmosphere, atmospheric disturbances can make things even more interesting. Baffles and other anti-slosh devices do help, but they add mass. And like bladders, anti-slosh devices add failure modes to the vehicle.

Another solution to the slosh problem is to ensure the spacecraft control system software and hardware do not do things that excite slosh modes. Every spacecraft I've worked on has tried to address the issue of avoiding slosh excitement by the control system. Lots and lots of analysis goes into this.

  • $\begingroup$ Agreed. Virtually all large maneuvers I’ve seen are preceded by a a series of pulses, usually from dedicated attitude control engines, called a “settling burn” to facilitate propellant management. $\endgroup$
    – cms
    Oct 17, 2021 at 1:06

Ways to do propellant tanks, with a view on preventing the propellant from "sloshing about". This sloshing could be either actual sloshing of liquid under gravity/acceleration in a tank, or the need for ullage control in a gravity-less tank.

  1. The naked case: just a spherical/capsule shaped tank containing liquid:
    Pro: easy to make, very light.
    Con: The propellant will indeed slosh around a lot, causing interesting movements of the tank as a whole and causing any outlet to emit an erratic mixture of propellant and gas, which is a bad thing. Even if the output is all liquid, it will be at wildly altering pressures, making life hard for the plumbing downstream. In zero-g, the propellant could be anywhere in the tank, so before use one needs to apply some gravity via an external force, to settle the propellant (ullage)
    This design is usually used by new rocket designers only on their very first attempt, before they go dOH! and revert to the baffled approach below. Some very small rockets can get away with this.

  2. The baffling case: The same tank, but with mechanical baffles on the inside:
    Pro: easy to make, not too much of a mass penalty.
    Con: The mass penalty is nonzero. And baffles not only are useless in zero-g, but could actually somewhat impede ullage during a zero-g restart.
    The propellant will still slosh around, but it will rapidly dissipate its mechanical energy on the baffles, greatly reducing the disturbance on the tank as a whole. The propellant will also tend to stay closer to the "down" side of the tank, as long as there is any gravity. In zero-g, the propellant dissipates all through the tank, and during ullage must find its way past the baffles before setting, increasing the amount of ullage force needed.
    This approach is used by 99%(margin of error 1%) of all current ground-to-orbit launchers.

  3. The bladder case: put your propellant inside a bladder, and the bladder inside a support structure which we will call a tank for convenience (but it does not need to be a sealed tank!)
    Pro: absolutely guarantees your propellant stays contained, and does not allow voids to be ingested by the engine. Even in zero-g. Especially in zero-g!
    Con: The bladder and supports are heavy. Making a bladder capable of containing multi-ton quantities of liquid is difficult. Making a bladder capable of handling cryogenic propellants is problematic. Making a bladder contain a propellant with any significant vapor pressure, especially the huge vapor pressures that cryogenic propellants generate when they start overheating, is impossible (or self-defeating, if you allow a void to form inside the bladder, you are back in the unmoderated case(1) above)
    Bladder fuel systems are very commonly used for the storage of Thruster propellant, where longterm, zero-g storage of relatively small quantities of stuff like Hydrazine is needed.

  4. Gaseous storage: the propellant is stored not as a liquid but as a gas, under pressure.
    Pro: Absolutely guaranteed no sloshing and no voids under any gravity, and it even allows pressure-fed engines.
    Con: Pressurized gas tankage is heavy. Ridiculously so!
    This is normally only used for gasses that will in turn pressurize other systems, such as the Helium COPV's, or for inert-gas storage for electric thrusters.

  5. A piston pushing the fuel. In concept like an refashioned bicycle pump, mechanically pushing your fuel. Pro: Guaranteed no sloshing (because these is no void in the tank), and it is self-pressurizing due to the mechanism.
    Con: This approach is heavy. Even heavier than the high-pressure-gas approach. You need not only the mechanism, its power source, and actuators, but you also need a tank wall that can provide sufficient strength to resist the pressure and mechanical forces of the piston. Additionaly, you get serious complications if the vapor pressure of your propellant is at all significant, as this will force a void into the tank, reverting this design back tot he (1) unmodified capsule tank above.
    Despite all this, this approach is used. By some small and microsats, for their thrusters/RCS, simply because at a small enough scale it is a beautifully simple self-pressurizing system powered by a ridiculously simple electric motor-and-gear system.

Here is where the lateral thinking begins:

  1. Solid rockets: If your propellant is not a liquid, then it is never going to slosh!

  2. Hybrid solid/gas: Same thing goes if your fuel is a solid, and your oxydizer is a gas, for example the rubber+nitrous approach that Scaled Composites used on their SpaceShipOne.

  3. Solid-fuel electrical systems:
    There are electrical thrusters currently in use that use a consumable, teflon rod as their propellant. There is simply nothing to slosh in that setup!


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.