I just read this question and answer, which seems to confirm my fears that an incendiary bullet will still destroy you if it hits your fuel. In other words, a self-sealing fuel tank will not protect against incendiary bullets. The wing will be punctured, and the fuel will either explode, or catch fire and remain on fire. (Correct me if I'm wrong.)

So the question is, why have them at all? It seems so easy to defeat the system by just using incendiary bullets, which must be a whole lot cheaper than the rubbery fuel tank.

I know there is accident protection and flak fragmentation protection. However, if you're taking flak, it seems to me you've got many problem-areas under the aircraft to worry about. I remember reading about a lot of bomber pilots in WW2, who took hits in the legs from flak, and some of them started sitting on their flak jackets all the time.

I realize this is a complex question, because cost-benefit analysis is a complex thing. But maybe someone can explain the rationale held at the time. Self-sealing fuel tanks added cost and weight, yet seem easily defeatable with incendiary bullets. Maybe I am missing something?


3 Answers 3


Your description of a self-sealing fuel tank remaining on fire when lit is the exact situation of the sealing mechanism failing. As soon as the tank is punctured, regardless of a fire or not, the rubber in the middle layer will react with the fuel and swell until the hole is reasonably obstructed. This both extinguishes the fire and seals the leak. Consider a case study: in the Pacific theater, Japanese fighter designs did without self sealing tanks for the cost and weight saving benefits. American naval designs such as the Hellcat however, had every possible damage-control system of the day installed everywhere possible to make the planes as rugged as they could. In the end, Japan mainly lost control of their own airspace due to the fact that they simply ran out of materials to build new planes from. The rate at which they lost planes was incredibly high due to both the small .50 caliber rounds' ability to set the unprotected tanks on fire, and the poor training of Japanese pilots towards the end of the war. If you look at statistics, per battle the IJAAF and the IJN lost more planes per squadron and battle group than any other country's air service in the war. In fact, many of the IJAAF's main fighters including the Mistubishi A6M Zero and the Kawasaki Ki-61 Hien were infamous among pilots of both sides of their ability to catch on fire from what should have been inconsequential hits and particularly their inability to stop burning. Another point of interest is the design of the self sealing fuel tank:

This shows the layering of the rubber tank. Much of the material going into the design is specifically for structure and flame-retardation.

This all goes to prove strongly that NO, self sealing fuel tanks are a definite must in combat airplanes, and a very important safety feature in other aircraft types. I'm confused however as to why your question switches from bullets (incendiary?) to flak as anti-aircraft guns such as the Flakvierling88 were not meant to ignite fuel tanks, they were supposed to destroy the whole plane, fuel tanks and all. If you want some more numerical data backing up my position, Wikipedia references many sites that give a bit more data on why self sealing tanks are a must for fighter planes. The Defense Systems Information Analysis Center also provides some interesting information on the design and how they weren't all too much of a cost/weight penalty later in the war. Hope this answers most of what you're asking.

In addition, there is a specific part of the article of my second link I would like to draw attention to:

This simple composition and design, which are inexpensive, lightweight, and easily produced in large volumes, confronts the deceptively complex ballistic dynamics of a bullet perforating a fuel tank. When a normally oriented bullet pierces a fuel tank, it leaves a small residual hole less than the diameter of the projectile. But this small hole is just the beginning of the challenge. The bullet, travelling at 3,000 ft/s, begins to unload its tremendous kinetic energy on the fuel inside. This unloading generates a high-velocity wave of pressure in the fuel that is reflected back on the wall milliseconds after the bullet tears through. This pressure wave, known as hydrodynamic ram, blasts a jet of fuel back through the entrance wound. The bullet, still moving at a high velocity, and now followed by the same pressure wave, begins to tumble as it travels through the fuel. When it reaches the back wall of the tank, the bullet erupts sideways through the material and leaves an elongated gash that is often torn further by the trailing wake of hydrodynamic ram.

Initial fuel tank prototypes tested by Navy engineers were discouraging failures. Engineers had not anticipated the challenging dynamics, and often the entire back walls of prototype tanks were blown out by the hydrodynamic pressure wave. However, these early development tests shed some light on the mechanisms that brought down so many pilots and aircraft during the World War I. The Navy and rubber company engineers persisted and established remarkably effective design features and manufacturing processes that ultimately saved untold lives not just in World War II but in each of the armed conflicts since then.

This statement strongly backs up the cost/weight penalty not being an issue position, if this clears anything up.

  • $\begingroup$ I brought up flak because the shrapnel could pierce a wing with a small hole, so self-sealing would obviously fix that. I'm still not 100% clear on incendiary bullets, tho. An inc. bullet that pierces into the tank is going to ignite something, regardless of what kind of tank it is. I noticed the wiki article said ...were able to withstand .50 in (12.7 mm) bullets and, on occasion, 20 mm (0.79 in) cannon shells. but it didn't say if those were inc. bullets. I guess I should have specified something about size. Obviously there is a limit to how big a hole a self-sealing tank could seal. $\endgroup$
    – DrZ214
    Mar 4, 2018 at 4:50
  • 11
    $\begingroup$ Keep in mind that you need three things to start a fire: an ignition source, fuel, and oxygen. An incendiary bullet provides the ignition source in the form of heat, and the aircraft provides the fuel, but you won't get a fire unless there is also adequate oxygen in the tank. Self-sealing tanks will reduce or eliminate external oxygen sources by sealing the entry holes, so you only have to worry if there is oxygen already present in the tank/bladder. Even if there is, it will be consumed relatively quickly and the fire will burn itself out -possibly before further damage is incurred. $\endgroup$
    – MikeB
    Mar 4, 2018 at 18:32
  • 1
    $\begingroup$ Yes but keep in mind, 20mm shells had a purpose more akin to flak than a smaller mg caliber round. German shells especially were designed to rip apart the structure i.e. the minengeschoss 20mm HE round. $\endgroup$
    – Jihyun
    Mar 4, 2018 at 19:08
  • $\begingroup$ Actually, Japan had a fair number of aircraft at war's end. It was pilots that they ran out of - takes time to train one. During the Battle of Leyte Gulf, their four remaining carriers were used as decoys, because they didn't have enough trained carrier pilots to outfit the carriers with an attack group. $\endgroup$
    – tj1000
    Mar 5, 2019 at 5:26

An incendiary bullet penetrating a self sealing tank won't necessarily cause a fire. Gasoline needs oxygen to burn. The Allies developed a self sealing bladder that not only plugged holes, but gradually collapsed as the fuel was used up, to suppress the mix of air and gasoline vapor that replaces the liquid fuel as the fuel is consumed and the level drops... in an unsealed tank.

Why is that important? A gasoline vapor/air mix in an enclosed space doesn't burn... it explodes with great force. In the case of a partially empty aircraft fuel tank, enough force to blow the plane apart.

This is why the Japanese aircraft in WW2, that didn't use self sealing tanks, often exploded when hit. A half empty unsealed tank could blow up from a single incendiary hit, if the vapor/oxygen mix was conducive to an explosion (not always, but often). Allied aircraft, that took measures to suppress the creation of gas/air vapor in an enclosed fuel tank, weren't nearly as prone to blowing up with only a few bullet hits.

Even if a vapor explosion didn't happen, and it didn't always happen, the fuel spraying out from an unsealed tank that had been hit could be ignited by anything... the next bullet, electric sparks from damaged wiring, even the flames coming out of the aircraft's engine exhaust... doesn't take a lot to ignite highly volatile gasoline vapor.

Finally, consider what Japan did with its later aircraft designs.

Towards the end of the war, Japan brought out some new fighters. The most capable, and an equal of the better US fighters, was the Kawanishi N1K2 Shiden Kai, which featured 4 20mm cannon, armor plate and bulletproof glass for the pilot, and... self sealing fuel tanks.

Late in the war, it appears that even Japan, the one nation who did not use self sealing fuel tanks on their aircraft, changed their mind.

Additional notes: two somewhat recent air disasters illustrate how the problems of WW2 combat aircraft are still with us.

The hydro shock problem of a bullet hitting a full fuel tank and creating a shock wave was revisited in the Air France Concorde crash. A piece of a tire flew up and hit the Concorde's fuel tank. It didn't penetrate the tank, but the shock wave caused the tank to break open and leak a large amount of fuel, that caught fire from the Concorde's afterburners.

After this crash, the Concordes were given new fuel tanks with a bladder similar to the self sealing bladders of WW2, to keep the tank sealed in the event of hydro shock from an impact, and prevent this from happening again.

TWA flight 800, a 747 that crashed off of Long Island, is believed to have been brought down by an explosion of its center fuel tank. The center tank was nearly empty, and had an air conditioning unit next to it that had been running for some time while the plane was on the ground, heating up the fuel vapor/air mixture. While this was kerosene and not the more volatile gasoline, the nearly empty tank exploded not long after the airplane departed the airport, possibly due to an electrical fault plus the high temperature of the vapor mix.

After this crash, commercial airliners were required to vent inert gas into the tanks rather than atmosphere. Venting with inert gas is also required procedure for oil tankers, after a few of them exploded when a spark ignited the oil vapor in nearly empty oil tanks.

  • 3
    $\begingroup$ Misread that as "420mm cannon" and was mightily impressed for a moment. :-) $\endgroup$ Mar 4, 2018 at 18:49
  • $\begingroup$ P.S., do you have a source for this: Japan, the one nation who did not use self sealing fuel tanks on their aircraft, changed their mind. I've never heard of Soviet aircraft from WW2 using them, but then, I've hardly read much about the R&D of Soviet WW2 aircraft. If they did use self-sealing fuel tanks, I want to read about it. $\endgroup$
    – DrZ214
    Mar 5, 2018 at 0:30
  • 2
    $\begingroup$ @DrZ214 Self-sealing tanks don't all shrink. The self-sealing functionality is orthogonal to bladderized tanks, though it's easier to make a bladderized tank self-sealing since it must be rubber anyway. $\endgroup$ Mar 5, 2018 at 4:02
  • 3
    $\begingroup$ @DrZ214 The self-sealing part is important, even without the tank being bladderized. If the tank is self-sealing, even if there is air in the tank, any ignited fuel will quickly exhaust the available oxygen, which will extinguish the fire due to lack of oxygen. The exception, of course, is if the fuel vapor/air mixture actually explodes... preventing that is where the bladderized tank comes into play. $\endgroup$
    – reirab
    Mar 5, 2018 at 17:07

Based on DoD Ballistic Research Laboratories document Effectiveness of Incendiary Ammunition Against Aircraft Fuel Tanks, especially page 10:

It should be noted that both the fracture of the nose and the tumbling of incendiary projectiles influence the self-sealing property of tanks.

However, so-called "plugging" of the tank, where a piece of the tank is actually removed, will usually result in its defeat. An incendiary bullet may assist this "plugging" action by burning of rubber and by virtue of the break-up of the nose, resulting in a penetration by a jagged blunt nose.

So it seems that a hit by incendiary bullet is worse than a hit by a normal bullet, but there is still possibility that the tank will be able to self-extinguish. After all there will not be that much oxygen inside the tank, so the fire will eventually run out of it if the self-sealing works. There is also the option of filling the tanks with nitrogen or other non-combustible gas, apparently used at least on Russian fighters.

On page 61, on fuel tank fires, it is noted:

  1. If the opening caused by the round is small, a low-order explosion may occur by the expansion of the resulting gases. The fire may extinguish itself due to lack of oxygen and no serious damage to the structure will result.

  2. If the opening is large enough, air can enter to support combustion and a fire may result which may cause a kill.

  • 4
    $\begingroup$ Inerted Fuel Tanks (i.e. filling them with nitrogren/other) have I think been standard in military applications for years (everyone, not just the Russians), and more recently have been mandated in commercial applications. I was surprised to read that they only keep the oxygen content down to 10% or so, but I suppose that implies that fires don't readily start without an abundance, suggesting igniting the tanks is harder than the OP might be presuming. (Inert filler also has the benefit of keeping the fuel under pressure) $\endgroup$ Mar 4, 2018 at 8:24
  • 8
    $\begingroup$ Gasoline is surprisingly hard to ignite. Gasoline vapor is quite easy. That's why we have fuel injectors designed to atomize the fuel, i.e. turning it to vapor as much as possible. Heck, even in a typical car engine you have unburned fuel - from inside a literal explosion! $\endgroup$ Mar 4, 2018 at 12:39

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .