Space Shuttle External Tank Foam Insulation


Since the loss of the Space Shuttle Columbia in 2003, we have received many questions about the foam used on the Shuttle's External Tank. These questions have continued as NASA has struggled to eliminate the problem of foam shedding during subsequent return to flight missions. Several of the more common questions are addressed in this article.


The Space Shuttle consists of three major components. The Orbiter carries the crew into space and returns to land on Earth. Two Solid Rocket Boosters provide the majority of the thrust needed to lift the Orbiter into space. Holding these major components together is the External Tank. The primary purpose of this tank is to carry the liquid propellants used by the Orbiter's three main rocket engines. The tank is the only part of the system that is not reused. After the main engines of the Orbiter have been shut down, the External Tank is released to re-enter the atmosphere where it burns up.

Components of the Space Shuttle
Components of the Space Shuttle

The tank is made up of three major sections. At the top is a tank containing liquid oxygen used as an oxidizer in the main engines. At the bottom is a tank for liquid hydrogen burned as fuel by the main engines. In between these two sections is a region called the intertank that contains various instrumentation and processing systems for the propellant tanks.

Sections of the External Tank
Sections of the External Tank

Both oxygen and hydrogen exist as gases at standard temperature and pressure. Since their density in this state is quite low, the amount of these substances required by the Space Shuttle would take up an enormous volume. The only way to carry sufficient propellant in a reasonable amount of space is to increase the density of the gases by cooling and pressurizing them until they become liquids. The liquid oxygen is cryogenically cooled to -300F (-184C) while the liquid hydrogen is chilled to -423F (-253C). These liquids must be kept at high pressure and very low temperature or they will boil back to a gaseous state.

As the Space Shuttle rises through the atmosphere, however, the outer surface of the tank is heated by friction with the external air that rushes past the vehicle at very high speeds. A thermal insulation coating is needed on the outer surface to maintain a constant temperature inside. Without this insulation, the cryogenic fuels would become warmer and start to convert back into a gas, raising the pressure inside the tank and possibly causing it to disintegrate. The thermal insulation is a 1 inch (2.5 cm) thick polyisocyanurate foam sprayed on the outer surface of the tank.

This insulation is also important for reducing the formation of ice on the outside of the tank while it is sitting on the pad prior to launch. Ice has always been one of the greatest concerns of NASA engineers since it would likely fall off the tank during launch and could impact the Orbiter. Large chunks of ice can be very heavy and dense, so any impacts against the heat tiles on the lower surface of the Orbiter could easily damage these critical surfaces.
- answer by Justine Whitman, 17 December 2006


As shown below, the External Tank was indeed painted white on very early Space Shuttle flights. This paint job was only added to make the tank look better and was discontinued after the first two missions. By eliminating the paint, NASA was able to save money and shed about 1,000 lb (455 kg) of weight, translating into improved payload performance.

Liftoff of the first Space Shuttle mission with a painted External Tank
Liftoff of the first Space Shuttle mission with a painted External Tank

To determine whether a coat of paint might prevent foam from shedding requires a better understanding of what causes the foam to fall off in the first place. Most people probably assume that the foam separates from the tank because the flow of air rushing past rips it away. This type of separation is called an adhesive failure because the foam does not properly stick to the bare aluminum surface of the tank. However, NASA research has found that very little, if any, of the foam comes off because of adhesive issues.

The problem is instead caused by a cohesive failure. In other words, the foam itself does not stay together. This cohesive failure is caused by voids created inside the foam as it is sprayed onto the tank. These voids create pockets of air trapped within the foam. Since the voids are formed at sea level, the air inside them is at sea level pressure. As the Shuttle climbs into orbit, the external atmospheric pressure continually decreases. This difference causes the relatively high pressure air trapped inside the foam to expand outward ripping off chunks of foam in the process. Further worsening the problem is the heat generated by friction between the thermal insulation of the tank and the air rushing past it as the Shuttle accelerates to high speeds. This effect heats up the air pockets within the foam raising the pressure and further encouraging the voids to rupture. This phenomenon is often referred to as "pop-corning."

Since the problem lies deep inside the foam rather than at its surface, an outer coating of paint would do little to stop this process. NASA records support this conjecture since the first two missions flown with a painted External Tank suffered just as much foam loss as those with an unpainted tank. Indeed, it is more likely that an extra coating on top of the foam would end up creating an even bigger problem by adding chunks of paint to the debris threatening the Orbiter.
- answer by Jeff Scott, 17 December 2006


NASA has actually studied placing some sort of net or cover over the foam insulation. However, NASA concluded that portions of this covering could also break away posing a debris hazard to the vehicle. The covering material might also be quite dense depending on the material it was made of, so its separation may pose an even greater threat to the Orbiter's thermal protection system than the foam does. The question also implies that some sort of material could be embedded within the foam itself. Perhaps adding high strength fibers or a similar material inside the foam would be a better alternative to an exterior coating but is still not without risk.

Foam from the External Tank breaking away and impacting Columbia
Foam from the External Tank breaking away and impacting Columbia

Whether this netting was an internal or external addition to the thermal coating, any such solution would require a great deal of research, testing, and new manufacturing techniques. Any change to the Shuttle is subject to a very lengthy and rigid certification process to understand what unintended effects it might have on the vehicle. This particular change would require exploring if and how the netting could fail, how much mass it might have, and how it might threaten the Orbiter. NASA has concluded that such a process would take several years and probably would not be completed before the Space Shuttle is retired in 2010. Instead, NASA has focused its efforts on improving the foam application process to minimize voids while also identifying and eliminating other potential debris sources elsewhere on the vehicle.
- answer by Jeff Scott, 17 December 2006


As in the previous question, any modifications to the Shuttle would require a level of analysis, testing, and cost difficult to justify given the small number of flights left. In addition, this idea raises a variety of questions about how it could be implemented. The shield would not only have to be strong enough to survive debris impacts but also to withstand the aerodynamic forces and thermal heating imparted as the Shuttle climbs through the atmosphere at supersonic speeds. The shield would also have to be attached to the Orbiter without interfering with the thermal protection system that covers the Orbiter's underside and leading edges. Furthermore, the shield would have to be released somehow once in orbit without doing any damage to the thermal tiles. Procedures would also have to be developed for astronauts to manually remove the shield during a spacewalk if it failed to separate on its own. When considering issues like these, a disposable debris shield sounds rather complex and would likely be deemed impractical.
- answer by Joe Yoon, 17 December 2006


Read More Articles:






Back Aircraft | Design | Ask Us | Shop | Search Home
About Us | Contact Us | Copyright 1997-2012