Special Feature: The F-117A Crash

Last Updated: 12/21/97 PM

Final Update

This will probably be my final update to this article. The cause of the crash has been determined. The fault did not lie with any composite components, and the cause was actually known within two weeks of the crash.

Introduction

On Sunday September 14, an F-117A Stealth fighter crashed in Maryland during an air show. The F-117A derives it stealth properties in large part from composite materials. Although it is too early to tell what caused the crash, the unique properties of composites will have an effect on the clean-up efforts and the investigation.

Over the next several weeks, I will follow the crash investigation, placing an emphasis on the role of composites. I will update this page as often as daily, depending on how much and how often information becomes available.

Some of what I write will be speculation based on my experience as a composites and aerospace engineer. I should note now that I have never worked on the Stealth program, so I won't be providing any inside information.

Crash Cause Discovered (12/21/97)

The Baltimore Sun reported that the Air Force has discovered the cause of the F-117A crash. The Sun covers the results in a series of three articles (thanks to Leon Chuck of the University of Dayton for pointing these out):

(These articles are no longer available online.)

An Amarillo Globe-News article gives a little more detail on the inspection process.

The wing design was modified several years ago to include an additional stiffening plate. During maintenance in January 1996, the wing was disassembled. When the wing was reassembled, four of five fasteners were not reinstalled. These fasteners were hidden by the stiffening plate, so standard visual inspections didn't turn up the problem.

Lockheed engineers had reached a preliminary conclusion on the cause of the accident only two weeks after the crash. The fasteners are installed through a U-shaped aluminum beam, and the beam is painted after fastener installation. As soon as the part was found, it was obvious the fasteners were missing.

The paint job confirmed the conclusion. Because the beam is painted after fastener installation, there should be a ring of bare metal near the fastener holes, where the washers and bolt head should be. Investigators found, however, that the paint extended all the way to the edge of the fastener holes. This verified that the fasteners had not been installed, and also allowed the review team to pinpoint when the oversight occured.

The findings seem to be consistent with the Aviation Week report of a loose elevon. According to theWashington Post, the stiffening plate was originally installed to correct a flutter problem in the elevons. However, the Sun articles seem to imply that the stiffening plate was obscuring the problem in another component. A press conference is supposed to be held soon, so these minor inconsistencies may be cleared up.

The Air Force issued a very brief press release on December 15, announcing that the cause of the crash had been found, but giving very few details.

Earlier reports confirm that the cause was known almost immediately after the crash. As early as October 4, newspapers were reporting that a "significant defect" in the wing support structure had probably caused the crash (see also the Air Force Statement discussed below). Details of the defect, however, were not announced until the investigation was completed.

Mishap Risk Control Guidelines (10/30/97)

In my original posting of this article, I presented some information about the Advanced Composite Mishap Recovery Program run by the Advanced Composites Program Office (ACPO). At that time, the Web page was more than a year out of date, and the mishap documents were not publicly available.

But now ACPO has updated the mishap Web page (as of 10/20/97). A new contact is listed, and the Mishap Risk Control Guidelines are now available in MS Word format (the document is actually in RTF format, even though it has a .doc extension).

The document states that personnel injuries and equipment damage does occur without adequate protection. In fact, illness and death after exposure to burning composites has been reported. Furthermore, many of the health effects of exposure to burning composites are unknown.

Hazards are divided into several broad categories:

1. Organic Compound/Matrix Hazards (resins, adhesives, residual solvents, organic fibers)
2. Smoke and Fume Hazards
3. Reinforcement Hazards
4. Specific Carbon Fiber Hazards
5. Electrical Hazards to Equipment
6. Hazards to the Environment

Unfortunately, there is no consistent set of guidelines for dealing with mishaps involving composite materials. The ACPO report provides some general guidelines, but it is intended to be used with other documents. Universal guidelines are not possible, because of the widely varying conditions under which accidents can occur, but more research will still be useful.

The report contains three checklists for response, containment, and clean-up actions. The first three items on the respons echecklist are emphasized:

1. Minimize unnecessary personnel; only firefighters with Self-Contained Breathing Apparatus (SCBA) should be in the immediate vicinity.
2. All unprotected personnel should avoid the downwind area of the crash site.
3. Areas affected by dense smoke should be evacuated, including the removal of mobile, critical equipment.

The guidelines for containment are actually fairly specific.For example, generic acrylic floor wax, mixed at a 10:1 water-to-wax ratio is recommended as a "hold-down" solution. Appendix A gives even more detailed instructions for mixing and applying this solution.

Finally, like any good document, this one gives a long list of recommendations for additional research. The bibliograpy is also several pages long, and many of the documents should be publicly available.

The Mishap Guidelines make interesting reading. They should be of use to composite manufacturers as well as mishap response teams. After all, who hasn't had an oven fire?

Structural Defect (10/14/97)

An Air Force Statement says that "Physical evidence found in the crash debris revealed a significant defect in a support structure in the left wing of the accident aircraft." No further details on the type of defect are given.

All remaining F-117As are being inspected for this defect and others. 33 aircraft have been inspected so far and no defects have been found. F-117A flights resumed on October 2. Only those aircraft which have passed the inspection are flying.

Cause of Crash (09/26/97)

According to the September 22, 1997 issue of Aviation Week and Space Technology, the cause of the crash was excessive elevon vibration leading to failure of the wing.

Videotape shows the elevon going through several rapid oscillations just prior to the crash. This caused the left outer wing panel to fall off. The aircraft then pitched to a high angle of attack and was unable to recover.

Because the flutter occured on only one side, it is unlikely that the vibration was caused by the flight control system. Rather, a former Skunk Works engineer speculates that the problem was a loose elevon.

Aerodynamic stability is a structural problem, but it is not unique to composites. As discussed below, damage to the composites may make the crash investigation more difficult, but it does not appear they contributed directly to the cause of the crash.

Current Articles

Following are some links to articles related to the crash. Rather than linking to everything on the Internet, I have tried to link to articles which mention composites. These links will probably change rapidly; I will try to keep them updated, but some may become invalid before I can recheck them.

• Washington Post

Composites

If you're already familiar with composites, you can safely skip to the next section.

A significant portion of the F-117A structure is made from composite materials, so composites are sure to play a role in the crash investigation. But just what are composites?

In the broadest sense, a composite material is a combination of two or more materials which behave as one material. The class of composites used on the Stealth fighter are called reinforced plastics. These materials consist of a fiber (carbon, glass, Kevlar, etc.) embedded in a plastic matrix.

The exact composites used on the F-117A are classified, but it's possible to make some guesses. Many of the articles refer to carbon fibers and carbon composites. Furthermore, other synthetic fibers such as glass and Kevlar don't have the properties necessary for an aircraft structure. Metallic fibers, such as boron, aren't stealthy. Finally, carbon composites are black--the same color as the F-117A.

The plastic resin is more difficult to identify, but also isn't as important as the fiber. The plastic is either a thermoset, meaning it burns at elevated temperatures, or a thermoplastic, meaning it melts before it burns. At the temperatures at which jet fuel burns, either material will burn.

A composite structure is made forming fibers over a tool shaped like the final structure, filling the fiber form with a liquid resin (similar to the two-part epoxies found in any hardware store), then curing the resin under heat to harden the material. In effect, the material is made at the same time as the part.

The two most important things about the Stealth composites are: they contain small diameter fibers which won't burn at the crash temperatures; and these fibers are held together with a plastic that will burn at the crash temperatures. Some other composite properties will play a role in the investigation, but these are the important ones for now.

For more information about composites in general, see my previous feature entitled What's a Composite?, or check my list of links to Educational Materials.

Material Hazards

Several sources reported that firefighters and others near the crash scene became ill from fumes emitted by the fire. These sources also reported that a wax-like material was sprayed on the fire to contain the fibers.

The health hazards of uncured composites are fairly well known. The OSHA Technical Manual has an entire chapter devoted to the Health Hazards of Polymer Matrix Materials. (In addition to health information, this chapter is also a good general introduction to composites.)

Most of the hazards of uncured composites come from the resins. The fibers are generally too large to be inhaled. They will make you itch like crazy--and carbon is much worse than fiberglass--but that is only a temporary situation.

The resins pose the worst hazards and must be handled carefully. Depending on the exact system, the resins may give off fumes which are toxic (some contain the same solvents used in model aircraft glues, for example). All resins are also skin sensitizers, which means you will eventually become allergic to them if you handle them. The OSHA manual describes these hazards in great detail for many different resin systems.

Once a resin is cured, however, it is as harmless as any other plastic. The fibers are also firmly embedded in the resin, so composites can be handled without any itchy side effects.

Problems arise, however, when the composite is damaged. As in the uncured composites, the hazards of the cured composite are different for both the fibers and the resin.

As mentioned above, the resins will burn. As they do so, they emit toxic fumes. Many deaths in fires are caused by the fumes given off by burning plastics. At least some of the fumes reported in the press would have been caused by burning of the resin in the composite material.

The fiber hazards are actually worse in the cured damage condition than in the uncured condition. The fibers themselves won't burn, but impact damage from the crash will crack the fibers and release small particles and dust. These particles are very light and, with the heat generated by the fire, will easily become airborne.

The OSHA manual states that dust generated by machining is small enough to be respirable. This dust is also extremely irritating--I have been in shops with good ventilation and still walked away with a sore throat, scratchy eyes, and itchy skin.

Finally, although carbon fibers do not generally burn in fires, this is really dependent on the type of fiber. As the fibers heat up, they may oxidize and lose weight, in effect making the fiber particles smaller and increasing the chance that they may be respirable.

The larger fiber particles are irritating as in the uncured condition. Furthermore, fibers which are still encased in resin are extremely stiff and sharp. It is very easy to get a graphite splinter (composite workers call this a "snakebite") just by handling damaged composites. These splinters are much thinner than wood splinters and thus more difficult to remove. They also do not eventually rot out like wood.

Hazard Preparation

The unique hazards of composites in a plane crash have actually been studied by the aerospace industry and the military. The earliest study I was able to find is Assessment of Carbon Fiber Electrical Effects, NASA-CP-2119, December 4-5, 1979.

Graphite fibers are electrically conductive. When the dust or fiber particles come in contact with electrical equipment, they can cause shorts or other problems. This report covers not only the electrical hazards, but also the methods in which carbon fibers can be released. As the report states:

...a burning aircraft containing carbon composites releases smoke, soot and carbon fibers to be wafted downwind from the fire and, depending upon wind direction, have the potential of adversely impacting on transportation, manufacturing, and public service facilities and the power distribution systems.

This NASA study was really very preliminary, and not much was known about the behavior of composites in fires. The risk was taken seriously, however, and the Air Force sponsored two studies to develop guidelines for accidents.

These studies were sponsored by the Advanced Composites Program Office (ACPO) at McClellan Air Force Base, under the Advanced Composite Mishap Recovery program. The first study resulted in the Composite Aircraft Mishap Safety and Health Guidelines which were completed in 1992 and presented to the Air Force and other government agencies involved with composite aircraft.

The second study developed a more comprehensive set of requirements, the Mishap Risk Control Guidelines for Advanced Aerospace Materials, which was completed in 1993. These guidelines were coordinated throughout the Air Force, DoD, other government agencies, and private industry.

The Advanced Composite Mishap Recovery Web site is outdated, and little information beyond what I have printed here is available. I called the ACPO to find out more about the guidelines. The first set is not released to the public, and no one in the Office seemed to know about the second set. I will try to find out more about these guidelines from other sources and post an update here.

Accident Investigation

The unique properties of composites will also make the accident more difficult to investigate. The health hazards mentioned above (snakebites, carbon dust) will make physical investigation of the crash site both difficult and uncomfortable.

More importantly for the investigators, however, will be the destruction of physical evidence by the crash and fire.

Metals are a ductile material: they tend to bend and tear when damaged. Experienced investigators can examine a crumpled piece of metal and determine where the failure started and even what caused the initial failure.

Composites, on the other hand, are very brittle: they deform very little before breaking. In a controlled situation, it is still possible to determine where the failure started and what caused it. But that's when you have all of the pieces in relatively good condition.

The crash, though, would have pulverized much of the composites, breaking the fibers in the impact area into tiny particles and perhaps even dust. The fire would have burned off much of the resin, destroying the characteristic markings of the initial failure. Many of the bare fibers would also be carried away by wind, water, and firefighting materials, further destroying the evidence.

Updates and Feedback

Check back here for further details over the next several days and weeks. As more information becomes available, I will summarize it here. In the near term, I plan to do a more careful reading of the NASA study, and to look for more published information on the hazards of composites. I will also try to find out more information about the Mishap Guidelines. Finally, as the investigation proceeds, I will look for details about the role composites played in the crash.

If you have any information you would like to contribute to this, please write me at composite.guide@about.com. Let me know whether you want to be attributed as a source or if you prefer to remain anonymous.

Discuss in my forum