"Going down in flames," though scary, is not common
Fire may be the devil's only friend, but it is among the average pilot's most dreaded in-flight possibilities. When contemplating an in-flight fire, most pilots worry first about being overcome by smoke and flames before they are able to complete a successful emergency landing. Still others - perhaps those with vivid imaginations - envision parts of the airplane melting away beneath them, leaving them with little or no control as their chariot plummets to the ground.
In-flight fires are a bit like IRS audits - everyone's afraid of them, but only a small percentage actually gets the experience. Indeed, your chances of having an in-flight fire are quite slim. In a search of the National Transportation Safety Board's files for the years 1996 through1998, we found 46 accidents with the keyword fire in the synopsis. Among those were several large aircraft - including the 1996 ValuJet crash in the Florida Everglades started by improperly marked and stowed oxygen generators. General aviation's share was 20 accidents; perhaps surprisingly, only five of these resulted in fatalities.
If you suspect a conflagration has broken out, one of your first duties is to discern its origin. Fires can be started and prolonged by a variety of sources. Fuel and oil fires can erupt from broken hoses or lines, but they still need some kind of ignition source. There have been many incidents in which a broken oil or fuel conduit has flooded the engine compartment and yet no fire broke out; these are the fortunate ones. A bit of flashover from the ignition system or contact with a hot exhaust system often is enough to commence the inferno.
Electrical fires can start ahead of the firewall or in the cabin. The worst are those that are started and sustained by a circuit that either is not protected by a fuse or circuit breaker or one that cannot be broken from the cockpit. Generally, these amount to high-current circuits, such as the battery main, starter supply line and, in some cases, the alternator output. If the alternator breaker is inside the cabin - which it normally is - there must be a length of its supply wire that is unprotected.
Good electrical-system design seeks to isolate these high-current subsystems and make them as robust as possible. It also seeks to protect current-carrying lines downstream as much as possible. Most of the time, it all works.
According to the NTSB accident briefs, the pilot of a Cessna 210 reported an in-flight electrical problem and requested a descent for an immediate landing; he crashed 12 miles from the airport. The pilot and his passenger were killed.
Post-accident investigation revealed evidence of an in-flight fire, with burned or charred control wheels, glareshield, and instrument panel. The windows were rendered opaque by soot. A California Highway Patrol officer witnessed the airplane descending with smoke coming from the cabin and a "paper-like debris" falling from the airplane.
An inductive noise filter fitted to the alternator was pinpointed as the source of the fire. Accident investigators and electrical lab technicians agreed that the capacitor shorted internally and arced on the firewall. This capacitor had been installed with neither circuit protection nor electrical isolation from the airframe. The NTSB report mentions the lack of FAA parts manufacturer approval (PMA) for the item and the fact that it is not a part of the original electrical system design. Yet this item had been on the airplane for three and one-half years.
Electrical-system woes brought down another airplane, an elderly Ercoupe. Its pilot radioed the tower that he had an electrical problem and needed to land immediately. He made a base entry to the runway from which he'd just departed. Witnesses saw the airplane continue its turn into a descending spiral that continued until it impacted the ground. Smoke was observed trailing from the aircraft.
This Ercoupe's electrical system had been extensively modified - among the changes, the battery had been moved from the baggage bay to the front of the firewall. According to the accident report, "Examination of [two runs of 11-gauge wire] revealed a frayed/worn-through section that matched an area of the firewall penetration hole. At another location on the wire was multistrand copper wire that had become molten and necked down to a break…the wires were located on the battery side of the master relay and were without circuit protection." For this unlucky Ercoupe pilot, there was no way to stop the electrical fire once it started.
At the first sign of an electrical fire - which often can be identified by a distinctive ozone-like odor - turn off the master and alternator switches. Attempt to determine if the fire is ahead of the firewall or inside the cabin. When you have extinguished the fire or stopped the smoke, reset the electrical system with care and by the book. Recommendations vary, but usually include keeping the avionics master and all other accessory switches off. Turn on the master first, then the alternator, and then, one by one, bring subsystems back on line. If a circuit breaker pops as you're doing this, do not reset it. It's good practice to complete the flight with minimum electrical load. Do not attempt to return the electrical system to service unless you have absolutely identified the cause and can isolate the circuit.
If you can see flames after power has been turned off, use your fire extinguisher first; only after you are sure the flames have been extinguished should you ventilate the cabin. (There have been cases of minor fires fanned into major ones by the opening of a window or vent.) You should also know how to turn off all ventilation and heating from ahead of the firewall.
Electrical systems do not bear the full brunt of the in-flight fire tally. Failures of exhaust-system components can lead to in-flight fires just as easily. Typically, a part failure will allow hot exhaust gases to impinge upon other engine-compartment components, including cowlings, engine controls, and oil and fuel lines. In one such accident, a Piper Warrior crashed when one of the exhaust stacks separated from the muffler. The unrestricted exhaust stream then impinged on the carburetor fuel-supply line and created an engine-compartment fire. According to the NTSB report, "Witnesses observed the airplane at very low altitude 'banking side to side with white smoke trailing from it.' The witnesses further stated that they saw the white smoke turn to black smoke and one witness saw 'orange flame coming out the left side of the engine compartment.'"
In a similar scenario, a homebuilt Velocity crashed after flames erupted from the engine compartment. Post-accident investigation discovered several oil and fuel hoses to be insufficiently tight and leaking. Because fuel and oil deposits were so thoroughly mixed, the investigators could not determine the exact cause of the fire. Incidentally, this pilot managed to get the airplane near the ground under control, but the airplane struck a pole on the crash landing.
In both accidents, the concept of poor maintenance practically leaps from the page. The Piper's exhaust system was held together with automotive clamps that lacked the locking pin found on standard aviation pieces; this pin helps to keep the components held together even if the torque on the clamp should fail. The Velocity's accident appears to have stemmed from poor attention to detail during the engine installation. In the case of the Piper, it's likely that turning off the fuel would have stopped the fire eventually, but not before consuming the fuel in the carburetor bowl and the supply lines downstream of the fuel selector.
Another fuel-related in-flight fire that had a far happier ending resulted from a broken fuel-injector line. The pilot of a Piper Turbo Saratoga noticed a drop in manifold pressure and a massive reduction - from 21 to 10 gph - in fuel flow. He turned on the boost pump and began a descent from his cruise altitude of 8,500 feet. Descending through 2,500 feet, the pilot noticed flames coming from the cowling vents and initiated an off-airport landing; there was but one minor injury to the four aboard. Later investigation revealed a broken fuel-injection line; a 1993 airworthiness directive requires repetitive inspection of some stainless fuel-delivery lines, and this airplane had just undergone the inspection.
What's important to note about this accident is that the pilot kept his cool and put the airplane on the ground safely, but also that the signs were there that a fuel line had broken. A sudden drop in fuel flow on certain fuel-injected engines - those whose flow gauges are in fact calibrated pressure instruments, not true flow gauges - usually is the result of a broken or loose line. In fact, the fuel pump was providing normal flow to the remaining cylinders, but the break caused the indicated pressure to become abnormally low. Cross-checking fuel flow with exhaust-gas or turbine-inlet temperature is a good way to determine if the indicated flow is correct. A multiprobe EGT instrument also would have told this pilot that a single cylinder was off-line as well.
Although the bulk of the accidents in our search pinpoint electrical or fuel-system maladies as the prime culprits, there were other causes of in-flight fires. A Piper Apache crashed after one of the left engine's cylinders separated in flight. An engine fire spread to the adjacent fuel tank, eventually causing the wing to fail. An in-flight airframe failure is among most pilots' worst nightmares, but they occur infrequently; most often in fatal accidents, the pilot is either overcome by smoke or panics and loses control of a perfectly intact airplane.
Preventing in-flight fires is largely a matter of consistent and meticulous maintenance and inspection. And surviving an in-flight fire is mainly in the ability to maintain composure and control and in a thorough understanding of the airplane's systems. It's hard to stay cool when your feet are hot, but it's the only way out of a dicey situation.
Links to additional information on fires and other in-flight emergencies may be found on AOPA Online ( www.aopa.org/ pilot/links/links9909.shtml).
