This summer marked the 25th anniversary of the most complex and vexing aircraft accident in aviation history – TWA Flight 800. The vintage Boeing 747-100 had just departed from New York on its way to Paris when it suddenly exploded off the coast of Long Island. All 230 people on board were killed. As a rookie NTSB field investigator at the time, I was not involved in the years-long effort to solve the accident, but I sure do remember its impact, especially later when I taught at the NTSB Training Center where the TWA 800 reconstruction was housed (see top image at right).
The NTSB concluded that the center wing fuel tank of the 747 exploded, likely due to wire chaffing and arcing outside of the tank, which led to a short circuit that allowed excessive voltage to enter the tank through the wiring for the fuel quantity indication system (FQIS). Additionally, NTSB examinations of wiring on 26 other airplanes of varying ages (ranging from new to 28 years old) revealed that all of the older airplanes exhibited numerous examples of mechanically damaged, chafed, cracked, and contaminated wires (see image at lower right). Sharp-edged metal drill shavings (which can damage wire insulation), fluid stains, and other potentially hazardous material were also found in or near the wiring of old and new airplanes.
While not considered a “maintenance related” accident, the TWA 800 tragedy focused the MRO industry’s attention on the pitfalls of aging and improperly installed electrical wiring. Unfortunately, several more similar accidents and incidents followed in the late 1990s.
SWISSAIR Flight 111
Two years after the TWA disaster, another wide-body jet was felled by wiring problems. This time, it was a McDonnell Douglas MD-11, operating as Swissair Flight 111 on a flight from New York to Geneva, Switzerland, with 215 passengers and 14 crew on board. And, this time, I was involved in the investigation.
The date was September 2, 1998. Swissair Flight 111 was cruising at 33,000 feet about an hour after departure from JFK Airport when the flight crew reported a smell of smoke. As the smoke became dense and entered the cockpit, the crew attempted an emergency landing at Halifax International Airport in Nova Scotia, Canada. But it was too late. The MD-11 impacted the Atlantic Ocean a few miles off the coast near Halifax. There were no survivors.
Freshly promoted to the NTSB “go-team” as a systems engineer, I launched to Halifax to assist the Transportation Safety Board (TSB) of Canada since the aircraft was designed and built in the United States. I spent several weeks on a floating barge (see top right) with other investigators, sifting through piles of shredded wreckage brought up from the ocean bottom by a massive crane and bucket.
The Canadian TSB’s meticulous investigation rivaled that of the NTSB’s effort with TWA 800, including a 3-dimensional reconstruction of the front of the airplane (see lower right). Evidence revealed that a fire raged above the ceiling in the front area of the aircraft. The Swissair MD-11 was modified with an in-flight entertainment network (IFEN) for first class passengers, and it was connected to aircraft power in a manner that was not compatible with adequate emergency electrical load-shedding.
The IFEN was an approved, but inadequately reviewed, Supplemental Type Certificate (STC) installation.
A review of IFEN system installation records revealed discrepancies in the drawings and supporting documentation. No details for wire routing were provided. Additional inspections of wiring around the cockpit overhead circuit breaker panel in other MD-11s revealed loose wire connections, inconsistent wire routings, broken bonding wires, small wire bend radii, and cracked and chaffed wire insulation.
The investigation led to numerous recommendations regarding insulation flammability, crew checklist procedures, and system design. More importantly for the readers of this magazine, the accident highlighted the importance of guidance contained in FAA Advisory Circular (AC) 43.13-1B, “Acceptable Methods, Techniques, and Practices—Aircraft Inspection and Repair,” and AC 65-15A, “Airframe and Powerplant Mechanics Airframe Handbook.” These “best practices” rely heavily on the training and experience of the maintenance professionals who perform the installation work to determine proper wire routing.
More Incidents of Arcing and Sparking
Because of the TWA 800 and Swissair 111 tragedies, the aviation industry was “on edge” to say the least. Every report of a potential electrical arc or short was scrutinized. As the “on call” NTSB systems investigator in September of 1999, I was dispatched to examine a Delta Air Lines MD‐88 that performed a precautionary landing near Cincinnati, Ohio, after declaring an emergency due to a cabin fire. The flight attendants reported that there was a sulfurous smell followed soon after by smoke in the forward cabin. During the descent to land, a flight attendant discharged a Halon fire extinguisher into a floor grill where she saw a flame.
The MD-88 has two heater plates on both sides of the fuselage that are flush-mounted against the static air pressure sensing ports to ensure that the ports do not become blocked by ice (top image page 40). The heaters are powered by 115-volt alternating current through a 10-ampere circuit. The investigation revealed that a spark from the right static port heater plate ignited a small fire that propagated by consuming the sidewall insulation blankets surrounding the heater (lower image page 40). Examination of the heater revealed localized soot on the thermostat case, and on the lead wire that carries the 115 volts to the thermostat. The lead wire was bent sharply around the thermostat case, and its conductor was exposed at the bend (top image page 41) due to a manufacturing flaw. In response to the incident, Delta Air Lines initiated detailed visual, electrical, and functional inspections of the static port heaters on its entire MD-88/MD-90 fleet of 136 airplanes. What they found was frightening: 11 percent of the airplanes had at least one heater installation that exhibited some type of damage. Nine of the heater installations had arced, burned, or melted parts in the area of the electrical connector. Two heaters had charred and exposed elements on the heater plate.
In addition to the Delta MD-88 incident, I investigated another event involving an electrical fire in a World Airways MD-11. A vigilant mechanic at an MRO in California discovered evidence of sooting while removing several floorboards in the MD-11’s cargo hold. A wiring harness was routed onto a frame without the required support bracket/clamp, which allowed a wire bundle to chafe against the frame. The finding led to an FAA airworthiness directive (AD) that required visual inspections of certain MD-11 airplanes to verify that a bracket and nylon clamp were installed to support a specific wire bundle, repair any damage to the bundle, and install a protective wrap around it. Who knows what could have happened if that MRO mechanic ignored this discrepancy.
Rearing its Ugly Head in San Francisco: ABX Air 767-200
The airline industry kept serious wiring issues down to a dull roar during the years immediately following these events. But then, in late June 2008, while filling in for the vacationing director and deputy director of the NTSB’s Office of Aviation Safety, I was notified of an ABX Air 767-200 freighter that experienced a ground fire at San Francisco International Airport just before engine start. The pilots evacuated through the cockpit windows and were not injured, but the airplane’s crown had severe fire damage (see middle image on page 41). I facilitated the launch of a small go-team to investigate the event. I remember thinking: “But for the grace of God that the fire occurred on the ground rather than in the air.”
The 767 was converted from a passenger to a cargo configuration four years prior to the event by a company that performed an STC modification to the “supernumerary” crew seating area behind the flight deck. The STC included flexible hoses for a supplemental oxygen system (see lower image page 41). During postaccident inspections of other ABX Air 767 airplanes that were modified by the same company, some installations were found to have electrical wiring routed above and in direct contact with the oxygen tubing, even though the STC provided for positive separation. The investigation of the fire aboard the accident airplane found that a short circuit from electrical wiring was the most likely source to energize a coil spring inside an oxygen system hose, causing the hose material to ignite.
Advisory Circulars 43.13-1B and AC 65-15A state that no electrical wire should be located within ½-inch of any combustible fluid or oxygen line and that, if the separation is less than 2 inches, back-to-back clamps or a polyethylene sleeve should be installed to ensure positive separation. However, this guidance was not followed in the installation and inspection of the STC in the ABX Air 767, and a fire ensued.
History Repeats Itself: A Piper Cheyenne II in-flight fire.
Later in my career, while serving as the director of FAA’s Accident Investigation Division, another accident occurred due to the lack of separation between electrical wiring and other systems. On July 29, 2016, a Piper PA‐31T Cheyenne II twin-turboprop — while operating as an air ambulance — broke up in flight over McKinleyville, California shortly after the pilot reported smoke in the cockpit. The pilot, two medical personnel, and the patient were killed.
The wreckage was located several hours later in heavily forested terrain. Portions of the burned and fragmented wreckage were scattered along a half-mile debris path. The center fuselage and cockpit areas were largely intact and displayed no evidence of fire (see top image page 42); however, a large area of thermal damage to the forward fuselage and circuit breaker panels were found. An aluminum stringer in this location exhibited “broomstrawing” indicating that it was heated to near its melting point prior to impact. A single wire located in the area exhibited “notching” consistent with mechanical rubbing (see lower image page 42) and exhibited evidence of electrical arcing. Four hydraulic lines servicing the landing gear were located in this area, and all the lines were partially burned, melted and missing sections of material.
Prompted by these disturbing findings, six other airplanes of the same make and model were examined.
Sure enough, they all had instances of electrical wires and hydraulic lines in direct contact with each other in the area of the main bus tie circuit breaker panel. Some of the wires were chaffed. The NTSB stated the cause was: “An inflight fire in the floor area near the main bus tie circuit breaker panel that resulted from chafing between an electrical wire and a hydraulic line and/or airplane structure.” But even before the cause was determined, the NTSB issued an urgent safety recommendation to address the unsafe wiring conditions. The FAA quickly issued an AD that required repetitive detailed visual inspection of the wiring below the circuit breaker panels in Piper PA-31T series airplanes.
Prevent Fires by Checking Wires
The lessons learned from these accidents and incidents are plentiful. If you are a maintenance professional who wants to prevent opportunities for aircraft fires, you need to heed the guidance from the FAA and the manufacturer. More importantly, if you “see something” similar to the items listed below, then you need to “say something” so that an electrical fire can be prevented:
• No wire should be located within ½-inch of any combustible fluid or oxygen line.
• If the separation is less than 2 inches, back-to-back clamps or a polyethylene sleeve should be installed to ensure positive separation.
• To prevent chaffing, wiring harnesses should not be routed onto a frame without the required support bracket/clamp.
• Generally, clamps should not be spaced at intervals exceeding 24 inches. In high-vibration areas or areas requiring routing around structural intrusions, the clamping intervals may need to be reduced in order to provide adequate support.
• Routing of wires with dissimilar insulation, within the same bundle, is not recommended.
• Accumulation of dirt and lint near electrical wires creates a potential for smoke and fire.
• The minimum radii for bends in wire groups or bundles must not be less than 10 times the outside diameter of their largest wire.
• Metal stand-offs must be used to maintain clearance between wires and structure. Employing tape or tubing is not acceptable as an alternative to stand-offs.
Many wire defects may be difficult or impossible to detect through visual inspection alone, automated test equipment (ATE) inspection systems are available to supplement visual inspections. These systems include electrical continuity or resistance tests, insulation resistance and capacitance tests, and time-domain reflectometry (TDR).