Reports of smoke filling the cabin of Boeing’s new B787 Dreamliner on a test flight, forcing an emergency landing and deployment of the inflatable slides to speed evacuation, raise a pertinent question: is the flying public going to be as vulnerable in the next 40 years as it has been in the past to smoke and fire? (See Aviation Safety Journal, “Boeing’s New B787 Suffers Electrical Fire in Flight”)
The B787 incident has parallels to an electrically induced fire that occurred on a United Airlines B777 at London’s Heathrow airport in February 2007. Good thing British investigators were on the case. United filed a service difficulty report (SDR) that indicated only “smoke,” not “fire,” so there was no reason for the Federal Aviation Administration (FAA) to press further. As for the U.S. National Transportation Safety Board (NTSB), its accident/incident database characterized the event as “fire” but no further information was given.
The case was duly and glibly passed to the U.K.’s Air Accidents Investigation Branch (AAIB). The AAIB, fortunately, was on the scene and produced a thorough and juicily detailed investigation in February 2009. The AAIB noted, “The electrical power system is designed to isolate a fault or failed device selectively while minimizing power interruption to a functioning system.” Then it went on to describe how the protections utterly failed to prevent a fierce and fiery conflagration:
— An internal failure of the Right Generator Circuit Breaker or Right Bus Tie Breaker contactor on the P200 power panel inside the Main Equipment Center resulted in severe internal arcing and short-circuits which melted the contactor casings.
— The open base of the P200 power panel allowed molten droplets from the failed contactors to drop down onto the insulation blankets and ignite them.
— The aircraft’s electrical protection system was not designed to detect and rapidly remove power from a contactor suffering from severe internal arcing and short-circuits.
— The contactors had internal design features that probably contributed to the uncontained failures.
The passengers were debarked by a mobile step way, as the airplane was taxiing to its takeoff position at the time of the electrical fire. The B787 incident was obviously more serious, as the occupants got off the plane via the evacuation slides. This recourse suggests that there wasn’t time to haul a mobile stairway up to the plane for a more deliberate debarkation.
In the B777 incident, the AAIB concluded:
“The RGCB [right generator circuit breaker] and Right Bus Tie Breaker (RBTB) suffered from severe internal arcing and short circuits which generated temperatures in excess of 1,000º C [1,800º F], and resulted in uncontained failures.”
Damage to the P200 power panel.
“Molten copper and silver droplets from the failed contactors dropped down through the open base of the P200 panel and ignited the insulation blankets below.”
“The insulation blanket fire spread underneath a floor panel to the opposite P205 power panel, causing heat and fire damage to structure, cooling ducts and wiring.”
Burnt structure and insulation blankets directly below the P200 power panel.
As in this B777 event, the B787 reportedly suffered from a spreading insulation blanket fire.
Now consider the three requirements for a fire: oxygen, an ignition source and fuel.
In the B777 emergency at Heathrow, and the B787 emergency landing at Laredo, TX, electrical arcing provided the ignition source and the insulation blankets provided the FUEL for the fire.
From 2000 to 2008, the FAA issued more than 20 airworthiness directives (ADs) directing the removal of insulation blankets that burn. From 1988 to 2010, at least eight FAA Technical Center reports and two FAA Safety Conference Papers were issued that addressed electrically induced fires or the insulation blankets that give fuel to the fire.
In 2002 an FAA Aging Transport Systems Rulemaking Advisory Committee (ATSRAC) presentation in Atlanta concluded:
— In-flight fires in hidden areas are a risk to aviation safety.
— Most hidden fires are caused by electrical problems.
— Non-compliance with safety regulations has been uncovered.
— Fire safety problems and improvements are in various stages of correction and study.
— It is impossible to predict the relative risk of serious fires occurring in hidden areas or locations.
March 1999: electrical fire on World Airways MD-11F
September 1999: in-flight fire damage on a Delta MD-88
August 2000: in-flight fire on an AirTran DC9
November 2000: in-flight fire on a B757 at Copenhagen, Denmark
November 2000: fire damage to an American Airlines MD-80
November 2000: cabin damage from fire on an AirTran DC9
The expression “various stages of correction” can mean anything, and the term “study” reveals the lack of a sense of urgency. Thus, the B787 crew had an urgent reason to land, and an urgent reason to evacuate the airplane as rapidly as possible, but the issues of electrical fire and flammable insulation blankets remain under “study” by the FAA – with no deadline for resolution.
After its investigation of the 1998 in-flight fire that downed Swissair flight 111, the Canadian Transportation Safety Board (TSB) said airliners ought not be designed and built with flammable components. If there is nothing to burn, the danger of electrical arcing igniting anything is hugely reduced. This TSB recommendation has obviously not been taken to heart. The B787 Dreamliner is a post-Swissair flight 111 design, yet flammable materials are still used in building new airliners.
Instead of this “strategic” approach, which implies a tightening of the certification standards for airplanes like the B787, the FAA remains stuck in the “tactical” mode, which is to say not much has changed. ADs are issued with overly generous compliance times in years. One of the latest ADs illustrates the FAA’s “slackadaisical” approach. AD 2008-23-09 requires flammable AN-26 insulation blankets to be replaced on a raft of Boeing designs, from the B727 to the B747. The rationale for the AD is given as:
“This AD results from reports of in-flight and ground fires on certain airplanes manufactured with insulation blankets covered with AN-26, which may contribute to the spread of a fire when ignition occurs from sources such as electrical arcing or sparking.”
The AD was issued in November 2008 and operators have 96 months to comply.
Actually, the AD is more accommodating to the industry than hinted at by this summary. The industry had ample notice that the AD was coming. From the time the Notice of Proposed Rulemaking (NPRM) was published in April of 2004, to the “comply-by date” of December 2014, the industry had fully 19 years and 8 months to work with, and nibble at the requirements to reduce the airplanes affected. This is seen in the “Cost of Compliance” section of the AD:
“The number of airplanes is reduced from those in the NPRM because of airplane retirements or changes from U.S. to foreign operation. A substantial decrease in estimated cost results from the net change of increasing parts and labor cost, but reduced number of airplanes, and a changed assumption of service for the entire fleet.
“All passenger airplanes in the AD fleet will reach 25 years of passenger service at most three years prior to the end of the compliance period, at which time we assume they will be converted into cargo service [which is 40% cheaper].”
If an airplane is sold into foreign operation, the provisions of the AD can be ignored. In other words, let foreigners be exposed to risks passengers will (belatedly) not be exposed to here in the U.S. Allowing eight years for the work on U.S. registered airplanes allows for much less cost than, say, a four year deadline.
The AD is typical. A model-by-model approach, over years to minimize the cost, while the root problem cited by the TSB of Canada goes unaddressed: the very materials used across many manufacturers and model applications.
A strict ban on flammable materials, as recommended by the TSB, should have been applied starting with the new B787. It wasn’t, and the emergency evacuation resulted.
This airplane is going to be in service for the next 40 years. Instead of future item-by-item corrective ADs, the vulnerability of the aircraft to in-flight fire should be designed out now.