The McDonnell Douglas MD-11:

What kind of accident/incident history does it have?

The MD-11, manufactured by the McDonnell Douglas Co., is a derivative of the DC-10.  Less than 190 * MD-11s have been built, making it a commercial failure.  Boeing, who bought out McDonnell Douglas, is shutting down the production line in the near future.  Its low sales volume has been attributed to the competition from twin-engine airliners.  Undoubtedly that was a factor, but I suspect the previous bad design/accident history of the DC-10 also had a significant effect on airline management purchase decisions.

ACCIDENTS:

1992, December 7.  A China Airlines MD-11 encountered moderate turbulence at 33,000 ft. above sea level (FL 330).  “The airplane subsequently departed controlled flight and sustained damage to the left and right outboard elevator skin assemblies, portions of which separated from the airplane.”  The cost to repair the elevator damage was $312,000.  When the turbulence hit, the autopilot automatically disengaged, probably because of excessive roll rate.  The captain attempted to regain control, but it took almost 10 minutes.  During that time, both roll and pitch exceeded 30 degrees.  The flight data recorder (FDR) indicated the plane had stalled at least 4 times before recovery.  Amazingly, there were no injuries.

The MD-11 was designed with a smaller horizontal stabilizer than other airliners.  That, plus the shifting of its center of gravity further aft, all to reduce drag and thus fuel burn, causes it to be unusually light on the controls.  That design, known as “relaxed stability,” is common to fighter planes but is not normally found in the pitch axis of a civilian airliner.  It makes it more likely that the pilot will overcontrol and exacerbate the situation, during a recovery attempt after a high altitude upset. That deficient design, combined with a lack of pilot training for upset recovery in the MD-11, was the cause of the accident. 

1993, February 15.  A Brazilian Viaco Aerea MD-11 blew seven of 10 tires upon landing at San Francisco International Airport.  Debris from the tires and at least one brake assembly littered the runway for over 5,000 ft.  This kind of accident usually indicates some sort of defect in the anti-skid system.  Repair costs could easily have exceeded $100,000.

1993, April 6.  A China Eastern Airlines MD-11, on a flight from Beijing to Los Angeles, suffered an inadvertent deployment of the leading edge wing slats, while in cruise flight about 950 miles south of Shemya, Alaska.  The autopilot disconnected and the captain attempted manual control.  The plane progressed through several pitch oscillations, losing 5,000 ft. of altitude.  Two passengers were fatally injured; 149 others received various injuries ranging from light to severe.  Seven crewmembers were also injured, including one flight attendant who suffered severe brain damage.  The captain declared an emergency and diverted to the Air Force base at Shemya.  There was no external structural damage, but the interior cabin did sustain substantial damage.

During its investigation of this accident, the NTSB found “12 incidents of inadvertent or uncommanded in-flight slat extensions and 2 events on the ground involving MD-11 airplanes.”  The NTSB found the probable cause of the accident to be:

 …the inadequate design of the flap/slat actuation handle by the Douglas Aircraft Company that allowed the handle to be easily and inadvertently dislodged from the UP/RET position, thereby causing extension of the leading edge slats during flight.  The captain’s attempt to recover from the slat extension, given the reduced longitudinal stability and the associated light control force characteristics of the MD-11 in cruise flight, led to several violent pitch oscillations.

 Contributing to the violence of the pitch oscillations was the lack of specific MD-11 pilot training in recovery from high altitude upsets, and the influence of the stall warning system on the captain’s control responses.  Contributing to the severity of the injuries was the lack of seat restraint usage by the occupants.  [passengers did not have their seat belts fastened]

The stall warning system was cited as a contributing factor because its activation told the pilot that he must push the nose down even more, but that was the wrong action for a plane with relaxed stability designed into the pitch axis.  The investigation into this and the 1992 accident revealed:

  • There had been at least 5 other high-altitude upsets in the MD-11, including one that took place during the certification process.  These previous upsets all involved the failure of the pilots to adjust their control inputs properly when the autopilot disengages. Douglas Aircraft had “…not demonstrated, by flight tests, MD-11 stall recovery from abrupt high altitude, high speed upsets, nor were they required to do so as a part of the certification process.”

  • The FAA did not require pilots to “…receive hands-on training that demonstrated the light control forces encountered when manually flying at high altitudes and at high speeds in the MD-11.”

  • The design of the horizontal stabilizer (FAA approved, of course) was such that the stall buffet “…produced a dynamic load on the outboard elevators that resulted in structural overload and failure of portions of the outboard elevators.”  There had been 3 other incidents where MD-11s had suffered damage to their composite elevators, following stall buffets. 

Finally, the NTSB discovered something else about the MD-11:

The fire-retardant material used on the passenger seats had deteriorated and no longer provided fire protection to the seat cushions.  Although this deficiency did not directly compromise the safety of the passengers in this accident, it could potentially jeopardize the safety of passengers in accidents that result in interior cabin fires.

It is interesting to note that an Aeromexico DC-10 stalled at 29,800 ft. during the climb, as a result of pilot incompetence, on November 11, 1979.  It too, suffered elevator damage [the elevator provides the pilot with pitch control of the aircraft] as a result of entering the stall regime:

Approximately 4 ft. of the outboard ends of both outboard elevators were missing and had separated at almost identical locations…  The elevator outboard hinge fitting mounted on the elevator spar and the eyebolt mounted on the hinge filling had separated with the outboard end of the elevators…  no indication of prior cracking.  The fracture surfaces were clean and bright…

In my view, it is folly to certify a design that will permit external structural damage to the aircraft, especially to critical control structures such as elevators, just by the act of stalling the aircraft.  But then, we should trust the Government to know what is best for our safety….

1994, November 4.  A Fed Ex MD-11 freighter made a hard landing, and a tail strike at the Anchorage, Alaska airport.  After selecting 50 degrees of flaps the first officer, who was flying the plane, was not able to stabilize the approach in the pitch mode.  The attitude of the plane varied approximately 2 degrees with corresponding elevator position changes.  The captain, because of the high sink rate, grabbed the yoke and pulled back.  The plane landed hard, bounced, and oscillated at least three times, reaching a maximum pitch up attitude of 12.3 degrees.  The tail struck the runway during the oscillations.  This was also the plane that crashed at Newark in 1997.

1996, May 16.  A Fed Ex MD-11 freighter encountered wake turbulence from a preceding 747 as it was landing at the Anchorage, Alaska airport and suffered substantial damage.  When the plane entered a high sink rate, the captain tried to go-around but the lower aft fuselage hit the runway and bounced.  The captain discontinued the go-around and the plane bounced two more times, causing substantial damage to the aft pressure bulkhead. 

Prior to this accident, Fed Ex did not have formal tailstrike awareness training for its MD-11 pilots.  After this accident, however, Fed Ex developed a tail strike awareness training program that included bounced landing recovery in its simulator training.  That program limited pitch attitude to 7-1/2 deg for recovery from a bounced landing.

1997, June.  Thirteen people were seriously injured when a Japan Airlines MD-11 experienced severe pitch oscillations.  One passenger went into a coma and died, 20 months later.

The pilot attempted manual recovery when the autopilot failed to detect the plane was flying too fast after an encounter with wind shear.  The pilot’s repeated attempts to stabilize the altitude caused the severe oscillations that injured passengers and crew.  The autopilot was blamed for the accident because it contained a programming design defect that commanded it to respond to average velocity calculations instead of actual speeds.

1997, July 31.  A Fed Ex MD-11 bounced on landing at Newark airport and then flipped upside down off runway 22 R.  The two pilots and three passengers managed to escape before the plane was destroyed by fire.  The investigation is focusing on the failure of the right main gear, which allowed the right engine and wing to dig into the ground, flipping the plane over. 

          

          

Wreckage of the Fed Ex Newark accident, July 31, 1997

There is now some question about the structural strength of the wing box; whether it was strong enough to absorb loads that were well within the spar’s limits.  If it wasn’t as strong as it should have been, the next question will concern the same structures on other MD-11s.  Are they deficient too, or was it just a defect in the manufacture of that particular one?  The same plane had been involved in two other hard landing incidents, prior to this accident.  They are also examining the possibility of the pilot overcontrolling as he tried to correct after the first bounce.

It is obvious that Michael Crichton’s 1996 novel, AIRFRAME, is based on the checkered history of the MD-11, and in particular, on the deficient designs of its slats and pitch stability systems.  Crichton’s N-22 (his fictitious name for the MD-11) comes out as a stellar airplane that suffers only from bad press.  Likewise for the DC-10.  His discussion of the misfortunes of the DC-10 (beginning on page 181, in the paperback edition) is greatly distorted, chiefly by what he leaves out of the discussion.  In my view, the DC-10 is one of the worst designed airplanes to ever come down the pike and the MD-11 is not much better. 

1998, September 2.  A Swissair MD-11 plunged into the Atlantic Ocean, with the loss of all 229 onboard, after the pilots reported smoke in the cockpit.  That investigation is ongoing and they now have evidence of a fire fore and aft of the cockpit bulkhead.  Recovered parts, from the cockpit area, included a portion of the sheepskin cover from the F/O’s seat, an armrest, air filter, melted aluminum, electrical wires with melted copper, charred or missing wire insulation, and smaller parts that were discolored by heat.  Those parts are undergoing analysis to determine the temperature levels and heated gases to which they were exposed, according to the TSB (Transportation Safety Board of Canada).  That accident has prompted many questions received at this web site.  See www.airlinesafety.com/faq/faq8.htm

           

One recovered engine from Swissair 111

 Although the cause of that fire has yet to be determined, the FAA has nevertheless issued an AD (Airworthiness Directive), effective October 13, 1999, that:

…prohibits installation of a certain In-flight Entertainment Network system. This amendment is prompted by the results of a special certification review of the in-flight entertainment system installed on a Model MD-11 series airplane that was involved in a recent accident. The actions specified in this AD are intended to prevent possible confusion as the flightcrew performs their duties in response to a smoke/fumes emergency, which could impair their ability to correctly identify the source of the smoke/fumes and subsequently affect the continued safe flight and landing of the airplane….

The current design of the IFEN system electrical power switching is not compatible with the design concept of the MD-11 airplane with regard to the response by the flightcrew to a cabin or flight deck smoke/fumes emergency. In addition, the current IFEN system design does not provide the flightcrew and/or cabin crew with the ability to remove electrical power by a means other than pulling the system’s circuit breakers. The airplane manufacturer’s design concept of the airplane results in power being removed from the main cabin systems when the "CAB BUS" switch is engaged during a smoke/fumes emergency. However, the design of the IFEN system installation circumvented flightcrew procedures for responding to a smoke/fumes emergency by connecting the IFEN system to an electrical bus that is not de-energized when the "CAB BUS" switch is activated. Although the power to the IFEN system would eventually be removed via activation of the SMOKE ELEC/AIR rotary switch, the flightcrew would expect that selection of the "CAB BUS" switch would isolate all non-essential power to the cabin. Also, the cabin crew is able to only deactivate individual in-seat video displays (ISVD) from the IFEN system management terminal, deactivation does not remove electrical power from the ISVD’s and other IFEN system components. These conditions, if not corrected, could result in possible confusion as the flightcrew performs their duties in response to a smoke/fumes emergency, which could subsequently impair their ability to correctly identify the source of the smoke/fumes and subsequently affect the continued safe flight and landing of the airplane.

It is interesting to note that Swissair, on its own 111 accident web page, says that it disabled this IFEN system on all of its airplanes, after the 111 accident, as a precaution and because it was not necessary for the safety of the flight.  Swissair, nevertheless points out that the system was “certificated by the US Federal Aviation Administration.”

That is true; it was approved by the FAA.  But now, with this new AD, the FAA admits that it should not have been certified and installation of that system is now prohibited.

1998, September 10.  A China Eastern MD-11 made an emergency landing at Shanghai after its nose gear jammed.  It circled for 3 hours, dumping fuel during that time, and then landed on a foamed runway with the nose gear still jammed in the retracted position.  All 137 onboard evacuated safely while emergency fire trucks sprayed the aircraft. 

1999, April 15.  A Korean Air MD-11 freighter crashed into an industrial development six miles southwest of Hongqiao airport, Shanghai, China, 2 minutes after takeoff, killing the crew of 3.  Six on the ground were also killed, while thirty or more were injured.

1999, August 22.  A China Airlines MD-11 crashed while landing at Hong Kong’s Chek Lap Kok airport, during a rain storm with strong, gusting crosswinds.  The right gear struck the runway very hard and then broke off, allowing the right engine and wing to strike the ground.  The right wing then broke off and the plane flipped upside down.  That is very similar to the type of damage incurred by the Fed Ex MD-11, when it crashed at Newark.  Three of the 315 people on board Flight C1642 were killed.  The rest owe their lives to the Hong Kong airport’s fire brigade that put out the fire before it engulfed those trapped in the wreckage.  It took almost 3 hours to remove all the survivors.

          

Photo of the China Air crash at Hong Kong, courtesy of www.Airliners.net 

Failure of the wings, in the area of the fuselage attach points, has called into question  the structural integrity of the wing box structure of the MD-11 and is part of the focus of investigation in both accidents. 

It was the same plane that was involved in the turbulence accident on Dec. 7, 1992, described above.

1999, October 17.  A Fed Ex MD-11 touched down at Subic Bay, in the Philippines and then ran off the end of the runway and sunk in Subic Bay.  The plane was a total loss, but the 2 pilots escaped with only minor injuries.  The report found the probable cause to be:

The failure of the flight crew to properly address an erroneous airspeed indication during descent and landing, their failure to verify and select the correct airspeed by checking the standby airspeed indicator, and their failure to execute a missed approach. These failures led to an excessive approach and landing speed that resulted in a runway overshoot. Contributing factors to the accident were clogged pitot tube drain holes, the MD-11's insufficient alerting system for airspeed anomalies, and the failure of the SEL ELEV FEEL MAN and SEL FLAP LIM OVR D checklists to refer the crew to the standby airspeed indicator.

             

Fed Ex, Subic Bay.  The pilots were very lucky to escape this accident with their lives.  It would have been a much different story if the plane had been packed with 300 passengers.........

2001, November 20.  An Eva Air MD-11 made a hard landing at Taipei, Taiwan.  The first officer was flying the plane as it hit hard and bounced.  The captain immediately took control and initiated a successful go-around.  After the second landing, investigation revealed substantial damage to the nose wheel well structure and one of the two nose tires had failed.

INCIDENTS:

1995, June.  While the crew of a Chinese airline was preparing to start engines, they noticed smoke from the electrical and electronics bay (E/E) that is below the cockpit in the MD-11.  The investigation revealed that “molten metal, from arcing wires in the bay, had fallen on the blankets of insulation under the bay, igniting them.”  The Chinese official, Wu Xiangru, who sent a report to the American FAA, said that “There was extensive flame propagation from the insulation blankets up to the E/E bay with widespread damage.”  The Chinese officials did their own testing on the supposedly fire-safe insulation blankets.  They discovered the insulation would burn if it was exposed to high heat.  Thus, they suggested to the FAA that it review its testing standards which allowed such a fire to happen.

The FAA’s response was all too familiar:  it maintained its testing standards were adequate, noting that the Chinese tests were more extreme than the FAA’s requirements.  That, despite 3 other incidents in Douglas Aircraft types, where insulation blankets caught fire after being exposed to arcing electrical fires or hot shavings from a mechanic’s drill motor. 

Although the FAA failed to act, after learning of those 4 fires, Douglas did put out a service bulletin which recommended the replacement of the metalized Mylar insulation blankets, at the first maintenance opportunity, with another type.  That service bulletin was issued about one month after a heavy maintenance check was finished on the Swissair MD-11 that crashed in Nova Scotia.  Swissair decided, after receiving that bulletin, not to replace the insulation because “…it was a non-priority recommendation,” that had never been mandated by the FAA.

I cannot help but wonder if those 229 souls would be alive and well today if the FAA had acted promptly on the new information it had received.  Any wonder it is called the Tombstone Agency?

1998, October 8.  The crew of a Delta Airlines MD-11, detected an electrical odor in the cockpit, while climbing to FL350 out of Manchester, England.  There were two pilots, one jump seat rider, eleven F/As and 213 passengers on board at the time.  The flight diverted uneventfully into Shannon, Ireland.

1998, November 27.  A Swissair MD-11 experienced smoke in the cockpit upon climbout from Singapore.  The Singapore government is conducting the investigation into that incident.

1999, January 31.  An American Airlines MD-11 made an emergency landing at the Seattle-Tacoma International Airport , after discovering smoke in the cabin.  None of the 78 aboard was injured.

The flight was enroute from Seattle to Narita, Japan and was airborne for about 1 hour and 10 minutes when a "buzz" was heard over the PA system.  The pilots reset the PA circuit breaker.  Shortly thereafter, smoke was observed in the first class cabin.  An emergency was immediately declared and the flight returned to Seattle. 

The source of the smoke was located in an overhead bin just forward of door 2 Right.  The smoke dissipated after a halon fire extinguisher was discharged on a video system control unit (VSCU). 

Examination of the VSCU by FAA investigators, revealed that part of a circuit board was charred.  Examination of the entire video system revealed internal damage to several video distribution units (VDUs) downstream of the VSCU.  There also was evidence of moisture damage and a short circuit between two pins of a cannon plug that linked the damaged components.

The video system, manufactured by Rockwell Collins Passenger Systems, was certified by the FAA’s Long Beach Aircraft Certification Office.  According to Rockwell Collins, the connector failure was the first of its kind.

1999, March 29.  While opening the aft cargo bay floorboards, during a scheduled "4 C" maintenance check,  maintenance personnel at Santa Barbara Aerospace in San Bernardino, California, discovered evidence of a fire on board an MD-11, operated by World Airways.  The official incident report stated that the time and circumstances of the fire were unknown.  Removing the floorboards (first time ever) revealed that:

…the insulation blanket between stations 1661 and 1681 was found burned.  A detailed inspection of the area revealed that a wiring harness, containing 20-guage wires insulated with Kapton, was routed across and onto frame 1681. One wire was separated, and the insulation of seven other wires were damaged and chaffed where they contacted the frame. The bundle emanated from the aft cargo loading system control box, which routes 115 volt 3-phase power to electric floor rollers when the aft cargo door is in the fully open position.

Evidence of wire chaffing and arcing was present on the wire bundle and the frame where the bundle was contacted it. The metalized Mylar that covered the entire insulation blanket (measuring about 60 inches feet by 20 inches) that fit into the bay between frame 1661 and 1681 had completely burned away, exposing partially burned insulation material beneath it. A 1.25-inch hole in the blanket was found underneath the chaffed portion of the wire bundle. The mating edge of the adjoining insulation blanket (forward of frame 1681) was also burned. The metalized Mylar is DMS 2072K, type 2, class 1, grade A, lot no.2024. The tape that held the Mylar in place is DMS 1984 tape. Two wire bundle "stand-offs" were installed on either side of the arced area of the wires.

The plane, a freighter, was manufactured in 1992 and accumulated about 18,300 hours since delivery. On February 22nd, 1999 a deferred maintenance item was noted in the logbook regarding an inoperative electric cargo loading system.  The operator elected to defer removal of the floorboards until the 4C check.  While it is probable that the fire occurred at the time of the failure of that electric cargo loading system, no one can know for certain.

AIRWORTHINESS DIRECTIVES:

1998, December 3.  AD 98-25-11, amendment 39-10937 MD-11, to inspect for damaged wires near forward passenger doors.

…inspection to detect discrepancies at certain areas around the entry light connector of the sliding ceiling panel above the forward passenger doors….prompted by a report indicating that damaged electrical wires were found above the forward passenger doors due to flapper panels moving inboard and chafing the electrical wire assemblies of this area….AD [actions] are intended to prevent such chafing, which could result in an electrical fire in the passenger compartment.

1999, February 12.  AD 99-03-02, MD-11, for inspection to detect discrepancies of the wiring and insulation in the cockpit and overhead drop ceiling panel areas.

On September 2, 1998, a McDonnell Douglas Model MD-11 series airplane was involved in an accident following takeoff from John F. Kennedy International Airport in Jamaica, New York. The cause of the accident has not been determined.

In support of the subsequent accident investigation, examinations were conducted on several Model MD-11 series airplanes; the examinations focused on the area from the cockpit to station 515 (near the forward doors of the airplane in the forward drop ceiling area). The… results of these examinations…revealed chafed, cracked, broken, and cut electrical and bonding wires in several of these areas. These conditions, if not corrected, could result in electrical arcing of wiring and consequent fire and/or smoke in the cockpit or cabin

Since an unsafe condition has been identified that is likely to exist or develop on other airplanes of the same type design, this AD is being issued to prevent electrical arcing of wiring, which could cause a fire and/or smoke in the cockpit or cabin. This AD requires…visual inspection to detect discrepancies (including loose wire connections, loose ground wires, broken bonding wires, small wire bending radii, cracked support brackets, and chafed and cracked wire insulation) of the wiring and insulation in the cockpit and overhead drop ceiling panel areas…

The FAA has determined that this regulation is an emergency regulation that must be issued immediately to correct an unsafe condition in aircraft

1999, April 16. AD 99-06-08, MD-11 and DC-10, inspect for proper lubrication of main landing gear bolt.

…applicable to certain…DC-10 and MD-11 series airplanes, and KC-10 (military) series airplanes, that requires inspection for blockage of the lubrication holes on the forward trunnion spacer assembly, and inspection of the forward trunnion bolt on the left and right main landing gear (MLG) to detect discrepancies…reports of blockage by opposing bushings of the lubrication holes on the forward trunnion spacer assembly, and reports of flaking, galling, and corrosion of the forward trunnion bolt….which could result in premature failure of the forward trunnion bolt and could lead to separation of the MLG from the wing during takeoff and landing….

The blocked lubrication holes do not allow lubrication to reach the trunnion bolt. This condition can accelerate corrosion damage to the bolt, which could lead to the identified unsafe condition. An airplane that was in service for eight years may not have been subjected to loads that could contribute to failure of the bolt. However, another airplane may be in service for an even shorter period of time and yet experience loads that could lead to failure of a corroded bolt. Therefore, the FAA finds that repair of any discrepant spacer assembly prior to further flight is warranted.

1999, May 5. AD 99-08-51, MD-11, to inspect for missing bracket and clamp and damaged wire bundle.

…inspections under the floorboards in the lower center cargo compartment at frame 1681 to verify that a certain bracket and a certain open face nylon clamp are installed to a specific support wire bundle and to detect damage of the subject wire bundle; repair of damaged wiring; and installation of certain silicone rubber coated with a glass cloth protective wrap around the wire bundle, if necessary….an incident in which the insulation blanket between frames 1661 and 1681 in the lower center cargo compartment was found to be burnt due to a missing wiring harness support bracket/clamp on the wire bundle at frame 1681.…a missing bracket/clamp could cause the wire bundle to chafe against the frame, which could result in sparks, smoke, and possible fire in the lower center cargo compartment.

…a wiring harness (including wires with a 115 volt alternating current) of the aft cargo loader control unit was contacting the insulation blanket and rubbing against frame 1681 above stringer R46. A wire was cut, and three other wires were missing insulation. In addition, frame 1681 had signs of arcing damage….the exposed moisture barrier material of the insulation blanket was burnt, and a hole was detected on the insulation blanket where the wiring harness was chafing against frame 1681. Further investigation revealed that a wiring harness support bracket/clamp on the wire bundle at frame 1681 may not have been installed during production of the airplane.

The FAA has determined that this regulation is an emergency regulation that must be issued immediately to correct an unsafe condition in aircraft

1999, May 7. AD 99-09-01, MD-11, inspection for damage to wire bundles in the center accessory compartment (CAC).

…requires inspection to verify that the channel flanges of the bracket installations are facing forward and to detect chafing or damage of the wire bundles of the center accessory compartment (CAC) ….  [It] is prompted by an incident in which sparks and smoke came out of the CAC during a functional test due to a wire bundle that had chafed against a support bracket installation, which was installed improperly during production of the airplane…. Improper installation of the brackets of the CAC could cause chafing of the wire bundles, which could result in sparks, smoke, and possible fire in the CAC….the source of the sparks and smoke [was] a wire bundle that had chafed against a support bracket installation. A similar condition was noted on the left side of the CAC. The cause of such chafing has been attributed to improper installation (i.e., flange facing aft) of the brackets during production of the airplane….corrective actions include removing bracket installations that are facing aft; retaining bracket attaching hardware and wire clamps; reinstalling the bracket with flanges facing forward; reinstalling clamps; and repairing chafed or damaged wire bundles.

The FAA has determined that this regulation is an emergency regulation that must be issued immediately to correct an unsafe condition in aircraft

1999, May 7.  AD 99-09-02, MD-11, requires relocating the support bracket and rerouting electrical wiring in the aft storage compartment drop ceiling structure.

…is prompted by an incident in which a burning odor was detected, and the rear galley power repeatedly tripped off line during flight of an in-service airplane, due to the sense wiring of the galley load control unit (GLCU) chafing against the support bracket. The actions specified in this AD are intended to prevent chafing of the sense wire of the GLCU due to the location of the support bracket of the aft drop ceiling, which could result in electrical arcing, smoke, and possible fire in the aft drop ceiling area of the passenger compartments.…. the sense wiring of the galley load control unit (GLCU) located in the aft drop ceiling of the passenger compartments chafed against the light ballast; consequently, the wiring shorted. The cause of such chafing has been attributed to the location of the support bracket of the aft drop ceiling….

The FAA has determined that this regulation is an emergency regulation that must be issued immediately to correct an unsafe condition in aircraft…

1999, May 7.  AD 99-09-03, MD-11.  Inspection of wires in Main Avionics Rack.

…inspection of the wiring and wire bundles of the aft main avionics rack (MAR) to determine if the wires are damaged, or riding or chafing on structure, clamps, braces, standoffs, or clips, and to detect damaged or out of alignment rubber cushions inserts of the wiring clamps; and corrective actions, if necessary….is prompted by an incident in which the automatic and manual cargo door test in the cockpit was inoperative during dispatch of the airplane, due to the wiring of the MAR chafing against clamps as a result of the wire bundles being installed improperly during production of the airplane….Improper installation of such wiring and structure could cause chafing of the wire/wire bundles, which could result in electrical arcing, smoke, and possible fire in the MAR.

Investigation revealed the insulation of a wire located on the aft main avionics rack (MAR) was worn through, and that the wire shorted to a coax cable clamp. The wires that route from the main wire bundles to the MAR also were found contacting clamps at other locations of the MAR. The cause of such chafing has been attributed to improper installation of the wire bundles in the MAR during production of the airplane. (This incident is not considered to be related to an accident that occurred off the coast of Nova Scotia involving a McDonnell Douglas Model MD-11 series airplane. The cause of that accident is still under investigation.)

1999, May 7.  AD 99-09-04, MD-11, for incorrect installation of circuit breakers, during production.

…inspection to verify correct wire terminations of certain circuit breakers in the cockpit overhead switch panel; and correction of incorrect wire termination…. This amendment is prompted by incidents in which the wiring of circuit breakers on the overhead switch panel lighting were found to be terminated improperly during production of the airplane, which bypassed the circuit breaker protection. The actions specified in this AD are intended to prevent smoke and possible fire in the overhead switch panel lighting circuitry due to an overload condition, as a result of lack of circuit breaker protection.

…the FAA has become aware of an incident in which the wiring to a circuit breaker on the overhead switch panel lighting was found to be terminated improperly on a…MD-11….The bus assembly and the wire were connected on the same lug with nothing connected to the load side of the circuit breaker (i.e., bypassing the circuit breaker protection).

A subsequent line check of Model MD-11 series airplanes in production revealed that the wiring to three other circuit breakers on the overhead switch panel also were terminated improperly on some airplanes. Further investigation revealed that the MD-11 production build paper did not reference the wire hook-up chart for wire termination of the circuit breakers of the overhead switch panel lighting. (These incidents are not considered to be related to an accident that occurred off the coast of Nova Scotia involving a McDonnell Douglas Model MD-11 series airplane. The cause of that accident is still under investigation.)

Lack of circuit breaker protection for the circuit of the overhead switch panel lighting, if not corrected, could result in smoke and possible fire in the overhead switch panel lighting if the circuit breaker has an overload condition.

1999, August 11.  The FAA issued a press release noticing a proposed AD:  # APA 87-99 FAA to Order Insulation Replacement on Select Aircraft.

To reduce the risk of the spread of fire aboard aircraft, FAA Administrator Jane F. Garvey today said the agency is ordering operators of 699 aircraft to replace insulation blankets covered with metalized Mylar within four years. The FAA is also strongly encouraging operators to accomplish the insulation replacement during the earliest practical maintenance check.

The announcement follows eight months of extensive testing in support of the development of a new test standard for aircraft insulation.

The FAA is going beyond the current, acceptable level of safety and is proposing an even higher standard for testing insulation on all new aircraft. The new test standard was developed by the FAA with input from world-renowned fire experts. The agency plans to issue a proposal for all new aircraft later this year.

While other insulation materials in the current U.S. fleet are safe, tests show that metalized Mylar falls far below the new test standard. The proposed Airworthiness Directives (ADs) would affect DC-10, MD-11, MD-80, MD-88, and MD-90 aircraft.  They will require operators to remove metalized Mylar-covered insulation. Replacement materials must meet the FAA's new proposed flame propagation standard that is based on the American Society for Testing and Materials (ASTM) standard for flammability. Materials such as polyimide, certain polyvinylfluorides and certain fluoropolymer composites have been shown to be capable of meeting the ASTM test. 

‘The FAA's track record shows that we don't hesitate to have airlines retrofit the fleet when there is a threat to passenger safety,’ said Garvey. ‘We've weighed the benefit of replacing insulation, reviewed the service history of these aircraft and have made the right decision based on scientific data.’

Anytime an aircraft is taken apart, there is a possible risk of damaging aircraft wiring. Replacing aircraft insulation is complex and must be performed safely to avoid unintended consequences. Insulation is not easily accessible and replacement involves removal of overhead panels and floors. The work must be accomplished at the earliest maintenance check, but no later than four years. This allows for a safe and deliberative process designed to minimize the possibility of creating unintended safety problems….

Working with input from aviation experts around the world, the FAA replicated how different insulation materials behave in simulated fire situations. Using the new standard, FAA scientists measured a material's ability to prevent or contain the spread of fire.  Metalized Mylar fell short of an acceptable safety level and far below the new standard. It ignites much more easily than other materials and can spread fire because its properties are much different. The other materials performed better than originally anticipated and meet the acceptable level of safety. While these materials may not meet the new, higher standard, they do not pose a threat to aviation safety.

The estimated cost to U.S. operators is approximately $255 million, $380,000 to $880,000 per airplane.

KAPTON:

On its after-the-111-accident web page, Swissair answers an inquiry about Kapton wiring as follows:

Thirteen Swissair MD-11s have Kapton wiring insulation aboard, including the aircraft which was lost. Three of the fleet have the newer Teflon-Kapton-Teflon insulation material.

Up until 1995, almost all aircraft were equipped with Kapton wiring insulation.  From then onwards, a switch was gradually made to the Teflon-Kapton-Teflon alternative, which is lighter and less expensive.

Kapton has been rigorously tested and its use approved by the aviation authorities. It has never given cause for concern. It has been replaced by other materials on military aircraft, but this is because on these aircraft it is subjected to greater physical stress, owing to the aircraft's different operating conditions and methods of construction.

While it is true that Kapton was “rigorously tested,” the same can be said for metalized Mylar insulation, the Boeing 737 rudder PCU, the aft DC-10 cargo door, the fwd 747 cargo door and the all-hydraulic flight controls of the 747 and DC-10.  All were FAA approved and all those designs have failed and either cost lives or are suspected to have contributed to the loss of life (Swissair 111).

The “It has never given cause for concern” statement can no longer be viewed as accurate, in light of the above incidents and resulting ADs surrounding arching of that kind of wire.

One of the justifications for using Kapton was that it had superior flame-resistance to gasoline fires.  It also was supposed to have “good chafing resistance.”  But, under some conditions (especially if moisture is present) it can arc rather dramatically -- “much like a dynamite fuse,” – according to one aviation accident lawyer.  While I consider that somewhat of an overstatement, it nevertheless seems that the combination of metalized Mylar insulation, Kapton wiring, incompetent factory installation and poor quality control, produce an environment of inordinate risk to all who ride on the MD-11.  We know that from the June, 1995 experience of the Chinese airline that suffered substantial fire damage in the electronics bay when Kapton wire arched and the molten metal, from that wire, set the Mylar ablaze.  The other fires, itemized in the INCIDENTS section above, would also seem to indicate the logic of that conclusion.  

It is obvious  the NTSB also became concerned about the quality of wiring in the MD-11, during the early stages of the TSB investigation into the Swissair accident:

National Transportation Safety Board

Washington, D.C. 20594

Safety Recommendation

Date: January 11, 1999

In reply refer to: A-99-3

Honorable Jane F. Garvey, Administrator

Federal Aviation Administration

Washington, D.C. 20591

On September 2, 1998, at 2018 eastern daylight time, Swissair flight 111, a McDonnell Douglas MD-11 registered as HB-IWF, departed from John F. Kennedy International Airport in Jamaica, New York. Swissair flight 111 was a regularly scheduled passenger flight from New York to Geneva, Switzerland, operating under the provisions of 14 Code of Federal Regulations Part 129.

About 56 minutes after departure while at flight level 330, the flightcrew declared "PAN PAN PAN" and advised air traffic control (ATC) of smoke in the cockpit. The flightcrew requested to divert to a convenient airport and was cleared direct to Halifax International Airport in Nova Scotia, Canada. About 11 minutes after the report of smoke, the airplane’s electrical systems began to deteriorate. The flightcrew then declared an emergency, and communications between ATC and the flightcrew ceased shortly thereafter. Approximately 6 minutes later, at 2231 Atlantic daylight time, the airplane crashed into the Atlantic Ocean near Peggy’s Cove, Nova Scotia, Canada. All 14 crewmembers and 215 passengers were killed, and the airplane was destroyed. The Transportation Safety Board of Canada (TSB) is in charge of the accidentinvestigation, and the National Transportation Safety Board is participating in accordance with the provisions of Annex 13 to the Convention on International Civil Aviation.

Approximately 85 percent of the airplane’s wreckage has been recovered to date. Examination of the wreckage revealed evidence of considerable heat damage to ceiling areas both forward and aft of the cockpit bulkhead. This damage is consistent with the effects of a fire. Numerous sections of wiring from the cockpit overhead area also exhibited heat damage and burned insulation, and several of the wires from those sections showed evidence consistent with electrical arcing. 

Although some of the wires exhibiting arcing characteristics are from the entertainment system that is unique to the Swissair MD-11 fleet, others have been identified as original MD-11 wires.

On December 22, 1998, the TSB issued Aviation Safety Advisory 980031-1 to the Safety Board as the representative of the MD-11’s State of Manufacturer; a copy of the advisory was also sent to Switzerland’s Aircraft Accident Investigation Bureau, Transport Canada, and the FAA. The advisory points out that TSB investigators recovered two MD-11 electrical bus feed wires that show signs of arcing. The two wires are identified as the left emergency alternating current bus feed wire (wire number B205-1-10) and the left emergency direct current bus feed wire (wire number B205-4-6).

If the wires were in place, the area of apparent arcing damage would be located approximately 2 inches aft of the right cutout in the "tub" that encloses the overhead circuit breaker panel.  According to the advisory, the potential safety ramifications appear to be confined only to the MD-11 fleet.

TSB investigators also recovered the overhead circuit breaker panel and the upper avionics circuit breaker panel. Portions of each panel show evidence of substantial heat damage. The avionics circuit breaker panel is located along the right side of the cockpit behind the first officer’s seat, but the upper portion of that panel extends into the area near the overhead circuit breaker panel.

In addition, TSB investigators participated in examinations of several other MD-11 airplanes that concentrated on the area from the cockpit to station 600 (near the forward doors of the airplane).

The examinations showed the following:

· chafed, cracked, broken, and cut electrical and bonding wires in several areas, including the overhead and avionics circuit breaker panels and the forward drop ceiling area above the left (L) 1 and right (R) 1 doors and 

· inconsistencies in the routing of wires and wire bundles, loose terminal connections, excessively small bend radii, unsealed electrical conduits, and open smoke barriers between the cockpit and cabin areas.

On December 10, 1998, the FAA issued Airworthiness Directive 98-25-11, "McDonnell Douglas MD-11 Series Airplanes," requiring a one-time inspection above the L1 and R1 doors to address the wire chafing issue discovered as part of the accident investigation. Also in December 1998, Boeing issued two MD-11 Alert Service Bulletins—MD-24A068 Revision 1 and MD-25A194 Revision 4—which address the specific discrepancies regarding the door areas. Further, SR Technics, on behalf of Swissair and along with Boeing, has voluntarily developed an engineering order that defines a comprehensive inspection of the wiring in the forward areas of the Swissair MD-11 airplanes.

The Safety Board understands that the inspections completed to date have not uncovered any discrepancies that warrant regulatory action.

Although the apparent electrical arcing on Swissair flight 111 has not been determined to be a source of a fire, and Swissair’s voluntary inspections of its MD-11 airplanes have not uncovered serious discrepancies, the Safety Board is concerned about the recent discoveries of apparent electrical arcing damage to wiring near the accident airplane’s overhead and avionics circuit breaker panels, the heat damage to those panels, and the wiring anomalies discovered in TSB's examination of MD-11 airplanes.

Therefore, the National Transportation Safety Board recommends that the Federal Aviation Administration:

Require, on an expedited basis, an inspection of all MD-11 airplanes for discrepancies of wiring in and around the cockpit overhead circuit breaker panel (including the area just aft of the tub enclosure) and the avionics circuit breaker panel. The inspection should include examinations for loose wire connections, inconsistent wire routings, broken bonding wires, small wire bend radii, and chafed and cracked wire insulation.

(A-99-3)

Chairman HALL, Vice Chairman FRANCIS, and Members HAMMERSCHMIDT,

GOGLIA, and BLACK concurred in this recommendation.

[original signed]

By: Jim Hall, Chairman

 

CONCLUSION:

The MD-11 was designed with a smaller horizontal stabilizer than other airliners.  That, plus the shifting of its center of gravity further aft, all to reduce drag and thus fuel burn, causes it to be unusually light on the controls.  That design, known as “relaxed stability,” is common to fighter planes but is not normally found in the pitch axis of a civilian airliner.  It makes it more likely that the pilot will overcontrol and exacerbate the situation, during a recovery attempt after a high altitude upset or during a bounced/hard landing.

I have never flown an MD-11.  However, I do have a description of its handling characteristics from a pilot that has had considerable experience in that cockpit:

The MD-11 is not fly-by-wire. It is, however, fly by CONSTANT pilot input. The geniuses at MD decided to make the empennage 40% smaller than the DC-10 to save on both parasitic drag and induced drag by keeping the c.g.[center of gravity] near the aft limit during high-speed cruise.

This airplane doesn't really have a "slot" when you are on final; it doesn't seem to really stay at a trimmed AOA [angle of attack] /deck angle at a specific power setting/airspeed. As such, the pilot is constantly making little corrections, like flying a dynamically unstable fly-by-wire fighter with the computer out. This is unlike any transport aircraft I've flown. Part of the problem is a system called the Longitudinal Stability Augmentation System (LSAS) which is a computer that constantly trims the stab to make up for the shortcomings of the tail size. The landing is also unique. As soon as the plane touches down I have to push on the yoke to counteract a severe pitchup from the spoilers coming to 2/3 extension. Less than a second later, the autobrakes kick in, so you have to pull back on the yoke to gently lower the nose to the runway.

Somebody once said they should let Lockheed design all the airplanes, Boeing build them...and McDonnell-Douglas market them! And let the French guys stick to making Citroens and Peugeots...

In my view, that unstable pitch mode constitutes defective design, which is directly responsible for all the deaths and injuries that have occurred during high altitude upsets and the resulting violent pitch oscillations.

I also think it reasonable conjecture that the Anchorage, Newark and Hong Kong landing accidents might not have happened at all if the MD-11 was designed with a stable pitch mode as are Boeing airliners.   The kinds of conditions encountered in those accidents (heavy weights, short runway, wake turbulence, gusting crosswinds), are personally known to airline pilots of high experience.  That is when the skills of the pilot are put to the ultimate test, and the design quality of the plane is revealed.  It is precisely the time when the pilot, and all on board, need everything going for them. 

The pilot must be skilled enough to instantly recognize what control inputs are needed, and the plane must be designed so as to respond instantly to those commands without excessive oscillations.  The brain of the pilot and the brain of the design engineer must be simpatico.  If they are not, then in my opinion, you get what happened in some of the accidents listed above.  

See, also, the letter on the MD-10.

For additional information on MD-11 accidents, go to Report on Hong Kong Accident.

*While that number was correct when I first wrote this article, the final production number was 200, when the production line was shut down in the year 2000, according to Boeing.  "The last MD-11 was delivered Feb. 22, 2001."

October, 1999, revised September, 2002 & December, 2010

[All emphasis in this article is that of the Editor]

Robert J. Boser
Editor-in-Chief
AirlineSafety.Com

Return to FAQ page


HOME | UNIONS | FREQUENTLY ASKED QUESTIONS | WHAT'S NEW? | EDITORIALS | LETTERS TO THE EDITOR 

QUOTATIONS | AVIATION NEWS  | BOOK REVIEWS | SAFETY ARTICLES | LINKS | CONTACT US

All Material © 1997-2014  All rights are reserved. No part of this web site may be reproduced in any way without expressed written consent