You may be trying to access this site from a secured browser on the server. Please enable scripts and reload this page.
Turn on more accessible mode
Turn off more accessible mode
Skip Ribbon Commands
Skip to main content
Top Link Bar
NEWS & EVENTS
Speeches & Testimony
Most Wanted List
The Investigative Process
Data & Stats
General Aviation Safety
Assistance to Families & Victims
Operations & Policy
Administrative Law Judges
Strategic Plans & Reports
Safety Recommendation Details
The Investigative Process
Data & Stats
General Aviation Safety
On September 28, 2007, about 1313 central daylight time, American Airlines flight 1400, a McDonnell Douglas DC-9-82 (MD-82),2 N454AA, experienced an in-flight engine fire during departure climb from Lambert-St. Louis International Airport (STL), St. Louis, Missouri. During the return to STL, the nose landing gear failed to extend, and the flight crew executed a go-around, during which the crew extended the nose gear using the emergency procedure. The flight crew conducted an emergency landing, and the 2 flight crewmembers, 3 flight attendants, and 138 passengers deplaned on the runway. No occupant injuries were reported, but the airplane sustained substantial damage from the fire. The scheduled passenger flight was operating under the provisions of 14 Code of Federal Regulations Part 121 on an instrument flight rules flight plan. Visual meteorological conditions prevailed at the time of the accident.
TO THE FEDERAL AVIATION ADMINISTRATION: Require Boeing to establish an appropriate replacement interval for air turbine starter valve air filters installed on all MD-80 series aircraft.
Original recommendation transmittal letter:
Closed - Unacceptable Action
St. Louis, MO, United States
In-Flight Left Engine Fire, American Airlines Flight 1400, McDonnell Douglas DC-9-82
Addressee(s) and Addressee Status:
FAA (Closed - Unacceptable Action)
Safety Recommendation History
The FAA indicated that Boeing reviewed and, in February 2011, revised the Aircraft Maintenance Manual (AMM) procedure and task card with regard to the ATSV. The NTSB reviewed the revisions and found that the only mention of an ATSV-filter inspection is an instruction to “Inspect starter air shutoff valve inline filter for deterioration, corrosion, obvious damage and general condition.” The FAA believes that, in the American Airlines flight 1400 accident, the deterioration and failure of the ATSV air filter was not the cause of the ATSV failure and that filter failures are not a safety-of-flight issue. The FAA examined the service history of ATSV problems reported by MD-80 operators and concluded that operators who performed the Boeing-recommended maintenance for the ATSV air filter did not experience in-service failures. Therefore, the FAA does not plan to require a mandatory replacement interval for the ATSV air filter; rather, the FAA considers its action in response to this recommendation to be complete. In our previous letter classifying Safety Recommendation A-09-28, we stated that the FAA’s review of the service history alone would not be sufficient because this action did not take into account the findings of our investigation of the American Airlines flight 1400 accident. In the letter that transmitted this recommendation to the FAA, we explained that, during the accident investigation, American Airlines had sent seven serviceable ATSV air filters to the NTSB for evaluation. Three of the seven filters revealed evidence of the onset of fatigue in the filter elements’ outer mesh. American Airlines sent 15 additional, serviceable ATSV air filters to the ATSV air filter manufacturer for bubble testing and visual inspections, which were performed in accordance with the Component Maintenance Manual (CMM) and which revealed that 5 of the filters had a damaged mesh that required replacement. The NTSB also determined that the early-stage fatigue fractures that had occurred within the outer mesh were too small to be seen by the naked eye or when using the 5- to 7-power magnification recommended in the CMM. In addition, the ATSV filter manufacturer found that the approved bubble test method did not adequately detect early-stage fatigue cracks in the filter. We note that the February 2011 revisions to the AMM do not discuss any type of a detailed test or inspection for fatigue damage; rather, the AMM specifies an inspection for possible filter deterioration, corrosion, obvious damage, and general condition. The NTSB does not believe that such a maintenance task is likely to discover a developing fatigue problem with the filter mesh. The FAA considers its action in response to this recommendation to be complete even though we do not believe that the actions taken to date adequately address Safety Recommendation A-09-28; accordingly, this recommendation is classified CLOSED—UNACCEPTABLE ACTION.
CC# 201100227: - From J. Randolph Babbitt, Administrator: The results of the Board's investigation revealed that the ATSV on the accident airplane had been damaged by improper use and maintenance. After engine start, the damaged ATSV could not spring load fully closed. Once the engine spooled up, the damaged ATSV would have been forced open causing the air turbine starter to fail. Filter failures do not constitute a safety-of-flight issue. The deterioration and failure of the ATSV air filter was not the cause of the ATSV failure. Rather, as correctly identified in the accident report, the failure stemmed from incorrect troubleshooting and repeated application of an unapproved procedure to manually open the ATSV, which damaged the internal pin in the n1anual override button making the ATSV susceptible to inadvertent opening at high inlet pressure. Inadvertent uncommanded opening of the ATSV is an anticipated failure for which detection and mitigating procedures are in place. To mitigate possible hazards of an ATSV failure: 1. An indication of ATSV open is provided in the flight deck; and 2. Flight Crew Operations Manual procedures are in place which mitigate the hazards of starter failure and engine compartment overheat/fire. To mitigate the hazards of an engine compartment overheat or fire, § 25.1181 ensures that the engine compartment: 1. Is a designated fire zone; 2. Meets additional requirements for fire protection, flammable fluids, drainage and ventilation, shutoff means, and firewalls; 3. Provides a means for fire detection and extinguishing; and 4. Provides indication of a nacelle overheat condition to the flight deck. In the accident aircraft, all aircraft indication and fire protection systems performed as designed. This accident resulted from a chain of at least four errors on the part of the operator's maintenance and flight crew as noted by the Board. Following Boeing's recommended maintenance and operating procedures would have prevented this accident as well as limited further damage to the aircraft. Boeing reviewed and improved the Aircraft Maintenance Manual (AMM) procedure and the Maintenance Review Board (MRB) task card by clarifying the inspection for possible filter deterioration, corrosion, obvious damage, and general condition. The revised AMM procedure was issued in August 2010 and was revised again in February 2011 (enclosure 1). The revised MRB task card was issued in February 2010 (enclosure 2). Service history confirms that operators who perform the Boeing recommended maintenance for the ATSV air filter have not experienced in-service failures. Therefore, we do not believe it is appropriate to pursue a mandatory replacement interval for the ATSV air filter. I believe the Federal Aviation Administration has effectively addressed these safety recommendations, and I consider our actions complete.
The FAA replied that existing maintenance requirements for the ATSV-air filter stipulate that an operator clean the starter valve air filter at each C-check, which occurs every 15 to 18 months. As described in its response to Safety Recommendation A-09-21, the FAA reviewed the service history of the ATSV system and found that, since the issuance of AD 98-09-19, only two events have occurred involving uncommanded in-flight opening of the ATSV, including the American Airlines flight 1400 accident. The FAA found that both of these events were attributed to failed filters not serviced at the previous C-check as required. Therefore, the FAA concluded that operators who either cleaned or replaced the air filter at a C-check experienced no in-service failures and that, based on this extensive service history, the existing required C-check service interval has proven effective and appropriate. Therefore, the FAA does not plan to take the recommended action. The FAA’s review of only the service history is not an acceptable response as it ignores the findings of the NTSB’s investigation. In the letter that transmitted this recommendation to the FAA, the NTSB described that, as part of the American Airlines flight 1400 accident investigation, American Airlines had sent seven serviceable ATSV air filters to the NTSB for evaluation. Three of the seven ATSV-air filters revealed evidence of the onset of fatigue in the filter elements’ outer mesh. In addition, American Airlines sent 15 serviceable ATSV-air filters to the ATSV-air filter manufacturer for bubble testing and visual inspections, which were performed in accordance with the Component Maintenance Manual (CMM). The testing and inspections revealed that five of the filters had damaged mesh that required replacement. The NTSB also determined that the early-stage fatigue fractures within the outer mesh were too small to be seen by the naked eye or when using the 5- to 7-power magnification recommended in the CMM. In addition, the ATSV filter manufacturer found that the approved bubble test method did not adequately detect early-stage fatigue cracks in the filter. The filter inspection guidelines that are part of the C check do not require an inspection of the inner mesh because it is hidden from view and cannot be adequately inspected for evidence of fatigue. This recommendation was issued because of (1) the findings of fatigue in 3 of the 7 serviceable filters examined in detail by the NTSB and in 5 of 15 serviceable filters examined by the filter manufacturer and (2) the finding that ATSV-air filters cannot be adequately inspected, which may lead to future problems. The NTSB concluded that the existing inspection criteria for the ATSV-air filter are inadequate to detect early-stage fatigue fractures of the outer mesh of the filter element and that, because of the ATSV-air filter design, the inner mesh of the filter element cannot be inspected for evidence of fatigue. The NTSB is concerned that fatigue can be a difficult defect to detect using only service history data if, as with the ATSV-air filter, there is no periodic inspection for the development of fatigue. The current maintenance requirement is a check for blockage and/or cleanliness that does not examine for signs of fatigue. Therefore, the NTSB believes that the use of service history data is inadequate to prevent developing fatigue failures from becoming a problem. However, our investigation showed that, when the filter was inspected with a detailed examination unlikely to be done in the field, 3 of 7 (almost 50 percent) serviceable filters revealed the onset of fatigue in the outer mesh and that, of 15 serviceable units checked with the CMM test, 5 had damaged mesh and required replacement. The NTSB asks the FAA to reconsider the need for the recommended action. Pending the FAA’s completing the recommended action, Safety Recommendation A-09-28 is classified OPEN -- UNACCEPTABLE RESPONSE.
Letter Mail Controlled 8/31/2009 2:09:47 PM MC# 2090551: - From J. Randolph Babbitt, Administrator: The existing maintenance review board (MRB) report requires an operator to clean the starter valve air filter at each C-check, which occurs every 15 to 18 months. This requirement applies to both MD-80 and to DC-9 series aircraft, which have the same starter valve and filter. As indicated in our response to recommendation A-09-21, a review of the service history revealed that since the issuance of AD 98-09- 19 in June 1998, only two events were reported to Boeing involving uncommanded in-flight opening of the ATSV. Both of these events were attributed to failed filters not serviced at the previous C-check as required. Therefore, operators who either cleaned or replaced the air filter at a C-check experienced no in-service failures. Based on extensive service history, the existing required C-check service interval has proven effective and appropriate, therefore we conclude the current MRB task is adequate and does not need modification.
Strategic Plan, Performance & Accountability Reports & More
Directions to Conference Center
Web Policies & Notices
Annual Review of Aircraft