N13331

On September 27, 1997, at 0849 Pacific daylight time, Continental Airlines flight 1046 (a Boeing 737-3T0, registration number N13331), a scheduled domestic 14 CFR 121 passenger/cargo flight enroute from Seattle/Tacoma International Airport, Seattle, Washington, to George Bush Intercontinental Airport, Houston, Texas, returned to Seattle/Tacoma International when the crew noted immediately after becoming airborne that large amounts of aileron and rudder input were required to maintain wings-level flight. The crew was able to land at Seattle/Tacoma International without damage to the aircraft or injuries to the airline transport pilot-in-command, first officer, 4 cabin crewmembers, or 128 passengers aboard. The aircraft was on an instrument flight rules (IFR) flight plan.

The crew reported that at liftoff, "considerable aileron and rudder input" was required to maintain straight and level flight, and that when aileron input was reduced to zero, at least 5 units of left rudder trim was required to maintain wings-level flight (according to Boeing, a minimum of 16.4 units of rudder trim is available.) Upon returning to the gate, it was noted that the right aileron remained up with the control wheel centered. Postflight troubleshooting revealed that the ABSB-4 aileron bus cable was broken and the ABSA-4 cable was frayed. The cable break and fraying occurred where the cables ride over the right wing/body joint aileron pulleys in the right wing root/main wheel well area. Maintenance personnel did not find any discrepancies with the pulleys. There was also no evidence found to indicate that the cables were, or had become, misrouted.

According to Boeing's records, the aircraft was delivered on August 11, 1986. On its NTSB incident report, Continental Airlines reported the aircraft's airframe total time as 34,633 hours. Continental further indicated on its NTSB incident report that the aircraft had 378 hours in service since its last inspection, a continuous airworthiness inspection performed on August 15, 1997, approximately 6 weeks prior to the incident. Continental reported that this inspection was a segmented "C" check, and that aileron cables were to be inspected during the "C" check segment which was performed at that time. Continental further reported to the NTSB that it was unable to determine whether or not the parted and frayed cables were the original cables installed at the time of aircraft delivery.

The ABSA-4 (frayed) and ABSB-4 (fractured) right wing aileron bus cables, which are both similarly constructed from 3/16 inch diameter, 7x19 wire rope, were sent to the NTSB Materials Laboratory in Washington, D.C., for examination. Examination of the fractured ends of the ABSB-4 cable revealed severe amounts of wear between the individual wires and strands (the NTSB metallurgist's factual report stated that this type of wear is generally referred to as internal cable wear.) On the vast majority of parted wires, wear had reduced the individual wire diameters to knife edges with little or no perceptible fracture surfaces. The only wires showing significant fractures were those from the core strand of wires; the features of these fractures were reported to be typical of overstress separations. It was estimated by the NTSB metallurgist that over 90% of the cable's total section had been removed by the internal wear, which appeared to be present along about 2 inches of the cable with some wires showing several locations of severe reduction. Severe external wear was also noted on the cable adjacent to the location of the separation; however, no wires appeared to be fractured at this wear. Energy dispersive x-ray analysis of individual wires found wire composition and tin coating consistent with MIL-W-8342, composition "A" wire rope.

In the NTSB Materials Laboratory examination, magnified optical examinations of the frayed location on the ABSA cable uncovered many of the same features as at the separation on the ABSB cable, including severe internal wear of the wires and strands. Visual examinations of the cables also disclosed several other areas of locally severe external wear, as well as several locations where the overall diameter of the cable had been reduced without damage to the exterior cable surface, which the NTSB metallurgist characterized as indicative of internal cable wear. In some locations, the cable diameter was reduced by as much as 0.03 inches (corresponding to approximately a 30% reduction in cable cross-sectional area for a nominal 3/16 inch diameter cable.)

The ABSA-4 and ABSB-4 cables were subsequently examined at Boeing's Equipment Quality Analysis (EQA) Laboratory in Renton, Washington. This examination was performed with the NTSB investigator-in-charge (IIC) along with representatives of the FAA, Boeing, and Continental Airlines in attendance. Boeing's examination of the ABSA-4 cable found wear "on the exterior of one side of the cable, the side which contacted the OD of the pulley, and not on the other side." The EQA report also stated that internal wear was also evident on this cable. The EQA report stated that on the ABSB-4 cable, "the wear occurred on the exterior of one side of the cable; the side which contacted the OD of the pulley, and not the other side." The EQA examination also noted wear on the interior of the ABSB-4 cable, between the bundles. Boeing's conclusion was: "The cables exhibited external wear which is believed to have resulted from contact with their respective pulleys. This external wear is likely the cause of the cable fraying and separation....The existence of external wear was evidenced on several portions of the cable....The internal wear was likely subsequent to the excessive external wear...."

A similar B-737 incident to the one involving Continental flight 1046 occurred at Newark, New Jersey, on March 15, 1993 (NTSB incident number NYC93IA059.) In that incident, involving a B-737-130 series aircraft also operated by Continental as flight 1659, the airplane rolled left immediately after liftoff but the pilot controlled the roll with right aileron and was able to return to Newark and land without further incident. Post-incident examinations of the left wing ABSA and ABSB aileron bus cables from that aircraft revealed that the left aileron down cable had parted in the same location (but on the left side) and manner as the parted right wing ABSB cable on Continental 1046. The NTSB determined the probable cause of the 1993 incident to be "inadequate maintenance/inspection by company maintenance personnel, the manufacturer's inadequate inspection and/or replacement procedures for the aileron cables, and subsequent failure of the 'down' aileron control cable due to wear." Based on the March 1993 Continental Airlines B-737 incident at Newark, the NTSB issued Safety Recommendations A-94-64 through A-94-66 to the FAA as follows:

A-94-64. Issue an Airworthiness Directive (AD) to operators of Boeing 727 and 737 airplanes requiring periodic inspection of the aileron cables for both internal and external wear, and for broken wires, with particular attention to the area of the cable contacting the pulleys. The inspection should include releasing cable tension to better detect cable wear and wire breakage and establishing a maximum allowable reduction in cable diameter where pulley contact occurs. Based on the inspections, develop specific flight hour intervals for replacement of the cables.

A-94-65. Require that the Boeing Company examine the consequences of a 737-100 aileron cable failure, and provide appropriate flightcrew operational guidance for the best landing configuration in the event of such a failure.

A-94-66. Conduct a comprehensive study to determine the frequency of spoiler, rudder, and aileron cable failures on airplanes weighing 12,500 pounds or greater. Where the study reveals flight control inspection procedures to be inadequate, require appropriate revisions to those inspection procedures and/or issue Airworthiness Directives to mandate service life limits to assure greater reliability of those control cables.

In response to recommendations A-94-64 and A-94-65, the FAA responded on August 3, 1995, that it had conducted flight simulations in which it determined that the B-737-100 was controllable with a broken aileron cable, that no additional flightcrew guidance was necessary with regard to this condition, and that it did not consider an AD requiring periodic inspections in addition to the existing regular maintenance inspections to be necessary. The FAA also reported it reviewed the flight control cable failure rates for the B-737 fleet and found the failure rate for aileron wing cables to be 8 x 10-8 per flight hour, with the failure rate for aileron body cables being 6.4 x 10-8 per flight hour. The FAA stated that "this system performance further supports the conclusion that a broken aileron cable is an improbable occurrence." Based on the FAA response, the NTSB classified recommendation A-94-65 "Closed-Acceptable Action" on November 20, 1995.

In further response to recommendation A-94-64, the FAA stated to the NTSB on June 28, 1996, that it had examined the maintenance manuals for Boeing airplanes that utilize wire cable operated flight controls. The FAA stated that it found inconsistencies among some of the procedures, and that the best practices possible were not reflected consistently in all manuals. The FAA stated that as a result, Boeing had eliminated the inconsistencies and developed one standard inspection procedure for the Boeing family of airplanes. The new standard procedure, which the FAA stated was to be performed every 12 to 18 months, involved rubbing a cloth along the cable length to catch on broken cable strands and lock-to-lock control wheel rotation to expose cable hidden on the pulleys. Additionally, instructions for checking cable diameter wear were pro...