Alaska Airlines Flight 261

Information
Alaska Airlines Flight 261 was a flight from Licenciado Gustavo Díaz Ordaz International Airport in Puerto Vallarta, Jalisco, Mexico, to Seattle–Tacoma International Airport in Seattle, Washington, United States, with an intermediate stop at San Francisco International Airport in San Francisco, California. On January 31, 2000, the aircraft operating the route, a McDonnell Douglas MD-83, crashed into the Pacific Ocean roughly 2.7 miles (4.3 km; 2.3 nmi) north of Anacapa Island, California, following a catastrophic loss of pitch control. The accident killed all 88 on board: two pilots, three cabin crew members, and 83 passengers.

Pilots
The pilots consisted of Captain Edward (Ted) Thompson, 53, who had 17,750 flight hours, and had more than 4,000 hours experience flying MD-80s. First Officer William (Bill) Tansky, 57, had 8,140 total flight hours, including about 8,060 hours as first officer in the MD-80.

Background
Alaska 261 departed from Puerto Vallarta at 13:37 PST (21:37 UTC), and climbed to its intended cruising altitude of flight level 310 (31,000 feet or 9,400 meters). The plane was scheduled to land at San Francisco International Airport. Some time before 15:49 (23:49 UTC), the flight crew contacted the airline's dispatch and maintenance control facilities in SeaTac, Washington, on a company radio frequency shared with operations and maintenance facilities at Los Angeles International Airport (LAX), to discuss a jammed horizontal stabilizer and a possible diversion to LAX. The jammed stabilizer prevented operation of the trim system, which would normally make slight adjustments to the flight control surfaces to keep the plane stable in flight. At their cruising altitude and speed, the position of the jammed stabilizer required the pilots to pull on their yokes with approximately 10 pounds (44 N) of force to keep level. Neither the flight crew, nor company maintenance, could determine the cause of the jam. Repeated attempts to overcome the jam with the primary and alternate trim systems were unsuccessful.

During this time, the flight crew had several discussions with the company dispatcher about whether to divert to Los Angeles, or continue on as planned to San Francisco. Ultimately the pilots chose to divert. Later, the NTSB found that while "the flight crew's decision to divert the flight to Los Angeles [...] was prudent and appropriate", "Alaska Airlines dispatch personnel appear to have attempted to influence the flight crew to continue to San Francisco [...] instead of diverting to Los Angeles". Cockpit voice recorder (CVR) transcripts indicate that the dispatcher was concerned about the effect on the schedule ("flow"), should the flight divert.

At 16:09 (00:09 UTC), the flight crew successfully unjammed the horizontal stabilizer with the primary trim system. However, upon being freed, it quickly moved to an extreme "nose-down" position, forcing the aircraft into an almost vertical nosedive. The plane dropped from about 31,500 ft (9,600 m) to between 23,000 and 24,000 ft (7,000 and 7,300 m) in around 80 seconds. Both pilots struggled together to regain control of the aircraft, and only by pulling with 130 to 140 pounds (580 to 620 N) on the controls did the flight crew stop the 6,000 ft/min (1,800 m/min) descent of the aircraft and stabilize the MD-83 at approximately 24,400 ft (7,400 m).

Alaska 261 informed air traffic control of their control problems. After the flight crew stated their intention to land at LAX, ATC asked whether they wanted to proceed to a lower altitude in preparation for approach. The captain replied: "I need to get down to about ten, change my configuration, make sure I can control the jet and I'd like to do that out here over the bay if I may". Later, during the public hearings into the accident, the request by the pilot not to overfly populated areas was mentioned. During this time, the flight crew considered, and rejected, any further attempts to correct the runaway trim. They descended to a lower altitude and started to configure the aircraft for landing at LAX.

Beginning at 16:19 (00:19 UTC), the CVR recorded the sounds of at least four distinct "thumps" followed 17 seconds later by an "extremely loud noise", as the overstrained jackscrew assembly failed completely and the jackscrew separated from the acme nut holding it in place. The aircraft rapidly pitched over into a dive. The crippled aircraft had been given a block altitude, and several aircraft in the vicinity had been alerted by ATC to maintain visual contact with the stricken jet. These immediately contacted the controller; one pilot radioed: "that plane has just started to do a big huge plunge"; another reported: "Yes sir, ah, I concur he is, uh, definitely in a nose down, uh, position descending quite rapidly." ATC then tried to contact the plane. The crew of a SkyWest airliner reported: "He's, uh, definitely out of control." Although the CVR captured the co-pilot saying "mayday", no radio communications were received from the flight crew during the final event. The CVR transcript reveals the pilots' continuous attempts for the duration of the dive to regain control of the aircraft. At one point, unable to raise the nose, they attempted to fly the aircraft upside-down in an effort to maintain control. However, the aircraft was far beyond recovery; it descended inverted and nose-down about 18,000 feet (5,500 m) in 81 seconds (151 mph; 243 km/h). A few seconds before 16:22 (00:22 UTC), Flight 261 impacted the Pacific Ocean at high speed about 14 miles (23 kilometers; 12 nautical miles) offshore, between the coastal city of Port Hueneme, California, and Anacapa Island. At this time, pilots from aircraft flying in the same area reported in, with one SkyWest Airlines pilot saying: "and he's just hit the water". Another reported: "Ah, yes sir he ah, he ah, hit the water. He's ah down."

The aircraft was destroyed by the impact forces, and all occupants on board were killed by blunt force trauma.

Investigation
Using side-scan sonar, remotely operated vehicles, and a commercial fishing trawler, workers recovered about 85% of the fuselage (including the tail section) and a majority of the wing components. In addition, both engines, as well as the flight data recorder (FDR) and CVR were retrieved. All wreckage recovered from the crash site was unloaded at the Seabee's Naval Construction Battalion Center Port Hueneme, California, for examination and documentation by NTSB investigators. Both the horizontal stabilizer trim system jackscrew (also referred to as "acme screw") and the corresponding acme nut, which the jackscrew turns through, were found. The jackscrew was constructed from case-hardened steel and is 22 inches (56 cm) long and 1.5 inches (3.8 cm) in diameter. The acme nut was constructed from a softer copper alloy containing aluminum, nickel, and bronze. As the jackscrew rotates, it moves up or down through the (fixed) acme nut, and this linear motion moves the horizontal stabilizer for the trim system. Upon subsequent examination, the jackscrew was found to have metallic filaments wrapped around it, which were later determined to be the remains of the acme-nut thread.

Later analysis estimated that 90% of the thread in the acme nut had already worn away previously, and that it had finally stripped out during the flight while en route to San Francisco. Once the thread had failed, the horizontal stabilizer assembly was then subjected to aerodynamic forces that it was not designed to withstand, leading to complete failure of the overstressed stabilizer assembly. Based on the time since the last inspection of the jackscrew assembly, the NTSB determined that the acme-nut thread had deteriorated at 0.012 inches (0.30 mm) per 1000 flight‑hours, much faster than the expected wear of 0.001 inches (0.025 mm) per 1000 flight‑hours. Over the course of the investigation, the NTSB considered a number of potential reasons for the substantial amount of deterioration of the nut thread on the jackscrew assembly, including the substitution by Alaska Airlines (with the approval of the aircraft manufacturer McDonnell Douglas) of Aeroshell 33 grease instead of the previously approved lubricant, Mobilgrease 28. The use of Aeroshell 33 was found not to be a factor in this accident. Insufficient lubrication of the components was also considered as a reason for the wear. Examination of the jackscrew and acme nut revealed that no effective lubrication was present on these components at the time of the accident. Ultimately, the lack of lubrication of the acme-nut thread and the resultant excessive wear were determined to be the direct causes of the accident.