AAIB Bulletin: 6/2007 G-OOSI EW/A2006/12/02

Full document text

46 © Crown copyrght 2007 AAIB Bulletin: 6/2007 G-OOSI EW/A2006/12/02 INCIDENT Aircraft Type and Registration: Cessna 404 Ttan, G-OOSI No & Type of Engines: 2 Contnental GTSIO-520-M pston engnes Year of Manufacture: 98 Date & Time (UTC): 6 December 2006 at approxmately 90 hrs Location: En-route from San Pedro Arport, Cape Verde Islands, to Dakar Arport, Senegal Type of Flight: Prvate Persons on Board: Crew -  Passengers -  Injuries: Crew - None Passengers - None Nature of Damage: None Commander’s Licence: Commercal Plot’s Lcence wth Instrument Ratng Commander’s Age: 2 years Commander’s Flying Experience: 504 hours (of whch 5 were on type) Last 90 days - 85 hours Last 28 days - 5 hours Information Source: Arcraft Accdent Report Form submtted by the plot, a company report and enqures by the AAIB Introduction The ncdent occurred over nternatonal waters and, n agreement wth the Portuguese Authortes the AAIB, representng the State of Regstraton, took responsblty for the nvestgaton. Synopsis When flying above 10,000 ft, the commander did not use contnuous oxygen. He was probably sufferng from hypoxa when he attempted to adjust hs engne controls and ths resulted n vbraton and an uncontrolled descent. He recovered full control at a lower alttude and made a successful dverson. The operatng company s mplementng changes to ther operatng procedures to prevent a smlar occurrence. History of the flight The arcraft had been operatng n the area snce the end of November 2006 and was beng used for map survey operatons. Durng ths tme there were no reports of any significant aircraft unserviceability. The aircraft was usually operated on survey flights with two people on board: a plot (the commander) and a camera operator (the passenger). Prior to the incident flight, the commander had noted that hs ntended route from San Pedro Arport to Dakar Airport included a portion with a minimum notified alttude of FL95. He therefore planned a cruse alttude of FL20 but ntended to descend below FL00 when wthn the Dakar FIR. The arcraft oxygen pressure indicator was showing 1,200 psi before the flight which 47 © Crown copyrght 2007 AAIB Bulletin: 6/2007 G-OOSI EW/A2006/12/02 would allow just over four hours oxygen use for two people at FL220. The commander subsequently stated that the arcraft occupants would need oxygen for approxmately 45 mnutes. There were three plot-style oxygen masks for the two occupants. Each had a rubber restranng strap and a mcrophone. However, the commander was aware that the mcrophone on at least one of the masks was “crackly” and he ddn’t ntend to use t for rado transmssons. Addtonally, the mask provded to the passenger had a broken strap held together wth adhesve tape. The passenger stated that a week before this flight, he was advsed (by a dfferent commander) that he should only use oxygen “as and when he needed t” to avod any possblty of dranng the system. However, the operator’s representatve stated that no such nstructon was necessary for the incident flight and the commander stated that he had ntended the passenger to use “as much oxygen as they felt necessary”. The operator also stated that the oxygen system was servceable and had been used on recent flights. There had been no noticeable leakage from the system and there was more than sufficient oxygen remaining for the planned flight. For the takeoff at 855 hrs, the passenger was seated mmedately behnd the commander n a seat facng aft. As the arcraft clmbed through 0,000 ft, the passenger was nstructed to move to the rear of the cabn, to optmse the arcraft’s CG poston, and to actvate the oxygen system. When seated at the rear of the cabn, the oxygen and ntercom leads were too short to allow hm to connect both smultaneously. Durng the clmb, the commander used hs oxygen mask ntermttently, albet more frequently as alttude ncreased. When not usng the mask, he placed t on hs lap. Once level at FL20 n the cruse, he engaged the autoplot. Durng the subsequent cruse, the passenger had the mpresson that the commander’s voce was “a lttle slurred” when transmttng to ATC. When asked, the commander confirmed that he was using oxygen and shortly afterwards he was seen to be adjustng the engne controls. The commander subsequently confirmed that he took off hs oxygen mask to adjust the controls n response to a perceved engne problem. Soon after, the passenger heard a change n engne nose and was aware of vbraton together wth the sensaton that he was beng pushed nto hs seat. Attempts to contact the commander by ntercom were unsuccessful and, wth the arcraft descendng at hgh speed and n a spral, the passenger called twce for the commander to transmt a ‘MAYDAY’. The commander responded to the second call and declared an emergency. The arcraft was stll descendng and, at around 5,000 feet alttude the passenger opened the emergency escape hatch n preparaton for a possble sea dtchng. However, the commander then reganed control of the arcraft and once t was fully stable, he requested a dverson to Amlcar Cabral Arport on the Cape Verde Islands. A safe landng was made there at 2005 hrs. Post landing actions After landng, the commander checked the arcraft and considered that it was fully serviceable. He was confident that the arcraft had remaned wthn normal operatng parameters durng the ncdent and that no negatve ‘g’ manoeuvres had occurred. On reflection, he considered that he had started sufferng from hypoxa durng the clmb. The perceved engne problem probably resulted from hm not correctly adjustng the engne controls at alttude. The commander contacted hs company engneer n UK to advse hm of the ncdent and also carred out uneventful engne ground runs the next day before 48 © Crown copyrght 2007