AAIB Bulletin: 9/2009 G-BOLF EW/C2008/10/03

outer right wing, and a twisting failure of the rear of the fuselage just forward of the empennage. This evidence correlated well with the witness evidence that the aircraft first impacted the sea with its right wing, before rapidly swinging to the right, inverting and sinking. The engine cowlings, right landing gear, cockpit windscreen and cockpit doors were torn off either during the impact or whilst on the sea bed prior to recovery of the aircraft. These items were not found. The propeller showed good evidence that it was not being driven by the engine when it impacted the sea. The wing flaps had been fully deployed, the fuel was selected to the left tank, the throttle was fully open, the fuel mixture was near the fully rich position, the carburettor heat was selected on, the fuel primer control was ‘in-and-locked’, the electric fuel pump, landing light, anti-collision light and pitot tube heater were selected off and the engine ignition was selected off. The fuel tank drains were found to be firmly closed. There was no evidence of an airborne fire. A detailed examination of the flying control system, fuel system and engine operating systems did not reveal any evidence of pre-impact failure, disconnection or restriction. The engine was taken to an overhaul facility for examination. external and internal examination showed no evidence of failure, disconnect or seizure. The general internal condition of the engine was found to be very good and consistent with the hours that the engine had operated since overhaul. During the disassembly of the engine it was found that the engine-driven mechanical fuel pump mechanism was stuck in an unusual position. During the disassembly of the fuel pump it became free to operate. A detailed examination of the pump’s internal mechanism revealed possible evidence of stiction on the central spindle. The central spindle was found to be slightly bent in the area of the stiction marks, which could account for why the pump mechanism was found stuck in an unusual position when removed from the engine. The fuel pump had been fitted to the engine as a new item approximately 1,500 operating hours prior to the accident and there had been no reported problems associated with the pump. Two other mechanical fuel pumps of the same type were internally examined for comparison and both had similar, but less pronounced stiction marks on the central spindle, but neither spindle was bent. it was not possible to determine how or when the spindle of the accident mechanical fuel pump became bent. The electric fuel pump was tested and, even after immersion in the sea for about 24 hours, it functioned satisfactorily. Carburettor icing The Met office aftercast for the robin hood’s Bay area at the time of the accident gave the air temperature, dew point and humidity from the surface to 3,000 feet in bands of 500 ft. When these figures were plotted on the Civil Aviation Authority (CAA) Carburettor Icing Prediction Chart it gave a prognosis that serious carburettor icing could occur at any power setting between 1,000 and 3,000 feet amsl (Figure 1). 64 © Crown copyright 2009 AAIB Bulletin: 9/2009 G-BOLF EW/C2008/10/03 Maintenance Documentation showed that the aircraft had sufficient fuel on board for the intended flight and that it had been maintained in accordance with cAA regulations. A 50-hour maintenance check was completed on the 13 october 2008. Pilot training The pilot had obtained his licence after 45 hours of training. he made his first solo flight after 17.5 hours and he was considered by his instructor to have made satisfactory progress throughout his training. The examiner who conducted the pilot’s PPL Skills Test on 15 september 2008 considered that the test was flown to a good standard. Investigation flight During the investigation the AAiB conducted a flight in a PA-38-112, following the route of the accident flight as closely as possible. This showed that when the MAYDAY call was transmitted, it was theoretically possible for the pilot to achieve a forced landing on land. however, the nearest land would have been behind the aircraft and hidden by the wing. it would not, therefore, have been immediately visible to the pilot. Pathology The post-mortem examination revealed that the pilot had survived the impact with the water and had died from drowning. This evidence, and the fact that his body was found outside the aircraft, suggests that he was able to extricate himself from the aircraft after it impacted the water, but had then drowned. Figure 1 CAA carburettor icing probability chart -20 Serious icing - any power 100% Relative Humidity 80% 60% 40% 20% -10 0 +10 +20 +30 +40 -20 -10 0 Dewpoint ºC +10 +20 +30 World Wide Approximate upper limits of dewpoint NW Europe Moderate icing - cruise power Serious icing - descent power Serious icing - descent power Light icing - cruise or descent power FOG/CLOUD Courtesy of CAA SafetySense Leaflet No 14 2000’ 3000’ 1000’ 65 © Crown copyright 2009 AAIB Bulletin: 9/2009 G-BOLF EW/C2008/10/03 None of the witnesses to the accident reported seeing the pilot on the surface of the water. he was not wearing any form of flotation aid. There was no evidence of significant pre-existing natural disease which could have caused or contributed to the accident and toxicology revealed no evidence of alcohol or drugs in the pilot’s body. Discussion It was apparent that the aircraft had suffered a loss of engine power. Given that the conditions at the time were conducive to carburettor icing at any power setting, the possibility of loss of engine power due to carburettor icing could not be ruled out, but this does not appear consistent with the witness report of hearing an engine increasing to high power, then decreasing and increasing again. it could not be determined how or when the spindle of the fuel pump had become bent. it was not possible to positively determine the reason for the loss of engine power from the available evidence. When the engine failed, the pilot could have achieved a forced landing on land, but the location of the nearest land would not have been obvious to him as it was behind the aircraft and hidden by the wing. he may have been initially occupied with making a MAYDAY call and trying to restore engine power. By the time he had completed these actions the aircraft may have been beyond gliding distance of land. it appears that the aircraft stalled just prior to entering the water, causing the right wing to drop into the water and the aircraft to cartwheel into the sea. cAA safety sense leaflet No 21 contains advice on ditching an aircraft. it points out that the main cause of death after ditching is drowning, and that in many cases the deceased person either was not wearing a life jacket, or did not have one readily available. This accident reinforces the importance of wearing a suitable life jacket to maximise the chances of survival whenever there is a risk of ditching.