Delivery

Non-compliance by the crew of the technical procedures, configuration, speed and bank angle of the aircraft required for the completion of the initial turn of the Instrument Approach Procedure n°4 published in the AIP Ecuador, to Latacunga Airport, a failure that placed the aircraft outside of the protected area (published pattern), leading to high elevation mountainous terrain.
Contributing factors:
- Ignorance of the crew of the area which was under the approach path.
- Lack of documentation and procedures of the airline that govern the conduct of flights to non-scheduled and special airports.

Engine fire/failure for unknown reasons.

For reasons that could not be determined, the aircraft commenced a slight left angle of bank and drifted left after lift-off at a height from which the pilot was unable to recover prior to striking trees to the left of the runway.

The following findings were reported:
- The pilot was appropriately licensed, rated and experienced for the series of flights undertaken.
- The aeroplane had a valid airworthiness certificate and had been released to service.
- There was nothing (other than the item in 3.5) to suggest that the aeroplane was operating abnormally on the final flight.
- The aeroplane was being operated at 14 000 feet pressure altitude without supplementary oxygen as required by CAR 91.209 and 91.533.
- The left front fuel filler orifice was observed to be leaking fuel before departure.
- There was no attempt made to further investigate or correct this fuel leak and the pilot stated that it would stop once he departed.
- On most other aircraft this would be true, once the fuel level dropped away from the filler orifice and was no longer affected by aerodynamic suction.
- On the 750XL, the fuel system design was such that the front tanks were continuously topped up.
- The fuel loss would continue until all fuel in the rear tanks and the ferry system was consumed.
- The front fuel caps are thus critical items to be checked before flight.
- The fuel quantity uplifted at Hilo indicated that the problem had existed on the previous leg with a loss rate of up to 125 litres (33.2 US gallons) per hour.
- A comparison of the uplift figure with the expected consumption on the previous leg should have provided sufficient warning to the pilot that a problem existed.
- The existence of the problem could have been detected on the final flight by the shortened top-up intervals and by comparing fuel used by the engine with fuel remaining.
- Cumulative delays, especially including the longer than normal final refuelling time, probably influenced the pilot’s decision to depart without further checking the reason for the fuel leak or the apparent discrepancy between fuel figures.
- Cumulative fatigue, circadian rhythm and hypoxia were probably significant factors in the pilot’s failure to detect the fuel problem in flight, in time to make a safe return.
- By the time the pilot announced that he had a fuel problem, the only course of action open to him was ditching the aeroplane.
- The search and rescue facilities were activated appropriately, and had the potential to effect a successful rescue.
- The water entry impact on ditching was reasonably severe and probably incapacitated the pilot before he could vacate the cockpit.

It is believe that the crew was forced to make an emergency landing following a general failure of the navigation instruments in flight.

On vacating FL140, the aircraft's climb rate was so erratic at 140 KIAS that it seems likely that by then, the aircraft had already gathered sufficient ice to seriously affect its performance. If all the
turbocharger inlets had become partially blocked, then manually selecting both engines to alternate air induction should have introduced warmer air into the turbochargers and restored power. The description of engine operation in the Superstar manual states:
'If manifold pressure continues to decrease after opening the manual alternate air, it is an indication that turbocharger inlets are still restricted and the engine may become normally aspirated through the automatic alternate air door located below the induction air filter'.
Normal aspiration reduces the manifold pressure to ambient or less and at FL140 the ambient pressure is about 17.6 inches which is less than half the climb rated manifold pressure. That might explain the inability to climb above FL 160 but it would also have deprived the pilot of pressurisation. There was no change in his voice consistent with donning an oxygen mask so he may not have lost pressurisation completely. Nevertheless, since he lost control at around FL160 and 110 KIAS, and because the aircraft initially turned to the right, a combination of airframe icing and asymmetric power loss seem the most likely explanation for the sustained loss of control. The split in the EDP diaphragm which almost certainly occurred during this flight may have contributed to an asymmetric power problem. Alternatively, the pilot might have become mildly hypoxic and decided to begin an emergency descent. If so, he did not declare an emergency at the time he started to descend, although he did utter the word "SCOTTISH" after control was lost, so he was conscious even if his mental abilities had been impaired by hypoxia. On balance, the tone of his voice and his initial failure to respond to ATC messages suggested that the descent was begun through loss of control rather than a deliberate act followed by loss of control. At the time of the accident the aircraft had been flying below the freezing level (8,000 feet) for about five minutes and much of the airframe and induction system ice may have melted. The would-be rescuers would have taken at least five minutes to reach the crash site and so the fact that none of them reported seeing or treading on any ice was not surprising. Witness and propeller evidence indicated that power had been restored on at least one engine but there seemed to be insufficient power to climb out of Glen Lyon. The aircraft was out of control when it crashed at low speed from a sharp turning manoeuvre. Before this manoeuvre the pilot may have had partial control, albeit with a power problem which prevented him from climbing, and he finally lost control totally when he attempted to turn around within the confines of the Glen. He had no choice but to attempt the turn since, had he not turned, he would have flown into the side of the hill above the crash site.

It was reported that the aircraft was unstable on final approach and not properly aligned with the runway centerline. This caused the aircraft to land hard in 'crab' configuration, causing the left main gear to collapse upon touchdown.