ENG711(A) - Fuel selector incorrectly installed

Initial Report

Following work to the undercarriage, the fuel selector handle was refitted incorrectly. It was fitted 90º out of position and with one bolt missing. When the selector handle was set to the main rear tank it was in fact drawing fuel from the left-wing tank, when the selector was set to the right-wing tank, the fuel selector valve was closed. If both bolts had been fitted it would not have been possible to fit the selector handle incorrectly. Following completion of repairs to the undercarriage and a taxi test by the Approved Maintenance Organisation (AMO), the aircraft was released for flight. The same afternoon I planned to undertake a short local check flight. The engine first started ok but only ran for a short time. Thereafter It would start when primed but would not run for more than a few seconds. I called one of the AMO engineers to look at the problem; he discovered that the fuel-selector handle had been fitted incorrectly and one of the fixing bolts was missing. At first, he thought the handle had been installed in reverse and, indeed, with the selector set to the off position the engine ran. However, following further checking he realised it had been installed at 90º to the correct position. I was lucky that the left wing-tank had all but run dry during the taxiing runs in the morning, otherwise it would have done so shortly after I took off. With the fuel selector set to the main rear-tank (which was full) but in fact pulling fuel from the left wing-tank, when that tank ran dry the engine would have stopped and I would have tried to select the right tank as an alternative, but this would have resulted in turning the fuel off and a forced landing.

comments

CAA Comment

CHIRP Comment

This report was filed to CHIRP as a GA report but it is also worthy of publication in the Air Transport FEEDBACK. The Human Factors issues in this report all lined up to create an extremely dangerous situation. Fortunately, the left wing-tank running dry on taxi saved the day. The pilot filed an MOR but the AMO did not. Refitting or replacing components in the incorrect orientation is not new of course, and there is a danger that good aircraft design to make installation fool proof is making us lower our guard against such errors. The overriding question is, even if the component was orientated correctly and a bolt was left over, why did that not raise a big red warning flag? We all know of situations away from work where the odd fastener is left over after carefully erecting Swedish flat-pack furniture, but this was a component critical to the safe and correct operation of the aircraft.

In commercial air transport, would this issue be less likely because larger organisations are able to provide sophisticated risk mitigation, large training budgets and a permanent Quality/Safety presence? Large organisations benefit from a Human Factors advantage: that of mentoring, coaching and advising – what one might call “Good” Peer Pressure. If you are a certifying engineer with thirty licenced colleagues, then you have thirty people to ask for assistance if needed. Equally thirty people can say “you are wrong” before you make the mistake. GA should be using the latest revision of the AMM and recording the ATA Chapter and Section reference, just as Air Transport have done for a number of years. One difficulty in GA though is that the AMM is often not as sophisticated as that of an Air Transport aircraft. Dimensions, limits, fit and tolerances will be there, but how to change a component may be left to the experience and judgement of the engineer. In GA and some corporate jets in Air Transport, there is no Structural Repair Manual (SMM). The OEM provides this information at a considerable cost, on an aircraft-by-aircraft (Serial Number) basis.

Those trained in Maintenance Error Investigation know not to assume anything. Considering maintenance error in a different section of the industry than our own may be full of assumptions until we consider the working environment of the person who made the error. Perhaps some Human Factors are more common in one part of the industry compared with another. An Aircraft Maintenance Licence often leads to supervisory status but there is no exam module for man-management. Now that HF training has largely moved to Computer Based Training, is our knowledge really being refreshed? Has Safety Culture become stale? Safety Culture should be more than something only the Authority and Training focuses on – it is for all of us to support and aim to improve. Human Factors is for all of us to consider all and every day, not just on recurrent training day or on quiet days when there is time, but also on the busy days when operational, management, time and adverse peer pressures plus distractions and multi-tasking all present competing challenges.