History of the Flight
The instructor and student were conducting takeoff and landing practice on Runway 23 at Durham Tees Valley Airport. The student's first touch-and-go was normal, and the second touchdown was also normal. While applying power for the next takeoff, the instructor felt a slight wobble, took control, and closed the throttle. The right wing dropped, and the aircraft slewed to the right, coming to rest on the runway. After shutting down the engine and electrics, both occupants vacated the aircraft.
Aircraft Examination
Examination revealed that the right main landing gear leg had separated from its wing spar attachment. All eight attachment fasteners—four upper screws and four lower bolts—had failed. Three upper screws were recovered and had failed in overload at the screw head. One lower bolt was recovered with a corroded head and a fracture at its thread, but the fracture surface was smeared, preventing determination of its failure mechanism. Two additional bolt or screw ends were recovered, which had failed at the thread just above the nut. Both exhibited fracture surfaces consistent with fatigue failure. By process of elimination, at least one of these was from a lower attachment bolt. One nut was not 'in safety' as no threads protruded beneath it.
Maintenance History
The aircraft had accumulated 10,634.4 hours at the time of the accident. Its last annual inspection was completed 35 hours prior, on 24 February 2012. The manufacturer's maintenance schedule recommends inspecting the attachment bolts every 100 hours for condition and security. The maintenance engineer stated he inspected the bolts during the annual and found no anomalies, but no torque check was performed as none was required. About seven to ten years earlier, the right gear leg was replaced following a failure of the upper lug securing the torque links; the attachment bolts were reportedly not disturbed since then.
The maintenance engineer believed the failures resulted from repetitive heavy landings, but the aircraft owner stated that no heavy landings had been reported to the flying school.
Information from the Aircraft Manufacturer
The manufacturer searched the FAA Service Difficulty Reporting database and found six reports of fixed-gear PA-28 landing gear attachment bolt failures (including shear and corrosion) over 18 years, but none mentioned fatigue. Two reports recommended torque checks every 100 hours. The manufacturer was asked to consider adding a torque check to the maintenance schedule, but responded that the data did not support changing the instructions for continued airworthiness.
Analysis
The separation occurred after a normal landing. All eight attachment fasteners failed; three upper screws failed in overload, while two recovered bolt ends showed fatigue. At least one of these was from a lower attachment bolt. The crack features indicated a combined tensile and bending load mode, which could result from loose bolts. The bolts had reportedly not been disturbed for over seven years; loosening could have occurred due to bolt stretch from heavy landings or from nuts backing off (one nut was found not 'in safety'). A visual inspection 35 hours prior would not have detected slight loosening or cracking above the nut, but a torque check could have. The manufacturer declined to mandate a torque check based on fleet-wide data.