What happened
On 2 March 2003, a De Havilland Canada DHC-8-311, registration G-BRYJ, was performing a scheduled passenger flight from Aberdeen to Manchester. During the climb through FL170, the crew observed that the aircraft failed to level at its assigned altitude. Despite the autopilot being engaged, the pitch attitude remained nose-up, and the crew noticed intermittent 'mistrim' warning messages.
Upon disengaging the autopilot, the pilots found that the control column offered very limited movement. Both pilots applied significant manual force to the controls, but the aircraft continued to climb to FL175. To address the suspected elevator jam, the commander activated the pitch disconnect handle. This action successfully separated the left and right elevator control circuits, allowing the commander to regain control of the aircraft, though with reduced authority. The crew declared a MAYDAY and diverted to Edinburgh. The aircraft subsequently completed a successful flapless landing on Runway 24 with no injuries to the 4 crew members or 29 passengers.
The investigation
The AAIB examined the aircraft's flight data and conducted a physical inspection of the control surfaces. Analysis of the flight data recorder revealed that while the left elevator responded normally to inputs, the right elevator remained fixed at a constant angle during the initial climb.
Post-flight inspections of the elevator spring and trim tab hinge bearings uncovered a substance with the consistency of 'wallpaper paste'. Laboratory testing of this residue showed a very low glycol content and a freezing point of approximately -1.1°C. The investigation also noted that the aircraft had been exposed to rain prior to the flight and had been treated with Type II anti-icing fluids in previous weeks. Investigators were able to replicate a similar restriction in a laboratory setting by using dry ice to cool the bearings and applying water.
Findings
- The flight control difficulties were caused by a restriction of the right elevator spring tab.
- This restriction was likely caused by the freezing of re-hydrated anti-icing fluid residue around the hinge bearings.
- Accumulated residues from previous applications of Type II fluids had re-hydrated due to moisture/rain and subsequently froze.
- The low glycol content of the residue resulted in a freezing point high enough to cause ice formation in the observed atmospheric conditions.