What happened
On 2 April 2000, a Boeing 747-200, registration ZS-SAL, was taxiing toward runway 9R at London Heathrow Airport for a scheduled passenger flight to Johannesburg. The aircraft was moving at speeds not exceeding 9 knots when the crew experienced a loud bang and a physical jolt. Simultaneously, anti-skid warning lights for the right wing landing gear illuminated on the flight deck. A passenger observed a tyre detaching from the aircraft and rolling across the grass area between the taxiway and the runway.
The aircraft stopped at the holding point for inspection. It was discovered that the inboard rim of the No 2 wheel on the left wing landing gear had detached, causing the tyre to separate from the hub. While the primary failure occurred on the left side, fragments from the failed rim were thrown by the tyre, striking the right main gear door, the right wing landing gear, and its folding door. This impact also severed a wiring conduit containing the anti-skid unit wiring. Following the replacement of the affected wheels, the aircraft was towed to the gate, and passengers disembarked without further incident.
The investigation
AAIB investigators examined the failed wheel assembly and the tyre. The tyre itself was found to be in good condition with no unusual damage. The wheel assembly consisted of an inboard and outboard half bolted together. Metallurgical analysis of the fractured rim revealed that the failure was caused by a high cycle tension fatigue crack in the bead radius of the rim. This crack had progressed to a point where the rim fractured into three separate pieces.
Further investigation focused on the manufacturing and maintenance history of the component. The manufacturer had originally used a process called shot peening to create compressive stresses in the bead radius to prevent fatigue. However, investigators found that a polishing operation performed during maintenance had removed approximately 9 inches of this protective layer. This polishing also introduced surface scores. The investigation determined that an embryonic fatigue crack had likely existed prior to this polishing and remained present in the newly smoothed surface.
Findings
- The failure originated from a fatigue crack in the bead radius of the inboard wheel rim.
- A maintenance polishing operation had removed the protective compressive layer created by shot peening.
- The polishing process also introduced surface irregularities that likely contributed to the crack's progression.
- The failed component was an older standard of wheel design, which lacked the improved fatigue resistance found in modern versions.