Incident Overview
On 13 June 2003, at approximately 0740 UTC, an HS125-700A aircraft (registration N125XX) was conducting a public transport flight from London Luton Airport. During the takeoff roll on Runway 26, at a speed of 60 kt, the crew heard a loud bang and the aircraft yawed to the right. The commander immediately closed both thrust levers, applied the brakes, and brought the aircraft to a stop. The right engine oil pressure warning light illuminated, and the engine fire warning bell sounded. The commander completed the engine shutdown drill. Air traffic control reported smoke emanating from the right engine. The aircraft was then taxied clear. No injuries were sustained by the two crew members or five passengers. The right engine and thrust reverser sustained severe damage.
Examination and Findings
Initial examination revealed an uncontained failure in the region of the low-pressure turbine (LPT) of the No. 2 engine. A hole was present in the engine cowling from approximately the 9 o'clock to 11 o'clock position (viewed from the front). The 1st stage LPT disc was missing; a segment equating to approximately 120° of the disc was recovered from the runway. The remainder of the disc could not be found despite extensive search. Other damage was limited to a small debris impact on the right wing flap.
Engine History
The failed engine, serial number P-80213, had accumulated 8,792 hours and 5,205 cycles since new. Approximately 107 hours and 98 cycles before the incident, it had undergone a Major Periodic Inspection (MPI) and Compressor Zone inspection at an overhaul agency in the USA. During this inspection, the LPT 1 disc and nozzle assembly were replaced. The disc was a new item from the manufacturer, while the nozzle assembly was an overhauled unit obtained from a spare parts dealer. The nozzle assembly had been overhauled by an FAR Part 145 approved repair station operating under a Designated Engineering Representative (DER) system. This repair station was not on the manufacturer's list of authorised organisations for this type of work.
The nozzle assembly (part number 3072319-11, serial number 7-01345-11066) was manufactured as a one-piece ring of 67 vanes. During overhaul, it underwent a flow check and adjustment of individual vanes to achieve a specific total nozzle area. The manufacturer (Honeywell) uses a computer programme called 'NAPOLI' to calculate the required vane deflections. This programme was developed in response to previous fatigue failures of nearly-new LPT 1 discs in the 1980s.
Service Bulletins and Repair Station Awareness
In September 1988, Honeywell issued Service Bulletin (SB) TFE731-72-3367, which outlined the potential consequences of operating with flow-adjusted LPT 1 nozzles and referred customers to authorised repair centres. This SB did not contain instructions for adjusting vanes. In July 1990, a further SB (TFE731-72-3369RWK) provided instructions for adjusting nozzle throat openings using 'NAPOLI' and Honeywell-devised tooling, but this document was proprietary and only issued to authorised repair centres. The repair station that overhauled the nozzle assembly was not authorised and was unaware of the manufacturer's requirements for vane adjustment and the 'NAPOLI' programme.
Laboratory Examination
Metallurgical examination of the recovered LPT 1 disc fragment identified four areas of high-cycle fatigue, originating alternately on opposite sides of the disc. No material defects or other anomalies were found. The fatigue cracking was consistent with operation of the disc with an incorrectly adjusted LPT 1 nozzle assembly.
Discussion and Probable Cause
In the absence of any other possible causal factors, it has to be assumed that the LPT 1 disc failed due to disadvantageous adjustment of the throat areas of individual nozzles of the A-5 assembly. This resulted in a resonant excitation of the disc, applying stresses beyond the design requirement and initiating and propagating high-cycle fatigue.