ASN logo
Accident description
Last updated: 19 January 2017
Date:Tuesday 19 January 1971
Time:13:55 UTC
Type:Silhouette image of generic COMT model; specific model in this crash may look slightly different
de Havilland DH-106 Comet 3B
Operator:Royal Aircraft Establishment - RAE
Registration: XP915
C/n / msn: 6100
First flight: 1954-07-19 (16 years 6 months)
Engines: 4 Rolls-Royce Avon 524
Crew:Fatalities: 0 / Occupants: 7
Passengers:Fatalities: 0 / Occupants: 0
Total:Fatalities: 0 / Occupants: 7
Airplane damage: Substantial
Airplane fate: Repaired
Location:Bedford-Thurleigh Airport (   United Kingdom)
Phase: Standing (STD)
Departure airport:Bedford-Thurleigh Airport (EGSV), United Kingdom
Destination airport:Bedford-Thurleigh Airport (EGSV), United Kingdom
A Comet aircraft, XP915, operated by the Blind Landing Experimental Unit (BLEU) of RAE Bedford, was about to make a training flight from Bedford-Thurleigh Airport. It was in the takeoff position on runway 27, on the centreline, about 500 feet in from the end of the runway, awaiting takeoff clearance.
At the same time, a Trident aircraft, G-AWZA, operated by British European Airways (BEA) was one of a number of civil aircraft flying in the Bedford aerodrome circuit and making use of the Instrument Landing System (ILS) of runway 27, for training. These aircraft were operating under radar surveillance and were being positioned and sequenced by the Bedford approach controller as required for the various exercises on which they were engaged. The local weather was reported as visibility 4 to 6 nautical miles, cloud base varying from 1,500 to 2,000 feet and surface wind varying from 190░ (T) to 210░ (T) at speeds from 10 to 15 knots. There were occasional flurries of light rain. Trident G-AWZA was engaged in conversion training. At the time of the accident the particular manoeuvre being practised, in simulated Category III weather conditions, was designed to familiarise crew members with various failure situations during an automatic approach to land using ILS. A screen in front of the left-hand pilot was being used to simulate Instrument Meteorological Conditions (IMC) and an aperture in this screen was being opened and closed at the relevant heights to simulate the appropriate visual reference limitations.
Prior to the approach on which the accident occurred four exercises had been successfully completed. These had involved automatic approaches down to a height of about 12 feet, each followed by an automatic overshoot; the last three exercises had included a simulated engine failure during the overshoot phase. In the fifth exercise the overshoot was originally planned to be initiated at a height of about 70 feet, the visibility shutter in the screen in front of P1 being closed at about this height, thus triggering P1 into calling for overshoot action as defined in the relevant BEA crew procedures. Almost simultaneously with overshoot initiation there was to be an auto-pilot 'failure'. The training captain had briefed P1, P3 and the flight engineer on the details of this intended exercise.
When the aircraft was inbound at the outer marker, approximately 3 nautical miles out on the approach, the Trident pilot reported that he was on finals for a low approach and overshoot. The controller acknowledged this by informing the pilots that the Comet aircraft was back-tracking on the runway and that they were therefore cleared only to make an overshoot; the Trident pilot replied with the word 'understood' . At about 900 feet the training captain introduced a 'failure' by switching off one of the roll channels of the auto-pilot system; both P1 and P3 noted it and took account of its consequential effects. In this case the primary consequence was to increase the decision - height (DH) from the 12 feet which is permitted with a fully triplicated autopilot system to a height of 100 feet which is appropriate for an auto-pilot system operating only in duplex. It was raining but forward slant visibility was good and the training captain could see the Comet lined up on the runway near to the threshold. His own vision was not impaired by the screens in front of P1 and as the aircraft seemed to be correctly aligned both laterally and vertically he was satisfied to continue the approach. Having established the Comet's position on the runway he decided not to continue the exercise down to the originally intended 70 feet (R) but to overshoot at 100 feet (R). He did not consider it necessary to inform the crew of this change and decided that he could still get some value from the exercise by having P1 overshoot from 100 feet (R). At 500 feet (R), P3 called '500 - RADIO' there by indicating to P 1 and P2 that subsequent height calls would be given by reference to the radio altimeters. At 300 feet (R) P3 called 'AUTOLAND - 100 feet' to indicate that the status of the relevant equipment was satisfactory for the approach to be continued to a landing on automatics subject to P1 making his decision by 100 feet (R). P1 repeated '100 feet' to acknowledge this call from P3. At about 190 feet (R) P1 went 'head up' preparatory to seeking external visual reference and at 150 feet the screen shutter was opened thus giving him the appropriate 'visibility' . At 130 feet (R) P3 called '30 ABOVE'; almost immediately P1 called 'LAND' and the Training captain (P2) said subsequently that immediately following this he had called 'OVERSHOOT' and opened the throttles fully. P1 said that when P2 called 'OVERSHOOT' he went `head down' and reverted immediately to his flight instruments checking speed and attitude and noted that the aircraft had rotated to about 9░ nose-up attitude ; however all crew members said that they felt the aircraft continue to 'sink'. At 70 feet the flight engineer closed the screen shutter as he had been previously briefed to do. The aircraft continued to descend and at a radio altimeter reading which P1 said subsequently was about 20 feet there was an impact; P1 disengaged the auto-pilot by pressing his thumb button, called 'AUTO-PILOT OUT', took over manual control and climbed the aircraft back to circuit height. The pilots of another Trident aircraft in the circuit then made a visual check of the damage to G-AWZA and established that it was confined to the port inner flap after which the aircraft made a successful landing at Bedford using a 'flaps up' technique as a precautionary measure. The Comet crew were uncertain of the extent of the damage to their aircraft and after shutting down the engines, left the aircraft with the assistance of the Fire Service who were quickly on the scene in response to ATC instructions. The aircraft was towed away and the runway cleared of wreckage so that the Trident could land. There was no fire in either aircraft.

Probable Cause:

CAUSE: "The cause of the accident was that the training captain did not ensure that the overshoot was made at the intended height of 100 feet (R)."



Add your photo of this accident or aircraft


languages: languages