Resized to 1280 pixels wide. No cropping. Responsive WebP copies resized without cropping.
Geographic context
London Luton, Runway 25 departure end
Athens
The flight departed London Luton for Athens. The Luton marker uses the Runway 07 threshold coordinates shown on the AAIB chart, at the departure end of Runway 25. Athens is shown as a city marker. Basemap: Natural Earth, public domain.Read supporting evidence
The crew prepared a full-length departure, then accepted Intersection A. The takeoff settings in the flight management computer were not correctly verified after that change. Maximum recorded thrust was 82.1%, against the 85.2% calculated for the intersection departure, and V2 remained unchanged from the initial entry. The aircraft lifted off 162 m before the end of the paved runway.
Read the full incident Follow what happened, from the flight to the investigation findings.
What investigators concluded
The AAIB attributed the long takeoff roll and slow climb to insufficient power after the crew changed the departure point and failed to verify the flight management computer entries. It identified a potential loss of stopping or obstacle-clearance margins if thrust had been lost during takeoff.
709 words / about 4 minutes to read. The official report for this event is short, so this account is shorter than most and covers only what the report documents.
Preparing the Athens flight
On 22 April 2025, G-CRUX was preparing for an early morning scheduled passenger flight from London Luton to Athens. The Boeing 737-8 carried six crew and 162 passengers. Built in 2018, it had two CFM International LEAP-1B27G05 turbofan engines. The AAIB recorded the serious incident at 0630 UTC. No injuries or aircraft damage were reported.
The pilots initially prepared the aircraft to use the full length of Runway 25. Their takeoff performance data therefore reflected that departure point. The commander was 61, held an Airline Transport Pilot's Licence and had 10,000 flying hours, including 3,000 on type.
The AAIB's stated information source was an Aircraft Accident Report Form submitted by the pilot. Its investigation also examined flight data monitoring, ATC observations and CCTV.
The departure plan changed before takeoff. The pilots checked the performance calculations to see whether they could use Intersection A. During the pushback clearance, they told air traffic control that they could accept an intersection departure. As they approached the holding point, they recalled checking the Onboard Performance Tool and believed they had updated the V speeds in the Flight Management Computer.
The Onboard Performance Tool calculated takeoff and landing performance for the aircraft, runway and conditions. The V speeds were defined airspeeds used during takeoff, climb, approach and landing.
The shorter starting point changed the required performance. Intersection A had a declared takeoff run available of 1,771 m and a calculated thrust requirement of 85.2%. For the full runway length, the corresponding figures were 2,116 m and 81.4%. Takeoff run available, abbreviated TORA, meant the runway length available for the aircraft's ground run during takeoff.
A full-length Runway 25 departure involved entering through holding point A1, turning left to the runway end and making a 180 degree turn where space was available.
Flight data monitoring subsequently showed that the takeoff settings had failed to follow the change in departure point. The V2 speed remained unchanged between the initial entry during flight preparation and takeoff. Maximum achieved thrust was 82.1%, below the 85.2% calculated for the intersection departure. The AAIB found that the crew had failed to verify that the takeoff performance was correctly entered in the flight management computer.
The achieved thrust was slightly above the 81.4% calculated for a full-length departure. The AAIB could not determine whether the autothrottle had set a slightly higher value than required or whether the pilot had set the thrust manually. The recorded maximum established the power used during takeoff, while the method of setting it remained unresolved.
The aircraft used a long ground run before becoming airborne. Flight data placed liftoff 162 m from the end of the paved surface. Figure 1 in the investigation showed that position by overlaying the flight data on a satellite image of the runway.
The climb after liftoff was very shallow. G-CRUX crossed the end of the runway at 13 ft above ground level. Air traffic control reported that the aircraft appeared on radar climbing through 900 ft above ground at approximately 0.8 nautical miles from the airfield. At that height, an aircraft would normally have appeared much closer to the airport.
CCTV from Luton air traffic control provided a further view of the departure. The AAIB included two images: the upper image showed G-CRUX, while the lower image showed a representative departure a few minutes later. Together with the flight data overlay and the radar observations, the images documented the shallow departure.
The AAIB concluded that accepting a different departure point without verifying the performance entries led to a lower power setting than required, a long takeoff roll and a slow climb. Its explanation of the performance calculations covered both stopping distance and obstacle clearance. The calculations provided for enough runway to stop after a failure at a critical speed and for minimum clearance from obstacles during the climb.
The investigation identified a potential consequence if the aircraft had lost thrust during takeoff. Following a decision to reject, it might have been unable to stop on the remaining paved surface. Following a decision to continue, it might have been unable to achieve the minimum height clearances during departure.
The operator reviewed its procedures and issued new guidance to crews. The AAIB published its completed correspondence investigation as AAIB-31017 in Bulletin 7/2026.
Intersection A provided 1,771 m of takeoff run and required 85.2% thrust. The full-length calculation used 2,116 m and required 81.4%. The crew failed to verify the revised performance entries.
What did the pilots recall checking before takeoff?
They recalled checking the Onboard Performance Tool and believed they had updated the V speeds. Flight data monitoring subsequently showed that V2 had stayed unchanged from the initial entry.
The AAIB could not determine whether the 82.1% thrust resulted from the autothrottle setting a slightly higher value than required or from a manual setting by the pilot.
Matching dossiers for Insufficient takeoff thrust: No other dossier in this collection.
RiskStopping and obstacle-clearance exposure
The AAIB found that a loss of thrust could have left insufficient paved runway for a rejected takeoff or insufficient height clearance if the takeoff continued.
Matching dossiers for Stopping and obstacle-clearance exposure: No other dossier in this collection.
Read alongside
These comparisons connect specific research topics. Shared circumstances do not establish identical causes. Evidence from both investigations is provided below.
Both investigations examined departures in which takeoff performance was compromised. At Luton, the AAIB found incorrect performance entries after a change to the departure point. In the Air Florida 90 accident, the NTSB identified failures to use engine anti-ice, takeoff with snow or ice on the airfoils, and failure to reject after anomalous engine instrument readings. The investigations identified different mechanisms affecting the takeoff.
Explore the records below for additional context. They open on the publisher's site and are not all included in our report search. Source reuse terms apply.
Completed AAIB correspondence investigation in Bulletin 7/2026, including the flight data overlay, CCTV comparison and takeoff performance calculations.
Sources and transparency
The AAIB completed its correspondence investigation in Bulletin 7/2026, published on 9 July 2026.
Source limitations
The AAIB investigation drew on a pilot-submitted accident report form, flight data monitoring, ATC observations and CCTV.
AAIB-31017: Boeing 737-8, G-CRUX
Air Accidents Investigation Branch Final correspondence investigation report, AAIB Bulletin 7/2026, published 9 July 2026
Official report PDF, hosted by assets.publishing.service.gov.uk
Kevin Murphy writes the incident accounts on AirlineIncidents.com, out of a long interest in aviation safety and the lessons each investigation leaves behind.
Prepared by
AirlineIncidents editorial desk
Report citation check
Independent expert review
Not completed
Citation checks compare supporting text with the indexed report pages. They do not constitute independent expert review. How these pages are prepared.
Revision history
: Published sourced incident account, timeline, photographs and investigation findings.
.jpg){kind=link}