Created Monday, Nov 22nd 2021 15:46Z, last updated Monday, Nov 22nd 2021 15:46Z
An Omni Air International Boeing 767-300, registration N378AX performing flight OY-531 from Shannon (Ireland) to Kuwait City (Kuwait), was holding on taxiway C following a rejected takeoff at about 06:08L (05:08Z) when the crew requested to return to the apron due to a problem with the brakes. Ground instructed the aircraft to stop at first, then instructed the aircraft to turn left onto runway 24 and left onto taxiway A for taxiing back to the apron. While the aircraft turned onto the runway tower reported smoke coming from the landing gear and subsequently, while the aircraft was taxiing on runway 24 towards taxiway A, instructed the crew to stop and evacuate the aircraft, there was fire from the left landing gear. The crew confirmed the brakes issue was with the left main gear, stopped and initiated an evacuation via the right hand slides. Emergency services responded and extinguished the fire quickly. The aircraft however is still disabled on the runway. One passenger received minor injuries during the evacuation.

The airport advises: "Due to an incident at the airport all flights have temporarily been suspended. We are working to remove the aircraft involved from the scene of the incident so we can resume safe operations on the runway. This may take some time. We are advising passengers to contact their airline for updates. More information to follow."

The airline reported the crew rejected takeoff and was safely evacuated, no serious injuries occurred.

The aircraft had arrived earlier on flight OY-531 from Oklahoma City,OK (USA) to Shannon, had refueled and had rejected takeoff from runway 24 due to an open cockpit window at about 05:45L (04:45Z). The aircraft vacated the runway via taxiway A, requested to return to the apron and requested time to have the brakes cool down. Tower inquired whether the crew would be happy to hold on a taxiway and after affirmative reply instructed the crew to taxi via taxiways D1 and D2 onto northern taxiway C and hold there.

The Irish AAIU reported they dispatched investigators on site. The Boeing 767 was evacuated on the runway following a reported fire in a main-wheel well.

On Nov 22nd 2021 the Irish AAIU released their final report concluding the probable cause of the serious incident (the brakes fire) was:

Fire in the brake assembly during taxi, following a high-speed rejected take-off which occurred 28 minutes earlier.

The AAIU reported the commander rejected takeoff due to abnormal noise increasing in volume that made communication between the flight crew difficult. The first officer indicated the noise came from the cockpit window, the commander, pilot flying, however obviously did not hear that information and called for a rejected takeoff at 149 KIAS (Vr at 146 KIAS) and pulled the thrust levers to idle at 152 KIAS. A V1 call could not be heard on the CVR.

The AAIU analysed:

During the take-off roll, the Flight Crew became aware of a loud and increasing noise. The Co-Pilot suggested the noise may be coming from a cockpit window and was heard commenting ‘Window ok’ on one channel of the CVR recording. However, the Commander did not appear to hear the Co-Pilot and was unaware of the source of the noise. Data from the flight recorders shows that the Commander’s callout to reject the take-off occurred at a computed airspeed of 149 kts. The thrust levers were retarded and foot brakes applied two seconds later when the aircraft was travelling at 152 kts. The Aircraft Manufacturer informed the investigation that the recorded parameter ‘computed airspeed’ generally matches the IAS used by pilots in the cockpit unless there are data bus errors or air data computer malfunctions. There were no recorded errors or malfunctions recorded for the subject occurrence. Therefore, it is probable that the action to retard the thrust levers occurred above V1.

The Aircraft Manufacturer’s Flight Crew Training Manual advises Flight Crews that an open flight deck window does not warrant a high speed RTO. However, the same paragraph also notes that the noise from an open window may interfere with crew communications. The Operator’s Flight Crew Operations Manual states that above 80 kts and prior to V1 a rejected take-off should be performed if, ‘the airplane is unsafe or unable to fly’. In this occurrence the Commander, who has sole responsibility for rejected take-offs, decided not to continue because of the unidentified, increasing noise. Unfortunately, as forewarned in the Manufacturer’s Flight Crew Training Manual, the noise itself prevented effective communications between the Co-Pilot and the Commander. It is therefore probable that the Commander did not hear the Co-Pilot querying the status of the left sliding window. A few minutes after the occurrence the Co-Pilot told the other Pilots of a previous experience where a window was open during the take-off roll. This previous experience may have assisted the Co-Pilot in quickly identifying the source of the noise.

The ‘Before Start Procedure’ contained in the Aircraft Manufacturer’s FCOM requires a Flight Crew to carry out two checks to confirm that a cockpit sliding window is closed and locked; the lock lever must be in the forward position and the ‘window not closed’ decal located under the window must not be visible. Subsequent to the occurrence the Commander confirmed that the normal procedure was carried out. In addition, the Commander stated that the Flight Crew also checked the window indicator located above the window. The Aircraft Manufacturer informed the Investigation that this window indicator indicates the position of the latch cam only, and is a secondary method of verifying that the window is latched after the checks required by the FCOM procedure have been completed.

Following the rejected take-off there was a period of approximately 28 minutes during which time the Flight Crew calculated a brake cooling time, carried out checks, and communicated with both the AMC and the Operator’s engineering team. During this time the aircraft taxied from the runway to a Taxiway, then to a holding point on a second Taxiway and then back onto the runway in order to taxi back to stand. These movements involved a number of left turns and would have required the use of the aircraft’s brakes to control the aircraft’s speed. It is possible that, following the high speed rejected take-off, the use of the brakes during taxi and stopping may have further increased the temperatures of the brake system.

There were no brake temperature sensors installed on this aircraft, nor were they required to be. This meant that the Flight Crew had no means of monitoring brake temperatures. In accordance with Operator’s procedures, the Flight Crew used the Aircraft Manufacturer’s Brake Cooling Schedule to calculate a cooling time for the brakes. However, the Flight Crew underestimated the speed of the aircraft at the time the take-off was rejected. The underestimated value of 120 kts corresponds to a brake cooling time of approximately 44 minutes on the brake cooling schedule. The Flight Recorders show that the aircraft was actually travelling at approximately 146 kts when the ‘reject’ call was made which requires a longer cooling period and may be in the ‘CAUTION’ zone of the Brake Cooling Schedule.

Had the Flight Crew been aware of this, they may have adopted a more cautious approach, in particular with regards to taxiing the aircraft. The underestimation of speed, coupled with taxiing and further brake applications, within the calculated cooling period increased the risk of an adverse event such as a fire.

There are inherent risks associated with carrying out a high speed RTO, with runway excursions, loss of control, and brake fires amongst the possible outcomes. Above 80 kts a pilot has very little time for decision making. Both the Operator’s Flight Crew Operations Manual and the Federal Regulation definition of V1 state that the first action to stop the aircraft must be made by V1. It is therefore critical that the IAS is monitored closely so that if an RTO is required, the required actions can be initiated prior to V1 and the correct actions can be taken following the RTO. These actions include the calculation of the brake cooling time. However, in this occurrence the Commander had a significant distraction in the form of an unidentified, increasing noise, just before V1, a critical point in the take-off roll. This distraction may have contributed to the subsequent underestimation of the speed of the RTO.

During the cooling period following the rejected take-off, and whilst the aircraft was taxiing on RWY 24, Shannon ATC Tower advised the Flight Crew that smoke was observed to be emanating from the aircraft’s left main landing gear. A fire on the left main bogie was observed at 05.15 hrs by the AMC who immediately informed the Flight Crew and requested they evacuate the aircraft. The Co-Pilot acknowledged that the fire was on the left side and that they would evacuate to the right, to which the AMC answered ‘Affirm, affirm’. The Commander immediately directed the passengers and cabin crew to evacuate, using the exits on the right side of the aircraft. The AMC subsequently informed the AFRS and gave them clearance to attend the aircraft. Five minutes later the AFRS reported that the fire had been extinguished. The Commander then reported that all passengers and Cabin Crew of the aircraft had been evacuated and that the Flight Crew were leaving the aircraft.


An examination of the main landing gear bogie, wheels, brakes and hydraulic lines carried out by the respective Manufacturers and witnessed by the FAA, found that there was no identifiable source of the fire. The examination found that all of the brakes on the left main gear were found to be within acceptable limits. Analysis of the flight recorder data by the Aircraft Manufacturer showed that the brakes functioned as expected for the weight and speed of the aircraft at the time of the occurrence and there was no evidence of brake ‘dragging’.

In summary, following a high-speed rejected take-off, and subsequent period of taxiing, a fire started in the left main landing gear area. The fire was rapidly extinguished by the AFRS, and passengers and crew were evacuated. Subsequent examination of the wheels, brakes and flight recorder data found no identifiable cause of the fire. It is therefore probable, that the combination of a rejected take-off at high speed, combined with a lengthy period of taxiing where brakes, particularly on the left side were used, caused the brake system to overheat and a fire to ignite in the left main gear bogie.

Related NOTAMs:
A2500/19 NOTAMN
Q) EISN/QMRLC/IV/NBO/A /000/999/5242N00855W005
A) EINN B) 1908150539 C) 1908150700

A2501/19 NOTAMN
Q) EISN/QMRLC/IV/NBO/A /000/999/5242N00855W005
A) EINN B) 1908150701 C) 1908180800 EST

A2502/19 NOTAMR A2501/19
Q) EISN/QMRLC/IV/NBO/A /000/999/5242N00855W005
A) EINN B) 1908150708 C) 1908150800 EST

A2503/19 NOTAMR A2502/19
Q) EISN/QMRLC/IV/NBO/A /000/999/5242N00855W005
A) EINN B) 1908150830 C) 1908150930 EST

A2504/19 NOTAMR A2503/19
Q) EISN/QMRLC/IV/NBO/A /000/999/5242N00855W005
A) EINN B) 1908150908 C) 1908151200

Related Flight: OY531, Twitter: #OY531, Omni Air International News