TK1975 - Accident: THY A21N at Dublin on Oct 18th 2024

A THY Turkish Airlines Airbus A321-200N, registration TC-LTL performing flight TK-1975 from Istanbul (Turkey) to Dublin (Ireland), landed on Dublin's runway 28L at 09:18L (08:18Z) but balked the landing, the tail contacted the runway, went around and positioned for another approach. The aircraft landed on runway 28L without further incident about 15 minutes later.

The return flight TK-1976 was cancelled.

A listener on frequency reported the crew requested a runway inspection and an inspection of the tail after arrival.

The aircraft is still on the ground in Dublin 11 days later.

On Oct 31st 2024 the Irish AAIU reported they have opened an investigation into the occurrence with this brief: "Fuselage hit the runway while aborting the landing."

On May 19th 2026 the AAIU released their synoptic report concluding the probable causes of the accident were:

The pitch attitude of the Aircraft exceeded the tail strike pitch limit during a go-around manoeuvre.

Contributory Cause(s)

- Startle and surprise as a result of the urgent, non-standard command to go-around and hearing the ‘PITCH’ aural warning.

- The application of full back stick following the command to go-around, combined with the pitch-up effects of TOGA thrust.

- The continued application of full back stick, which may have been due to an initial pitching down of the aircraft, in conjunction with a perceived need to take-off again as soon as possible.

The AAIU analysed:

Introduction

The Aircraft took-off from Istanbul International Airport (LTFM) at approximately 04:10 hrs and was approaching to land on RWY 28L at Dublin Airport (EIDW), following a four-hour flight.

The Flight Crew’s recollection was that the Aircraft began to float during the flare and drift to the left of the runway centreline. The Commander reported that if a go-around had not been initiated, the Aircraft would have departed the side of the runway.

The Flight Crew recalled that the main wheels made contact with the ground during the go-around manoeuvre. Main wheel contact following the initiation of a go-around close to the ground would not be unusual due to the inertia of a landing aircraft. However, the recorded data from the flight indicates that in this case, the go-around was initiated after the Aircraft had touched down on both main landing gears, following an initial touch down on the lefthand main landing gear and subsequent decompression/lifting. Therefore, the go-around manoeuvre performed was actually a rejected landing, although for consistency, it is referred to as a go-around throughout this Report.

The data indicates that the tail strike occurred approximately three seconds after the commencement of the go-around and just before the Aircraft became airborne again. As highlighted in the Aircraft Manufacturer’s Flight Operations Briefing Notes on the subject of ‘Landing Techniques Bounce Recovery – Rejected Landing’, a rejected landing is a challenging manoeuvre, ‘decided and conducted in an unanticipated and unprepared manner’. It is possible that due to the perceived float and the smooth nature of the initial touchdown (1.15 g), the Flight Crew was unaware that the touchdown had occurred, and their focus on the deviation to the left of the runway centreline in addition to the dynamic nature of the goaround, followed by a realisation that a tail strike had occurred, may have adversely affected their recollection of the timeline of the second touchdown.

Tail Strikes

When an aircraft is rotated beyond a critical angle, on or near the ground, a tail strike can occur. The geometry of the A320 aircraft variant is such that a tail strike will occur at an aircraft pitch angle of 11.7o (wings level) when the main landing gear shock absorbers (oleos) are fully compressed. This angle increases to 13.5o when the shock absorbers are fully extended. On the A321 (ceo/neo) variant, which is longer, the angles reduce to 9.7o and 11.2o, with an associated increased risk of a tail strike, as evidenced by the statistical analysis conducted by the Aircraft Manufacturer. IATA’s 2024 and 2025 annual safety reports cited tail strikes as the most frequently occurring non-fatal accident category.

There are several features on the A321 neo aircraft to mitigate the risk of a tail strike occurring during the landing phase, including a pitch rate limitation function; a pitch limit indicator on the PFD; and a ‘PITCH PITCH’ aural warning. In addition, during a normal landing, there are several nose-down effects, including ground effect; ground spoiler compensation; ground reaction due to main wheel contact with the runway; engine thrust reduction; and autobraking. However, once a go-around manoeuvre is initiated, and TOGA thrust is selected, the increase in engine power will produce a pitch-up effect.

In addition, auto-braking, which resultsin a pitch-down effect, is deactivated. Furthermore, to ensure full elevator authority during a go-around manoeuvre, the pitch rate limitation function is deactivated one second after TOGA thrust is selected, and the pitch limit indicator and ‘PITCH PITCH’ aural warning are inhibited. These aspects would suggest that there is an increased risk of a tail strike during a go-around/rejected landing on an A321 neo aircraft, when compared to a normal landing phase. However, the Aircraft Manufacturer’s statistical analysis indicates that on the A321 neo aircraft, 50% of tail strike events occurred on landing and 50% occurred during a go-around or a rejected landing. The Investigation notes that there is no indication in the available data of tail strikes having occurred on the A321 neo aircraft during a normal take-off.

Sequence of Events

Introduction

The following sequence of events is based on the flight data and cockpit voice recording from the occurrence flight, and on the reports from the Flight Crew and interview details. The occurrence was dynamic in nature and the recorded parameters are not all recorded at the same sampling rate; therefore, exact timings are not possible. It should be noted that sidestick inputs during the landing phase are also dynamic in nature, particularly in gusty conditions.

Flare Phase

The data indicates that at approximately 50 ft above the runway, following a stabilised approach, the Aircraft was on the correct lateral and vertical flight path and was in the landing configuration. The Aircraft was at the target speed for the approach (VAPP) and no excessive flight parameter deviation was recorded. The ND-displayed wind conditions, as reported by the Flight Crew, of one seven zero degrees at zero nine knots, when the Aircraft was at minimums, would have resulted in a crosswind component of approximately 8.5 kt and a tailwind component of approximately 3.1 kt, which were within the Aircraft’s operational limits at the time of touchdown.

When the Aircraft was at approximately 20 ft RA, the PF initiated a flare, with control inputs that were predominantly nose-up, including a full back application of the sidestick. This full back application may have been in response to an increase in the tailwind component, which occurred around that time and may have presented visually as an increase in ground closure rate.

The nose-up inputs increased the Aircraft’s pitch angle, which had been at 1o , to a pitch angle of 5.5o at initial touchdown. At approximately 10 ft RA, the thrust levers were retarded to the idle detent as normal. The 4.5o pitch increase in the flare in this case was only slightly above the typical increase of 4o outlined in the FCTM, while the pitch angle of 5.5o at initial touchdown was within the range of most common pitch angles on the A321 neo aircraft variant. The Manufacturer described the manoeuvre as a ‘slight over-flare’ and explained that the absolute value of pitch angles during the flare were normal, but that the timing and dynamics of the pitch-up significantly reduced the rate of descent to 50 ft/min prior to touchdown (the normal rate of descent in the flare is 120 ft/min to 300 ft/min).

Initial Touchdown

The data indicates that the Aircraft’s roll angle was -2.5o (left wing down) just before landing, which was consistent with control inputs for a crosswind from the left and resulted in the Aircraft initially and ‘smoothly’ touching down on the left main landing gear (+1.15 g). The drift angle was +5.5o (Aircraft nose toward the left of the track) and the Aircraft began to deviate to the left of the runway centreline just after the initial touchdown. The CVR indicates that around the same time, the Commander expressed a degree of concern, which may have been due to the deviation to the left. It is possible that due to its smooth nature, the Flight Crew was unaware that the initial touchdown had occurred. However, a right rudder pedal input of up to ⅓ full deflection was applied by the PF following the subsequent decompression/lifting of the left main landing gear, which was consistent with a normal de-crab action to align the Aircraft with the runway centreline and resulted in the drift angle decreasing from +5.5o to +1.5o. The Aircraft’s ground spoilers began to extend shortly after the initial touchdown of the left main landing gear in accordance with the system logic (PLD), to accelerate normal ground spoiler extension when both main landing gears are on the ground.

Second Touchdown

The PF’s sidestick input, which was approximately ¼ of full back (nose-up) at the time of initial touchdown, was increased to full back stick before the second touchdown, while at the same time, the Aircraft’s pitch angle had begun to reduce. The ground spoilers began to deploy after the initial touchdown (PLD) and it is possible that, as suggested by the Aircraft Manufacturer, the rearward stick position may have been in response to a lift decrease due to spoiler extension. The pitch angle reduction may also have been due to ground effect (as noted in the FCTM and in the Operator’s Recurrent Instructor Guide), decreasing engine thrust, or a combination of both.

The Aircraft touched down again on its left main landing gear, less than two seconds after the first touchdown. The Aircraft’s pitch angle had reduced from +5.5o at initial touchdown to +4o at the second touchdown. The right main landing gear touched down approximately one second later with a vertical load factor of +1.57 g, which is below the vertical acceleration criteria used by maintenance to declare a hard landing (i.e. the second touchdown was not hard). The Aircraft Manufacturer considered that the second touchdown was as a result of the partial extension of the ground spoilers (PLD), which caused the lift to decrease.

The data indicates that a brief sidestick roll input of up to approximately ⅕ full left deflection was made by the PF and the Aircraft’s roll angle reached -4.5o (left wing down) around the time of the second touchdown. The data also indicates that the crosswind from the left began to significantly decrease around the time of the initial touchdown and it is therefore likely that the decrease in crosswind, combined with the sidestick roll input, which would have been applied to counteract the crosswind, resulted in the increased left-wing down attitude at this stage, and the deviation to the left of the runway centreline continued.

Following the touch down of both main landing gears, the ground spoilers began to fully extend. Due to the position of the spoilers on the aircraft’s wings, spoiler extension results in a pitch-up effect. To compensate for this effect, an automatic nose-down elevator order is commanded once all three SECs validate ground spoiler function. The data indicates that ground spoiler compensation commenced when the ground spoilers started to fully extend.

Just after the second touchdown, the PF released the sidestick from the full-back (nose-up) position but maintained it in a rearward position. When both main landing gears were on the ground, the auto brake system activated. This would have generated a further pitch-down effect; however, in this case, the pitch angle began to increase from 4o. The Aircraft also continued to deviate to the left of the runway centreline. The Aircraft Manufacturer considered that the pitch angle increase was mainly due to the rearward position of the sidestick. The FCTM states that after touchdown, ‘the pilot must fly the nosewheel smoothly, but without delay, on to the runway, and be ready to counteract any residual pitch up effect of the ground spoilers’. The data indicates that the nose landing gear did not contact the runway at any stage during the occurrence.

Normally on touchdown, the PF will retard the thrust levers to reverse idle and further if required. In this case, less than two seconds after the second touchdown, the right thrust lever was briefly moved to reverse idle and the left thrust lever was briefly moved to between idle and reverse idle. The PF’s sidestick was in an approximately ⅖ of the full back position at this stage and the Aircraft’s pitch angle was increasing towards 8o. The Aircraft continued to deviate to the left of the runway centreline. The ‘NORMAL PROCEDURES – STANDARD OPERATING PROCEDURES – GO-AROUND’ section of the FCTM states that ‘The PF must not initiate a go-around after the selection of the thrust reversers’. This is because reverse thrust results in a decelerative force, in addition to that provided by autobraking. If the thrust levers are moved forward to initiate a go-around after the selection of reverse thrust, significant time is required for the reversers to stow and for the engines to spool up to allow the Aircraft to accelerate to the correct speed and perform a successful take-off before the end of the runway. However, the position of the thrust levers did not result in any significant adverse effect on the Aircraft behaviour in this case, as the thrust reversers had not fully deployed.

Rejected Landing/Go-Around

When the pitch angle reached 8o, the ‘PITCH PITCH’ aural warning sounded. Co-incident with the annunciation of the second ‘PITCH’, the Commander called for a go-around using urgent, non-standard phraseology. Following the go-around command, the PF moved the sidestick to the full back position; however, the pitch angle began to decrease from 8o and continued to decrease over the next two seconds approximately.

Approximately one second after the ‘PITCH PITCH’ aural warning, and after the PF’s sidestick had been moved to the full back position, the thrust levers were moved to TOGA. Selection of TOGA thrust caused the ground spoilers to begin to retract. It also inhibited any further ‘PITCH PITCH’ aural warnings and deactivated the pitch limit indicator. Initially, the pitch angle continued to decrease and reached 4.5o approximately two seconds after the PF’s sidestick had been moved to the full back position, and approximately one second after the thrust levers were moved to the TOGA position. Throughout this period, the PF’s sidestick continued to be held in the full back position. The Aircraft Manufacturer stated that ‘the ground spoilers retraction added a nose-down effect partially compensated by the removal of the ground spoiler compensation elevator order’. Ground effect may also have been a factor in the pitch angle reduction, despite the full-back (nose-up) position of the PF’s sidestick.

The landing phase pitch rate limitation, regardless of the sidestick position, is automatically deactivated one second after the selection of TOGA thrust to ensure full elevator authority during a go-around manoeuvre, and when the pitch angle had decreased to 4.5o, which it reached approximately one second after TOGA thrust was selected, it began to increase again.

The tail strike pitch rate limitation had been removed at this stage. The nose-up pitch moment induced by the engine spool was likely a factor in the pitch angle increase. The FCTM highlights that the PF may need to counteract the pitch-up effect of engine thrust during a go-around.

The FCTM also highlights that during a go-around, the engines can take a few seconds to spool up; therefore, an increase in engine power and the associated pitch-up effect would not be instantaneous.

The PF’s sidestick remained in the full-back position for approximately three seconds after the go-around command, before being progressively released to a third of the full-back position, over the next one second approximately. This progressive release may have been in response to the increasing pitch angle and rate. However, the pitch angle continued to increase, with the rate of increase reaching more than 6o per second. The tail strike occurred around this time, at a pitch angle of approximately 13o, just before the Aircraft became airborne again at a CAS of 112 kt, which was significantly less than the VAPP speed of 129 kt. The Manufacturer’s prescribed procedure is to maintain pitch until VAPP is reached, before rotating and monitoring the pitch.

The data indicates that the PF’s sidestick was then briefly moved to almost the full-back stick position, before being progressively released again. The pitch angle continued to increase during the take-off sequence and reached +15.6o, before the Aircraft became airborne and increased to 17.5o after take-off.

Deviation to the Left of the Runway Centreline

Both Flight Crew members recalled that the Aircraft drifted to the left of the runway centreline during the flare. The distance between the outer wheels on each main landing gear is 8.97 m and the runway surface damage due to the tail strike was approximately 9 m to the left of the runway centreline and approximately parallel to it. RWY 28L at EIDW is 45 m wide; therefore, at the time of the tail strike, the left main landing gear was approximately 13.5 m to the left of the runway centreline52 and approximately 9 m from the left-hand edge of the runway.

The 9 m distance remaining does not indicate an imminent risk of a lateral runway excursion at that stage. However, when the nose of the Aircraft was elevated during the landing phase and pointing to the left, with the Aircraft decelerating from a speed of 129 kt (239 km per hour/66 m per second), the Commander’s perspective from the left-hand side of the cockpit likely indicated otherwise, as evidenced by the urgent, non-standard command to go-around.

CCTV footage and ASMGCS data showed that the Aircraft continued to deviate to the left of the runway centreline during the subsequent take-off manoeuvre.

Startle and Surprise

The fact that the PF had moved the thrust levers towards reverse following the second touchdown indicates that the PF’s intention was to complete the landing. It is likely that the urgent, non-standard command to go-around at a critical phase of the flight, combined with the ‘PITCH’ aural warning, resulted in the PF experiencing a significant degree of startle and surprise. The Commander may not have appreciated that the thrust levers had been moved towards reverse, and the PF’s likely startle and surprise may have been compounded by the command to go around which conflicted with the procedural requirement not to initiate a goaround after the selection of the thrust reversers.

A study on ‘Startle effect Management’ carried out by the NLR of The Netherlands noted that ‘On the flight deck the disruption caused by the startle reflex can have detrimental effects, particularly when the startle is elicited when the pilot is performing flight essential tasks’. The study also noted that:

‘The effects of surprise are in part comparable to those of startle. Physiological responses to surprise include increased heart rate and blood pressure, cognitive responses include confusion and loss of situational awareness, and may involve the inability to remember the current operating procedures’.

Sidestick Inputs

In response to the go-around command, the PF moved his sidestick to the full back position (in addition to moving the thrust levers to the TOGA position) and held it there for approximately three seconds. The Aircraft Manufacturer’s article on ‘Preventing Tail strike During Go-around near the Ground’, notes that the ‘application of full back stick by the flight crew’ was reported ‘in many of the tail strikes during go-around near the ground events’ and that it was ‘a common contributor to these events as this action led to a high rate of rotation’.

The ‘GO-AROUND NEAR THE GROUND PROCEDURE’ as contained in the FCTM, states that ‘the PF should avoid excessive rotation rate to prevent a tail strike’ and that ‘[…] the flight crew may need to maintain pitch until the aircraft speed increases to VAPP’ (rather than immediately attempting to take-off again).

The PF’s recollection was that following the application of TOGA thrust, he applied a pitch-up (stick back) manoeuvre, ‘possibly more aggressive than usual to prevent a probable nose wheel impact’. However, it is possible that the adverse effects of startle and surprise, in addition to a perceived need to take-off again as soon as possible, contributed to the full back input. When the PF’s sidestick was moved to the full back position, the Aircraft’s pitch angle decreased for the next two seconds (i.e. opposite to the applied input) and this may also have contributed to the stick being held in the full back position.

The data indicates that the Commander’s SIDE STICK PRIORITY pushbutton was not pressed. However, when the Aircraft’s pitch angle began to increase, following its initial decrease, it did so rapidly, with a pitch rate reaching more than 6o per second (the normal rotation rate is 3o per second). This would have limited the time available to assimilate the information and react appropriately.

During training sessions, the Operator is required, as per Regulation (EU) 965/2012, to include go-arounds with all (both) engines operating during every EBT (Evidence-Based Training) module, and in this regard, the syllabi for the Operator’s Operator Proficiency Check (OPC)/Licence Proficiency Check (LPC) for the Airbus A320 aircraft series contain, amongst several items, a ‘DISCONTINUED APPROACH, GO AROUND OR REJECTED LANDING’ item. These checks were last completed by the Commander and First Officer in August 2024 and April 2024, respectively.

Cockpit Environment and Flight Crew Actions

The cockpit voice recording for the final approach phase of the flight does not contain any extraneous conversation and indicates that approach and landing checklist items were carried out appropriately.

The cockpit voice recording and the flight data indicate that following the tail strike and before the FO passed control to the Commander, the Commander asked the FO to inform ATC of the go-around, which the FO did; it is normally the PM who communicates with ATC.

Approximately 22 seconds after the transmission to ATC, the FO passed control to the Commander, and approximately six seconds later, the Commander requested the FO to engage the autopilot and AP1 was selected. The data indicates that from time 08:13:45, which was just after the Aircraft became airborne again, until the Commander took control at approximately 08:14:15 (30 seconds later), when the Aircraft was at approximately 1,200 ft RA, no sidestick inputs were made by either the FO or the Commander and the Aircraft continued to deviate to the left of the runway centreline as the Aircraft climbed during the go-around.

The data indicates that the flaps were moved to position 3 just after take-off during the goaround manoeuvre, as required, before being retracted further during the climb. No verbalisations of these actions were recorded on the CVR.

The data also indicates that the landing gear was not retracted until the Aircraft reached approximately 4,000 ft, whereas it should have been retracted once the Aircraft was established in the go-around with a positive rate of climb.

This sequence of events may have been due to distraction caused by the Flight Crew’s realisation that a tail strike had occurred, the subsequent communications with ATC regarding the Aircraft’s ability to maintain the standard missed approach and the correct ATC frequency to use, and their in-flight discussion of the occurrence.

Following the subject occurrence, several safety actions were taken by the Operator. These actions were reviewed by the Operator following the subsequent occurrence at EIDW and additional actions were taken. As a result of the safety actions, no Safety Recommendation is made to the Operator.

...

Related Flight: TK1975, Turkish Airlines News