Novair A21N at Billund on Feb 8th 2019, hard landing and tail strike

Last Update: June 28, 2019 / 13:47:47 GMT/Zulu time

Bookmark this article

Photo of Novair SE-RKA, Airbus A321-Neo — Novair SE-RKA, Airbus A321-Neo (Photo credit: jbarcena / Flickr / License: CC by-nd)

Incident Facts

Date of incident
Feb 8, 2019

Classification
Accident

Cause
Hard landing

Airline
Novair

Flight number
1I-158

Departure
Arrecife, Spain

Destination
Billund, Denmark

Aircraft Registration
SE-RKA

Aircraft Type
Airbus A321-Neo

ICAO Type Designator
A21N

Airport
Billund Airport, Billund

Airport ICAO Code
EKBI

A Novair Airbus A321-200N, registration SE-RKA performing flight 1I-158 from Arrecife,CI (Spain) to Billund (Denmark) with 205 passengers and 7 crew, landed on Billund's runway 27 in gusting crosswind at 19:08L (18:08Z) but touched down hard and struck its tail onto the runway surface. The aircraft rolled out without further incident.

Sweden's Haverikommission (SHK) reported while landing in crosswind conditions the aircraft made a hard landing, the aircraft's rear part hit the runway surface (tail strike). There were no injuries, the aircraft however received structural damage. The occurrence was rated an accident and is being investigated.

Late Jun 27th 2019 the SHK released their bulletin releasing following conclusion:

The technical and training measures put in place by the aircraft manufacturer have been effective in reducing and keeping the Airbus A321 tailstrike rate low on the global fleet.

As described in the operator’s FCTM, it is likely that a tailstrike is the result of a combination of factors.

The AIB considers the contributing factors to this specific tailstrike to be:

- Gusting and strong crosswinds in dark night.

- Moderate turbulence and downdrafts on short final to runway 27 at EKBI.

- Too high sink rate, just prior reaching the flare height (destabilization).

- No flight crew calls and/or decision on going around on short final or during the bounce.

- Two hard landings.

- Inappropriate thrust lever management.

- Inappropriate pitch control during the bounce resulting in a pitch angle increasing beyond the critical angle.

The SHK reported the aircraft touched down, bounced and touched down hard a second time and suffered the tail strike while landing in dark night and instrument meteorologic conditions.

The captain (54, ATPL, 15,846 hours total, 7,600 hours on type) was pilot flying, the first officer (38, CPL, 6,748 hours total, 5,300 hours on type) was pilot monitoring.

The SHK described the sequence of events:

The aircraft was configured for landing, with flap position 3, stabilized, and the autopilot engaged and the autothrust engaged and activated in SPEED mode.

At approximately 700 ft RH, the flight crew got visual contact with the runway.

The Calibrated Air Speed (CAS) from 500 ft RH down to initial touchdown varied between approximately 160 kts CAS and 145 kts CAS. The aircraft remained stabilized on both localizer and glidepath down to 100 ft RH.

On short final, the first officer informed the commander of a displayed crosswind of 36 kts.

At approximately 190 ft RH, the commander disconnected the autopilot and flew the final approach manually with the autothrust engaged and activated in SPEED mode.

At 100 ft RH, the first officer made a standard operating call on aircraft pitch attitude: Pitch 2.5.

From approximately 100 ft RH until 35 ft RH, the vertical speed increased from approximately 700 ft/minute (min) to approximately 1100 ft/min.

At approximately 25 ft RH, the commander used an abrupt and progressive aft sidestick input and pulled back on the sidestick to full aft deflection to stop the rate of descent.

The synthetic voice calls TWENTY, RETARD and the initial touchdown occurred almost simultaneously.

Before recorded activation of both main landing gear squat switches (weight on wheels - WOW), the commander advanced the thrust levers forward of the climb detent position.

The aircraft made an initial touchdown at approximately 148 kts CAS, with a recorded pitch angle of 5.63°, a rate of decent of approximately 750 ft/min, and a vertical acceleration of 2.59 g. Upon initial touch down, the ground spoilers did not extend, and the autothrust deactivated.

The aircraft lifted off the ground and up to 12 ft RH.

During the bounce, the commander repeatedly voiced his surprise on the lack of flare.

In the air, the engines spooled up, and the CAS increased to approximately 156 kts. Two synthetic voice calls on excessive pitch (PITCH), and one thrust lever retarding synthetic voice call (RETARD) sounded.

The commander retarded the thrust levers to flight idle resulting in ground spoiler extension.

The aircraft made a second touchdown at approximately 154 kts CAS, with a recorded pitch angle of 9.49° and a vertical acceleration of 2.56 g. The aircraft tail section struck the runway.

Upon landing, neither of the pilots perceived that a tailstrike might have occurred, and the flight crew continued taxiing to the apron.

During taxi o the apron, the commander briefed the passengers on the hard landing.

The SHK described the damage:

The aircraft tail section was substantially damaged.

The aircraft struck the runway at a recorded pitch angle of 9.49° and at a recorded G-load of 2.56 resulting in the following damages:

- The lower part of the aft fuselage skin was damaged between frame (FR) 60 and FR 69.
- The impact deformed several of the above-mentioned frames.
- Several floor beams cracked at their attachments to the frames.
- A vertical support strut attached to FR 65 sheared and indicated a significant vertical displacement.

The SHK analysed:

Based on the available weather information sources, the aircraft encountered moderate turbulence during the final approach, which in combination with strong crosswinds and dark night operations most likely affected the flight crew handling on short final and landing.

Minor and short-term speed variations outside the stabilized approach concept (target speed +10 kts) below 500 RH were recorded.

Despite strong crosswinds and gusts, Ground Speed Mini and autothrust in SPEED mode ensured an efficient thrust management and corrected minor short-term speed variations outside the stabilized approach parameters.

After disconnection of the autopilot, a headwind gust led the CAS to increase and thereby increasing lift.

Consequently, the rate of descent decreased. The pilot flying counteracted the decrease of the rate of descent with two consecutive nose-down control sidestick inputs leading the pitch angle to decrease.

Nose-down dynamics combined with a downdraft led the rate of descent to increase within approximately 4 seconds up to approximately 1100 ft/min reached at 35 ft RH. In general, on a constant minus 3° glide path and at a GS of approximately 146 kts, the rate of descent would be approximately 770 ft/min.

Though the approach was performed in rainy and dark night conditions, the flight crew, below decision height, kept visual contact with the runway and appropriate visual cues for the landing.

The destabilization before the flare did not trigger flight crew calls and/or a decision on going around.

In stabilized conditions, the flare height was about 30 ft, and the engine thrust levers should be retarded to flight idle between 30 ft RH and touchdown.

Though a synthetic voice call on RETARD sounded at 20 RH, the sequence of events apparently took the pilot flying by surprise.

Due to the unannounced and uncorrected rate of descent within approximately 4 seconds prior to the flare, the actual flare initiation did not take place in time to change the vertical aircraft flight path and trajectory sufficiently prior to the touchdown, which led to a hard landing (2.59 g).

A headwind drop and a downdraft just before the initial touchdown aggravated the situation.

Following revealed findings caused a high bounce:
- The energy of the hard landing partially restored by the main landing gear shock absorbers.
- A nose up sidestick order progressively released after touchdown.
- The thrust lever movement forward of the climb detent to maximum continuous thrust detent.
- The inhibition of ground spoiler partial extension.
- After touchdown and because of thrust lever movement forward of climb detent, the autothrust reverted from active to armed and consequently led the thrust to largely increase.

The AIB finds it probable that the pilot flying - in order to reduce the effect of a coming hard landing, though too late, - instinctively moved the thrust levers forward of the climb detent.

As pitch sidestick order by the pilot flying was released but still in nose-up position and associated with the pitch-up effect of the thrust increase, the pitch angle increased and activated the synthetic voice call PITCH for an excessive pitch angle.

The pilot flying applied a pitch-down order leading the pitch angle to decrease.

During the bounce, the thrust levers were retarded to the flight idle detent (within three seconds of the initial touchdown) leading to autothrust disconnection and consequently led the thrust to decrease.

The thrust lever movement to the flight idle detent enabled ground spoiler extension.

The ground spoiler extension and the nose-down order by the pilot flying led to a partial loss of lift with an increase of the rate of descent.

To counteract this partial loss of lift, the pilot flying applied a pitch-up order to an approximately full aft sidestick deflection, which together with a residual pitch-up effect of ground spoiler extension led to activation of the second synthetic voice call PITCH for an excessive pitch angle.

The approximately full aft sidestick deflection was too lately applied to have a significant effect on the vertical speed, which continued increasing up to a rate of decent of approximately 600 ft/min at the second touchdown.

At an excessive pitch angle, the aircraft was exposed to a second hard landing (2.56 g), and the aircraft tail section struck the runway.

Even though, the actual recorded pitch angle was less than the geometric limits published by the aircraft manufacturer, runway downslope and the flexibility effect experienced by the aircraft structure and the landing gear resulted in the tailstrike.

The range of aft sidestick inputs on a comparable and “standard” landing varied considerably compared with the control inputs in this tailstrike event. The QAR data of this event indicated a significant change of pitch angle, mainly consistent with the sidestick inputs applied during the dynamic phase of the flare and the touchdown.

According to standard operating procedures, a high bounce required a flight crew decision on going around.

However, in practice it is not necessarily easily perceivable to a flight crew, if an aircraft has made a high or a light bounce.

In this tailstrike event, landing performance was not a limiting factor (runway length was sufficient to stop the aircraft in time), which might have had a negative influence on the flight crew decision making.

Metars:
EKBI 082020Z AUTO 18019G34KT 150V220 6000 RA OVC009/// 07/06 Q0995=
EKBI 081950Z AUTO 19019G34KT 160V220 6000 RA OVC009/// 07/06 Q0995=
EKBI 081920Z AUTO 18018G34KT 9000 RA OVC009/// 07/06 Q0996=
EKBI 081850Z AUTO 18018G31KT 150V220 7000 -RA OVC008/// 07/06 Q0996=
EKBI 081820Z AUTO 19019G32KT 9999 -RA OVC008/// 07/06 Q0997=
EKBI 081750Z AUTO 19018G30KT 9999 -RA OVC007/// 07/06 Q0997=
EKBI 081720Z AUTO 19019G32KT 9000 -RA OVC008/// 07/06 Q0998=
EKBI 081650Z AUTO 19016G28KT 7000 -RA OVC007/// 06/06 Q0998=
EKBI 081620Z AUTO 19016G30KT 160V220 6000 -RA OVC007/// 06/06 Q0998=
EKBI 081550Z AUTO 19016G30KT 160V220 6000 -RA OVC007/// 06/06 Q0999=
EKBI 081520Z AUTO 19016KT 160V220 4500 -RA OVC006/// 06/05 Q0999=