Travel Service B738 at Lajes on Feb 22nd 2014, hard landing

Last Update: October 6, 2015 / 15:17:23 GMT/Zulu time

Bookmark this article

Photo of Travel Service OK-TVT, Boeing 737-800 — Travel Service OK-TVT, Boeing 737-800 (Photo credit: BriYYZ / Flickr / License: CC by-sa)

Incident Facts

Date of incident
Feb 22, 2014

Classification
Accident

Cause
Hard landing

Airline
Travel Service

Flight number
QS-4130

Departure
Prague, Czech Republic

Destination
Lajes Terceira Island, Portugal

Aircraft Registration
OK-TVT

Aircraft Type
Boeing 737-800

ICAO Type Designator
B738

A Travel Service Boeing 737-800, registration OK-TVT performing flight QS-4130 from Prague (Czech Republic) to Lajes Terceira Island (Portugal) with 164 passengers and 6 crew, landed on Lajes Airport's runway 15 for a planned refueling stop in strong gusting crosswinds at 06:43L (07:43Z), touched down hard and rolled out without further incident.

A post flight inspection revealed the fuselage received creases on the right hand side near the nose gear strut, the aircraft was unable to continue the flight to Montego Bay (Jamaica), the ultimate destination of the flight.

A replacement Boeing 737-800 registration OK-TVL was dispatched to Lajes, continued the flight and reached Jamaica with a delay of about 8 hours.

On Mar 6th 2014 Portugal's GPIAA released their preliminary report reporting, that the captain was pilot flying during the approach and landing at Lajes. The captain had performed "a great number of landings" into Lajes previously. The captain provided testimony that the approach to runway 15 was normal until about 5nm before touchdown despite wind speeds of up to 50 knots (at approach altitude) and severe turbulence throughout the entire approach. Below 1000 feet AGL the aircraft encountered effects consistent with windshear with "huge variations" of intensity. The captain opted to remain above glideslope in order to have an additional safety margin in case of gusts, the approach was flown with autothrust systems engaged. On short final the aircraft developed a high rate of descent and loss of airspeed which was counteracted by the captain with increased thrust and increased pitch, the aircraft nonetheless touched down hard on its main gear almost instantly followed by the nose gear and bounced off. The crew did not initiate a go around, the aircraft touched down a second time at +3.52G.

Post occurrence investigation at first revealed creases and deformations of the fuselage skin above the nose wheel gear bay on both sides. A closer examination subsequently identified folded braces inside the nose wheel bay with bent struts, cracks in fuselage frames "that had suffered structural damage from the vertical loads applied in the contact with the runway". The occurrence was rated an accident, the investigation is continuing.

On Oct 6th 2015 the GPIAA released their final report, combining both English and Portugese in one document, concluding the probable causes of the accident were:

Procedural

The flight crew did not comply with aircraft manufacturer procedures and company SOPs, which required a “go around” manoeuvre;

Actuation

The accident was due to an excessive control column forward input, causing aircraft negative pitch attitude which led to a very high impact loading on the nose undercarriage, leading to the severe damage of braces in the interior structure of the nose wheel bay, struts and fuselage frames. This followed a chain of events, which contributed to the accident including the wind gusts, turbulence, decision to fly above ILS G/P, the use of the A/T without A/P and the decision to land from an unstabilized final approach.

The following were considered as contributing factors:

- The aircraft approach was conducted under turbulent conditions;
- The PF established an approach profile of one dot of scale above the nominal ILS G/S;
- The PF did not disconnect the A/T after disengagement of the A/P;
- Deviation from aircraft “stabilized approach” profile;
- The PM did not provide the required call outs for the “stabilized approach” deviations.

In their analysis the GPIAA complained that there was no cockpit voice recorder recording available of the last 120 minutes of the accident flight, the recording had been overwritten by the repositioning of the aircraft from one stand to another in Lajes, disabling the investigation to look into possible cockpit resource management issues, that however surfaced through interviews with the crew. From the interviews it appeared that no calls to query the pilot actions were made, the captain (35, ATPL, 6,819 hours total, 5,548 hours on type) was pilot flying, the first officer (32, CPL, 1,724 hours total, 1,442 hours on type) was pilot monitoring. The crew, having three options to divert, go around due to unstable final approach and attempt another approach or continue landing, chose to continue landing.

The GPIAA reiterated that there are stringent requirements to electrically isolate cockpit voice and flight data recorders following an accident.

The GPIAA analysed that the captain disengaged the autopilot while descending through 1500 feet AGL but left autothrust engaged despite aircraft manufacturer and standard procedures' recommendation to also disconnect autothrust when the autopilot is being disengaged. The GPIAA annotated that the captain justified the decision to leave autothrust engaged with low speed protection due the turbulences prevailing. In addition, the captain decided to keep one dot above glide to have an additional safety margin in the severe turbulence encountered.

The GPIAA wrote: "When descending through “RA” 475 ft (07:41:09 UTC), the aircraft had a rate of descent of -1136 ft/min, thus abandoning the “stabilized approach” envelope, for about two seconds. At this stage would be advisable the flight crew to comply with the “go around” manoeuvre, as recommended by aircraft manufacturer and company SOPs. ... From 07:41:22 UTC (“RA” 168 ft) till 07:41:24 UTC (“RA 116 ft) the ILS G/S deviation augmented to 1.20 dot of scale DWN and further to 1.25 dot of scale DWN at 07:41:25 UTC (“RA” 93 ft) and 07:41:26 UTC (“RA” 71 ft) with a rate of descent attaining -1,040 ft/min. Due to G/S deviations more than one dot of scale, the aircraft abandoned again the “stabilized approach” envelope and the flight crew, for the second time, did not execute a “go-around”, as required by aircraft manufacturer and company SOPs. ... Between “RA” 169 ft (07:41:22 UTC) and “RA” 52 ft (07:41:27 UTC) the engine thrust has been reduced to “approach idle” with CAS within limits of “command speed”. This was a result of flying above the ILS G/S, and when the PF induced manually a -0.5ᵒ pitch down attitude the A/T maintained the engine thrust reduced to control CAS. In consequence, the rate of descent attained -1,184 ft/min (07:41:28 UTC), and once again the PF did not execute the “go around” manoeuvre. ... At this moment, missing three seconds for the aircraft to land the PF applied engine thrust and pulled the flight control column to increase the pitch-up attitude. Due to significant deviation from aircraft manufacturer recommended profile at threshold and flare heights, engine spool and late actuation of flight controls, the aircraft inertia didn’t allow in time, the avoidance of RWY impact with 3.52 G vertical acceleration. At the same time, when de-crabbing the aircraft with a right crosswind component, the landing occurred with a slight left bank instead of zero bank or right bank, as appropriate for the wind conditions. The manoeuvre was executed with insufficient input of aileron and rudder controls for the correct alignment of aircraft longitudinal axis with RWY centreline, which could have caused an extra aerodynamic drag. After the impact, the aircraft registered a bounce of 15 ft with 3.9ᵒ pitch-up attitude and the PF pushed forward the control column resulting in second impact of 2.75 G vertical acceleration with -1.9ᵒ pitch down attitude over the nose wheel causing a substantial damage in that area of the aircraft fuselage. The second bounce had a smaller amplitude resulting in aircraft attitude closer to manufacturer’s recommended touchdown profile and which was controlled by the PF."

Metars:
LPLA 221000Z 21030G42KT 9999 FEW009 SCT015 FEW018TCU 16/14 Q1010
LPLA 220930Z 21026G42KT 9999 FEW009 SCT015 16/13 Q1010 RESHRA
LPLA 220909Z 21029G40KT 3500 1500W SHRA SCT004 BKN011 16/14 Q1011
LPLA 220900Z 20026G36KT 6000 2000NW RA FEW008 BKN011 16/14 Q1010
LPLA 220800Z 20018G28KT 9999 VCSH FEW008 BKN011 15/14 Q1011
LPLA 220700Z 19022G32KT 160V230 9999 FEW008 BKN012 14/13 Q1013
LPLA 220600Z 18015G30KT 140V230 8000 -RA BKN014 BKN032 14/12 Q1015
LPLA 220500Z 19016G28KT 8000 -RA FEW015 BKN032 14/12 Q1016
LPLA 220400Z 19016G29KT 9999 FEW015 SCT035 14/11 Q1018