Avior B732 at Manaus on Nov 28th 2018, overran runway on landing

Last Update: July 25, 2021 / 10:45:10 GMT/Zulu time

Bookmark this article

Incident Facts

Date of incident
Nov 28, 2018

Classification
Incident

Cause
Overran runway on landing

Airline
Avior Airlines

Flight number
9V-1272

Departure
Caracas, Venezuela

Destination
Manaus, Brazil

Aircraft Registration
YV-2937

Aircraft Type
Boeing 737-200

ICAO Type Designator
B732

Airport
Eduardo Gomes International Airport, Manaus

Airport ICAO Code
SBEG

An Avior Airlines Boeing 737-200, registration YV-2937 performing flight 9V-1272 from Caracas (Venezuela) to Manaus,AM (Brazil) with 39 passengers and 5 crew, landed on Manaus' runway 29 at about 15:43L (19:43Z) but overran the end of the runway and came to a stop 100 meters past the end of the runway end safety area with all gear on soft ground. There were no injuries.

In July 2021 Brasil's CENIPA released their final report concluding the probable cause of the serious incident was:

Contributing factors.

- Control skills – undetermined.

The inadequate application of flight controls may have contributed to the excess speed at which the aircraft approached for landing on SBEG, which required greater distance for the aircraft to stop.

- Communication – a contributor.

There were flaws in the communication process established between the
crewmembers of the YV2937 and the TWR-EG, characterized by the lack of accurate information on the intensity of the existing rain, as well as on the change of runway threshold for landing operations.

These failures contributed to the landing on runway 29, with tailwind, favoring the runway excursion.

Thus, the control agencies were also unaware of the aircraft's failure condition, since there was no declaration of urgency or emergency, contributing to their failure to activate the ground support means as required by the Aerodrome Emergency Plan.

- Adverse meteorological conditions – a contributor.

The distance required to stop the aircraft was increased due to the weather conditions present, as well as making it difficult to visualize the runway exit, by the TWR-EG, which culminated in the delay of the ground support means.

- Crew Resource Management – undetermined.

It is possible that an inadequate cabin coordination has caused the crewmembers to not realize that they were approaching the runway with tailwind, as well as that they did not observe, as recommended in the QRH, that the aircraft could not be taxied after landing, which could be corroborated by the fact that the crew did not request any ground support from the control agencies.

- Use of phraseology by the crew – undetermined.

The fact that the crew did not expressly declare that the landing would be carried out in a contingency situation may have made it possible that the other agencies involved in the operation could not provide the necessary support for the landing of the aircraft.

- Use of phraseology by ATS – a contributor.

There was a failure in the transmission of meteorological information by the ATCO to the YV2937, not being informed that the threshold in use had already been changed to 11, which allowed the landing to be carried out on runway 29, with tailwind.

- Piloting judgment – a contributor.

There was an inappropriate judgment of the consequences inherent to the Loss of System A failure, mainly considering that the approach took place in adverse weather conditions and with tailwind, and without requiring any type of ground support.

- Aircraft maintenance – undetermined.

The sequence of failures observed at the time of the loss of the Hydraulic System A, followed by secondary failures, such as the incomplete lowering of the flaps by the emergency system and the failure to activate the #1 reverser, may have been caused by poor maintenance of these systems, however such a hypothesis could not be confirmed.

- Perception – undetermined.

A possible decrease in the level of situational awareness on the part of the crew may have created difficulties in adequately perceiving the adverse effects generated by the combination of the failures in the aircraft systems and the existing weather conditions.

Possibly, there was a misinterpretation about the severity of rain present at the Aerodrome, by the TWR-EG, which led to the transmission of inaccurate information to the crewmembers of the YV2937.

- Decision-making process – a contributor.

The inaccurate assessment of meteorological information and the impact of the hydraulic system failures culminated in the decision to proceed to landing at threshold 29, despite adverse circumstances, which favored the runway excursion.

CENIPA summarized the sequence of events:

After landing on SBEG runway 29, the aircraft passed the Landing Distance Available (LDA), leaving the runway from its opposite end (overrun), coming to a stop on unpaved land.

The aircraft had minor damage.

All occupants left unharmed.

CENIPA summarized the aircraft damage:

The only damage observed was the point where the leakage of the hydraulic fluid occurred in the body of the System A pressure transmitter, which occurred in flight.

No tire damage was observed that could characterize aquaplaning.

CENIPA reported the captain (ATPL) had accumulated 4,850 hours total, 2,000 hours thereof on type, the first officer (ATPL) had accumulated 10,100 hours total, thereof 4,000 hours on type.

CENIPA reported after occurrence maintenance: "After the aircraft was removed to the apron, a maintenance work was carried out on System A.

The main hydraulic reservoir was refilled and pressurized, when it was observed that the hydraulic leak occurred through the body of the System A Hydraulic Pressure Transmitter, Part Number (PN) ST-107R and Serial Number (SN) 7929A."

CENIPA computed that with the 170 knots over ground observed during the final approach, the observed flaps assymetry and flaps being below 15 degrees, landing weight at 90,000lbs, wet runway with good braking action and maximum manual braking the required landing distance was 1463 meters, which was available.

However, CENIPA added, due to the loss of hydraulic system A the inboard main wheel brakes would not be operating, the reversers would move slower and commented: "The factors mentioned were already foreseen by the manufacturer in its operating manual, however, the failures, as they occurred at the time of landing, that is, concomitant with “Loss of System A” and “Trailing Edge Flaps Disagree”, were not approached in a combined manner, which made it impossible for a precise calculation of the landing distance, under these conditions, to be performed."

The aircraft came to a stop 100 meters past the end of the runway.

The CENIPA analysed:

Until the approach for landing, the flight went on without any abnormality. When starting the aircraft configuration, immediately after placing the flap control lever in position 15, there was a drop in the hydraulic pressure of System A, which was responsible for providing hydraulic energy for the movement of the flaps, as well as for other aircraft systems.

After the loss of pressure, the crewmembers realized that the amount of hydraulic fluid in the reservoir of that system had dropped to “zero”, that is, there was no more hydraulic fluid in System A.

In view of this, the APP-MN was asked to wait in the position where they were to solve the failure, which was authorized by the control. The crew then proceeded to read the QRH - Loss of System A.

During the execution of the Checklist “Loss of System A”, the flaps were commanded to position 15, through the alternate system, however, the flaps lowered only to position 10.

After that, the landing gear was lowered by the emergency system, successfully confirmed by the three green lights on, indicating that the three landing gears were lowered and locked.

For the calculation of the reference speed (VRef), the crewmembers reported having used the QRH, obtaining approximately the speed of 144kt, however, the procedure for Loss of System A indicated that a speed of VRef 40 + 30 should be used, which, under those conditions, would be 150kt.

In the radar visualization provided by the APP-MN, it was observed that the final approach was flown with a speed on the ground of 170kt, about 20kt above the speed stipulated by the QRH. It is possible that part of that speed was caused by the wind component, which, at that time, was approximately 5kt tail.

The available weather information indicated that there was moderate rain on the runway. However, this condition was not passed on to the crewmembers, since the TWREG was limited to communicating that there was “light rain through the tower”.

During the investigation, it was found that this information was inaccurate, since, corroborated by the CCTV images of the airport administrator and the measurement performed by the EMS-EG rain gauge, the amount of rain that fell at that moment was not light, according to the ICA 105-15.

This fact showed that, possibly, the TWR-EG's perception of the weather conditions was inaccurate, which led to a misinterpretation regarding the intensity of the rain at that time.

In addition, the TWR-EG already had the information that the prevailing wind was favorable for landing on runway 11 and not on runway 29, but it did not inform the crew of the possibility of tailwind, serving only to transmit the reading of the present wind.

Thus, the runway change process failed, since the YV2937 was never informed that the threshold in use had become the 11. Both the TWR-EG and the APP-MN stopped transmitting this information to the YV2937, which could be crucial for the safe landing of the aircraft, given the direction and intensity of the wind (180°/16kt) informed by the TWREG in the landing authorization.

Possibly, due to the high workload to which they were subjected at that time, the pilots of the YV2937 did not pay attention to the fact that the wind informed by the TWR-EG, during the approach, indicated that the landing should be carried out on runway 11.

It is noteworthy that the meteorological conditions, combined with the hydraulic problem presented by the aircraft at that time, indicated that the landing might not occur satisfactorily, since most of the systems necessary for the aircraft's deceleration were inoperative.

Thus, it is possible that there was a decrease in the crew's situational awareness level, to the point of not being aware of the risks arising from the association between adverse weather conditions and the aircraft's inoperative systems, which, by themselves, already required additional care.

It is noted that the crew, before landing, did not declare an emergency or request any type of ground support, which reinforces the understanding that the pilots had not understood the gravity of the situation, since, according to the note “items inoperative” of the Loss of System A of the QRH, the aircraft could not be taxied by its own means.

As a result of the difficulties to understand the dynamism of that operation, the landing was made at a speed above the one predicted by the QRH, which resulted in an increase in the distance required to stop the aircraft.

In addition, only one reverser worked, which, added to the tailwind component, caused the aircraft to cover the entire length of the runway and cross the limit of the threshold 11, entering the grassy area and damaging the end of the runway markings.

The fact that the crew did not mention to the controllers the real situation to which the aircraft was subjected may have contributed to them not realizing that the flight needed additional care, including ground support.

The TWR-EG was unable to visualize the YV2937 aircraft passing the end of the runway, due to the position of the tower in relation to threshold 11, combined with the heavy rain present at the time of landing.

This delay in identifying this type of occurrence can be crucial in situations where the response time is essential for the support to be provided. The pilot himself, when asked if he had already cleared the runway, was the one who informed that he had exceeded his limit and needed support.

The MCA 100-16, in item 2.20.2, NOTE, established that the frequently used Q code groups, which were already considered a common part of the aeronautical terminology, could be used as a more adequate alternative instead of long and complex phrases.

In aviation, terminologies with the code “Q” are commonly used, in reference to the adjustment of the altimeter in relation to the ground, the average sea level and the local pressure, in addition to the codes QDM (Magnetic Heading) and QDR (Magnetic Marking) in the aerial radio navigation. These codes are used in radios with air traffic control agencies as an unambiguous abbreviation, where safety and efficiency are of vital importance.

It was found that there was no forecast for the use of the “QAP” code (waiting) by the tower, since the term was not an alternative for a long sentence, and the use of this code did not bring any operational gain in communications made with the Control Tower.

Still, in relation to the operation of Eduardo Gomes Airport, an attempt was made to identify how a Special Inspection of Water Level Measurement was carried out, which, according to the PLEM, should be performed whenever it rained more than 25mm / h.

There was also a failure of this procedure, reinforced by the fact that the Infraero did not have a rain gauge, therefore, it did not have access to the measurement of the rain that fell on the Aerodrome, having to consult the information available with the EMS-EG.

However, there was no provision in the MOPS that such contacts would be made with that service.

According to a survey carried out with the EMS-EG, it was found that there was no record of special inspections performed by the Aerodrome operator on the day of the occurrence, despite the fact that during the months of October to December 2018, these inspections should have been carried out.

Regarding the execution of maintenance services on the aircraft, the sequence of nonconformities observed at the time of the failure of the Hydraulic System A, followed by secondary failures, such as the incomplete lowering of the flaps by the emergency system and failure to activate the #1 reverser, may have been caused by poor maintenance of these systems. However, this hypothesis could not be confirmed.
Metars:
SBEG 282100Z 19002KT 9999 VCSH SCT018 FEW020TCU SCT080 24/23 Q1 008 RETS=
SBEG 282035Z 15005KT 9999 -RA SCT018 FEW020TCU SCT080 25/23 Q10 08=
SBEG 282010Z 12010KT 9999 -TSRA SCT012 SCT018 FEW020CB 25/23 Q1 008=
COR SBEG 282000Z 10014KT 1000 +TSRA BKN008 FEW020CB SCT025 25/2 3 Q1008=
SBEG 281920Z 23003KT 9999 TS VCSH SCT012 SCT018 FEW023CB 30/24 Q1008=
COR SBEG 281900Z 29003KT 9999 VCSH SCT012 SCT020 FEW025TCU 31/2 3 Q1008=
SBEG 281800Z 23004KT 9999 SCT020 FEW025TCU 30/24 Q1009=
SBEG 281700Z 27005KT 9999 SCT020 FEW025TCU 31/23 Q1009=
SBEG 281600Z 19005KT 9999 SCT020 29/23 Q1011=
SBEG 281500Z 23003KT 9999 BKN020 29/23 Q1011=