IndiGo flight followed a similar flight pattern and landed safely before the ill-fated Air India Express crashed

Image
An IndiGo AT-72 turboprop landed at Kozhikode airport 01hr45min prior to the ill-fated Air India Express B-737.  The IndiGo ATR followed a similar flight profile befoe landing on the easterly direction runway 10 at the airport. The IndiGo ATR too attempted to land on the westerly runway 28 but discontinued the approach. The reason could be due to poor visibility in rain and/or low clouds obscuring the pilots vision in an attempt to acquire visual references of the landing runway.The second attempt was carried out on the easterly runway 10, same as the Air India Express. At the time of landing, the prevailing visibility in rain was between 1500m-2000m. The instrument landing system which guides the aircraft through electronic ground based transmitter signals to 250 feet above ground requires the visibility of atleast 1300m or more.The IndiGo ATR may have faced less severe weather and the slower turboprop may have been more forgiving. Nevertheless, this pattern can provide some interesti…

Atlas Air accident could have been prevented but NTSB investigation soft peddles some aspects


Amazon Prime Air 767-300ER Crash near Houston ++ FSX - YouTube

The activation of Ground Proximity Warning System(GPWS) could have prevented the Atlas Air Boeing 767 accident. The GPWS of the Atlas Air Boeing 767,which is a system designed to alert pilots if their aircraft is in immediate danger of flying into the ground failed to activate as the aircraft dived into the swamps near Huston Airport, Texas.The National Transport Safety Board (NTSB), USA which is investigating the accident which occurred on 23rd Feb 2019, is nearing contemplation of the final report and would be releasing it in the next few weeks.

The activation of a GPWS warning could have pulled the co-pilot out of the sensory illusion in time for recovering the flight path. The crew could have attempted to recover the aircraft downward trajectory just in time to prevent the accident. The NTSB investigation has not touched upon these critical aspects.

The pilot flying the B-767 probably inadvertently activated the Go-Around mode which caused the engines to increase thrust to maximum power. The acceleration thus caused could have given the pilot a sensation of a steep climb. In order to prevent a presumably imminent stall due to the apparent climb, the pilot flying pushed the aircraft nose down into a deep dive. Read more on the somatogravic illusion.

The first implementation of Terrain Avoidance Warning System (TAWS) was (GPWS) and was introduced in the 1970s as a means to combat the high incidence of controlled flight into terrain (CFIT) accidents.GPWS was mandated in many countries and was responsible for a significant reduction in the number of CFIT accidents. CFIT accident occurs when an airworthy aircraft under the control of the flight crew is flown unintentionally into terrain, obstacles or water, usually with no awareness of the impending collision on the part of the crew.


TAWS
GPWS Mode 1 Warning

IATA & Honeywell guidance


The manufacturer of the GPWS, Honeywell along with the International Air Transport Association (IATA) issued a guidance material in 2019 on performance assessment of pilot response to EGPWS.

Following an EGPWS alert, flight crew should control the aircraft flight path with immediate maximum Required Obstacle Clearance (ROC) and maximum thrust to clear the obstacles threatening the flight. 

Warning level alert
  • Aggressively position throttles for maximum rated thrust.
  • Apply maximum available power as determined by emergency need. The pilot not flying (PNF) should set power and ensure that takeoff / go-around (TO/GA) power and modes are set.
  • If engaged, disengage the autopilot and smoothly but aggressively increase pitch toward “stick shaker” or Pitch Limit Indicators (PLI) to obtain maximum climb performance.
  • Continue climbing until the warning is eliminated and safe flight is assured.
Source: Performance assessment of pilot response to Enhanced Ground Proximity Warning System (EGPWS)

Pilot’s response time

During EGPWS alert events that pilots responded to, their response time is shown in figures below. For this analysis, a pilot’s response time was the time it took for the aircraft pitch to increase more than 1.4° after an alert due to limited availability of parameters that can be used to measure pilot’s response in the EGPWS flight history database.Data collected suggests that pilot’s response time is very similar regardless of EGPWS alert types or alert level (caution vs. warning). For that reason, statistics have been aggregated in the diagrams below.

Mode 1

Pilot response time to activation of Mode 1, Excessive descent rate

Mode 2
Pilot response time to activation of Mode 2, Excessive Terrain Closure

Source: Performance assessment of pilot response to Enhanced Ground Proximity Warning System (EGPWS)


Why did the GPWS fail to activate?


The NTSB,Systems Group Chairman’s Factual Report dated December 16, 2019 explains in brief the status of the GPWS as provided by Honeywell. The GPWS computer could not be recovered post the accident but the data could be verified from the flight data recorder(FDR) parameters. The FDR did not record the evidence of a warning and the discrete parameters which are used to warn of unusual situation were in "OFF" state for the entire accident sequence.


In a letter to the NTSB received on September 9, 2019, Honeywell indicated that the change of radio altitude values near the end of the FDR recording was considered excessive by the EGPWS and flagged for internal reasonableness. The flag caused the EGPWS simulation to disregard the radio altitude data for three seconds. This logic flag, when combined with the EGPWS time guard prior to the first issuance of a Mode 1 (sink rate/pull up) alert envelope penetration, would have delayed the issuance of a Mode 1 alert until after the end of the FDR recording.

The NTSB has not queried Honeywell in depth on the reasons for the failure of the activation of GPWS warnings or proposed system enhancement post accident, as they have done for other aspects of the investigation.

The contradiction

The B-767 is fitted with a Honeywell Mark V GPWS which has a Geometric Altitude function, based on GPS altitude. Geometric Altitude is a computed pseudo-barometric altitude designed to reduce or eliminate altitude errors resulting from temperature extremes, nonstandard pressure altitude conditions, and altimeter miss-sets. This ensures an optimal EGPWS alerting and display capability. Therefore contrary to what has been stated by Honeywell, the GPWS does not solely depen upon the radio height.

1.The Fly Dubai Boeing 737 accident on 19 Mar 2016 in Russia was probably caused due to disorientation of the Captain leading to a nose dive to the ground on final approach to land. In the final stages of the flight, the aircraft was descending at 18,000 feet per minute, a nose down pitch of 50° and the GPWS "Pull Up" warning was activated.


Honeywell, being the EGPWS manufacturer, upon the investigation team request commented that the accident flight actual parameters (rate of descent 18000 ft/min (91 m/s)) had been far beyond than the maximum values (7000 ft/min (36 m/s)), determined by the TSO C151b and DO-161A documents, in accordance to which the system had been designed.

2. The Lion Air B-737 Max accident on 29th Oct 2018 in Indonesia was probably caused due to activation of the MCAS. The maximum recorded values were, Pitch values greater than 40° nose down & Vertical speed values were greater than 33,000 ft/min.

The GPWS modes activated the warnings at slightly above 5000feet radio height.

Lion Air

GPWS activation height
GPWS activation height

The Atlas Air B-767 reached a max. rate of descent of 20,000 feet per minute at 2500' reducing to 13,000' prior to impact. If the GPWS warning activated in the Fly Dubai & Lion Air accidents, then why did the GPWS warning fail to activate in the Atlas Air accident?

Attempted recovery by the Captain


NTSB changed wording in Investigative Update on crash of Atlas Air ...
B-767 Elevator, Left & Right sides

The Atlas Air captain attempted to recover from the nose down attitude by pulling the control column back in order to pull the aircraft nose up. The recovery was hampered by counter forces from the first officer side, who was still probably under the impression that the aircraft was stalling. This caused a split in the elevators. The elevators are designed such that if both pilots move their control columns in the opposite direction, the left part of the elevator moves in the direction of the inputs given by the left control column and the right part of the elevator similarly moves with the right control column inputs. This creates a split and thenet  effectiveness of the elevators is reduced.
Elevator Split
Elevator split due opposite movements of control columns

Unanswered questions

  1. Why did the NTSB investigation not pursue the failure of the GPWS to generate warnings in greater detail? A single page write up on the GPWS is grossly insufficient simply because the GPWS is a safety system built primarily to prevent controlled flight into terrain (CFIT). CFIT is globally the second biggest threat after loss of control in air. NTSB has been pursuing the case for installation of Airborne Image Recorders but there is no mention of the need to improve the GPWS system.
  2. Whereas the GPWS has activated in other accidents which display similar aircraft trajectory, why did the GPWS not activate in the accident of Atlas Air?
  3. Honeywell in their reply have stated that the change in radio altiude values were considered excessive. The Honeywell GPWS V operator manual indicated that a geometric altitude is used and is far more accurate. Why did honeywell attempt to mislead by providing inaccurate information?










Comments

  1. The failure of the GPWS to trigger alert needs thorough investigation for any improvement that is possible. However with such an incompetent crew handling the aircaft an accident was always waiting to happen. SI is not a new discovery and yet these pilots continued to manoeuvre the aircraft in instrument conditions with extreme control inputs without any reference to instruments whatsoever. In Tatarstan GA case pilot didn't know AP had tripped, in FlyDubai case GPWS did trigger, in Atlas it didn't, but the result was same. As in AF447 stall warning issue these pilots are outmanoeuvring the safety system itself. All these mishaps occurred without any equipment failures. No system can be made idiot proof.

    ReplyDelete
  2. I am assuming that the final report will talk about the crews experience.
    And what the first Officer's experience, and especially his training history were.....

    ReplyDelete

Post a comment

Popular posts from this blog

IndiGo flight followed a similar flight pattern and landed safely before the ill-fated Air India Express crashed

Mangalore Accident recommendation for table top runways ignored. 10 years later another accident. Where does the buck stop?

Ethiopian Airlines cargo plane catches fire at Shanghai airport, no casualties