Safety Investigation Report 2018:2 Analysis/2.1 Diversion from Filed Flight Plan
SAFETY INVESTIGATION REPORT MH370 (9M-MRO)
2.1 DIVERSION FROM FILED FLIGHT PLAN
2.1.1 Seven Simulator Sessions
To analyse further on how MH370 had diverted from the Filed Flight Plan (FPL) route, the Team conducted a total of seven flight simulator sessions to recreate the two turns of MH370, i.e. six sessions on the left turn past waypoint IGARI and one session on the right turn on reaching the south of Penang Island. Three of the seven sessions were conducted at high speed and the remainders at low speed. The turns were based on the recorded primary radar data that recorded a primary target conducting a left turn from where the SSR code ceased, shortly after the aircraft passed waypoint IGARI. The following data (Tables 2.1A to 2.1F [below] on data input for the first six sessions respectively) were introduced to simulate an actual environment:
- • Actual weight and meteorological condition prevailing at the time of the turn (extracted from the computerised flight plan); and
- • Different speed and rate of turns to determine scenarios closest were made available to the Team.
1) Recreating the Left turn past Waypoint IGARI – Session 1
Initial conditions
| Fuel | 41,200 kg |
| Gross weight | 215,410 kg |
| Height | 35,000 ft |
| Speed | IAS 271 (475 knots ground speed) |
| Simulator setup | Speed/Lateral Navigation/Vertical Navigation (SPD/LNAV/VNAV), autopilot engaged, autothrottle engaged |
| Entry waypoint | N07.05.7 E103.47.1 |
| Exit waypoint | N07.12.7 E103.38.7 |
Table 2.1A - Data Input for Session 1
To get the ‘aircraft’ to track correctly, a flight path from waypoint IGARI to waypoint BITOD was generated with the entry and exit waypoints entered.
The simulation commenced before IGARI and the ‘aircraft’ turned right on LNAV and tracked to the entry waypoint. Once over the waypoint, the flight management computer (FMC) was directed to fly “direct to” to the exit waypoint. The ‘aircraft’ entered a left turn, with a maximum bank-angle of 26° (maximum bank-angle in LNAV is 25°).
About half-way through the turn, it was obvious that the ‘aircraft’ was not going to make it through the exit waypoint as it was overshooting as there was no tracking information in the FMC. The simulator session was then terminated.
2) Session 2
Initial conditions (identical to Session 1)
| Fuel | 41,200 kg |
| Gross weight | 215,410 kg |
| Height | 35,000 ft |
| Speed | IAS 271 (475 knots ground speed) |
| Simulator setup | SPD/LNAV/VNAV, autopilot engaged, autothrottle engaged |
| Entry waypoint | N07.05.7 E103.47.1 |
| Exit waypoint | N07.12.7 E103.38.7 |
| Additional waypoint | N05.15.6 E100.27.5 |
Table 2.1B - Data Input for Session 2
To get the ‘aircraft’ to track correctly, a flight path from waypoint IGARI to waypoint BITOD was generated with the entry and exit waypoint entered. A further waypoint was entered along a track of 244° at the commencement of the right turn south of Penang.
The simulation commenced before IGARI and the ‘aircraft’ turned right on LNAV and tracked to the entry waypoint. Once over the waypoint, the FMC was directed to fly “direct to”, to the exit waypoint. The ‘aircraft’ entered a left turn, with a maximum bank- angle of 26° (maximum bank-angle in LNAV is 25°).
The ‘aircraft’ made the exit waypoint; however, it took 3 minutes and 45 seconds to achieve it (the recorded radar time was 2 minutes 10 seconds).
3) Session 3
Initial conditions
| Fuel | 41,200 kg |
| Gross weight | 215,410 kg |
| Height | 35,000 ft |
| Speed | IAS 250 (425 knots groundspeed) |
| Simulator setup | SPD/LNAV/VNAV, autopilot engaged, autothrottle engaged |
| Entry waypoint | N07.05.7 E103.47.1 |
| Exit waypoint | N07.12.7 E103.38.7 |
| Additional waypoint | N05.15.6 E100.27.5 |
Table 2.1C - Data Input for Session 3
Following discussions, it was decided to reduce the speed in the turn to see if the rate of turn would increase. In this session, the speed was reduced to 250 knots IAS (ground speed of 425 knots). Similar set up as Session 2.
The simulation commenced before IGARI and the ‘aircraft’ turned right on LNAV and tracked to the entry waypoint. Once over the waypoint, the FMC was directed to fly “direct to” to the exit waypoint. The ‘aircraft’ entered a left turn, with a maximum bank- angle of 28° (maximum bank-angle in LNAV is 25°). The ‘aircraft’ made the exit waypoint. However, it took 3 minutes and 3 seconds to achieve it.
4) Session 4
Following further discussions, it was decided to further reduce the speed in the turn.
The simulation commenced before IGARI and the ‘aircraft’ turned right on and tracked to the entry waypoint. Once over the waypoint, the FMC was directed to fly “direct to”, to the exit waypoint. The ‘aircraft’ entered a left turn, with a maximum bank- angle of 23° (maximum bank-angle in LNAV is 25°).
Initial conditions
| Fuel | 41,200 kg |
| Gross weight | 215,410 kg |
| Height | 35,000 ft |
| Speed | IAS 220 (400 knots groundspeed) |
| Simulator setup | SPD/LNAV/VNAV, autopilot engaged, autothrottle engaged |
| Entry waypoint | N07.05.7 E103.47.1 |
| Exit waypoint | N07.12.7 E103.38.7 |
| Additional waypoint | N05.15.6 E100.27.5 |
Table 2.1D - Data Input for Session 4
The ‘aircraft’ made the exit waypoint. However, it took 3 minutes and 30 seconds to achieve it.
5) Session 5 (Manual Flying)
Following discussions, it was agreed that the turn could be executed in LNAV, but not in 2 minutes. It was decided that the bank-angle needed to be increased to reduce the time and that could only be achieved with the autopilot disengaged and the ‘aircraft’ manually flown, with the auto-thrust managing the speed. Similar set-up as Session 2.
Initial conditions
| Fuel | 41,200 kg |
| Gross weight | 215,410 kg |
| Height | 35,000 ft |
| Speed | IAS 271 (475 knots ground speed) |
| Simulator setup | SPD/LNAV/VNAV, autopilot engaged, then autothrottle engaged |
| Entry waypoint | N07.05.7 E103.47.1 |
| Exit waypoint | N07.12.7 E103.38.7 |
| Additional waypoint | N05.15.6 E100.27.5 |
Table 2.1E - Data Input for Session 5
The simulation commenced before IGARI with autopilot and autothrottle engaged and the ‘aircraft’ turned right on LNAV and tracked to the entry waypoint. Once over the waypoint, the autopilot was disconnected and the ‘aircraft’ manually turned to the left. Bank-angles around 30°-32° were used. As the entry and exit waypoints were displaced slightly laterally (i.e. not exactly aligned 180° apart), the ‘aircraft’s wings were rolled level when aligned to intercept the exit waypoint. This was at 2 minutes and 10 seconds. The ‘aircraft’ then intercepted the exit waypoint at 2 minutes and 40 seconds.
6) Session 6 (Manual Flying)
Finally, it was agreed that the same turn should be executed manually but at a lower speed of 250 knots with the autopilot disengaged and the ‘aircraft’ manually flown, with the autothrottle managing the speed. Same set-up as Session 2.
Initial conditions
| Fuel | 41,200 kg |
| Gross weight | 215,410 kg |
| Height | 35,000 ft |
| Speed | IAS 250 (425 knots ground speed) |
| Simulator setup | SPD/LNAV/VNAV, then autopilot disengaged, autothrottle engaged |
| Entry waypoint | N07.05.7 E103.47.1 |
| Exit waypoint | N07.12.7 E103.38.7 |
| Additional waypoint | N05.15.6 E100.27.5 |
Table 2.1F - Data Input for Session 6
The simulation commenced before IGARI with autopilot and autothrottle engaged and the ‘aircraft’ turned right on LNAV and tracked to the entry waypoint. Once over the waypoint, the autopilot was disengaged and the ‘aircraft’ manually turned to the left. Bank-angle around 35° was used (bank-angle warnings sounded several times).
At about half way through the turn (1 minute mark), the stick- shaker activated. The ‘aircraft’ intercepted the exit waypoint at 2 minutes and 28 seconds.
7) Analysis on Re-enactment Sessions (Sessions 1 - 6)
- a) From the various re-enactment sessions tested, it is apparent that the ‘aircraft’ could make the turn in LNAV, but took a longer time due to bank-angle limitations (25°) and also required the need to reduce speed (Session 3 was the closest at 3 minutes and 3 seconds) in the turn.
- b) However, there were issues with the entry waypoint being off the direct track IGARI to BITOD (to the south-east) and this resulted in the ‘aircraft’ being in a slight right bank when overflying the waypoint and then starting the left turn. This would have increased the time to make the left turn as the ‘aircraft’ had to roll through level, before rolling west.
- c) The ‘aircraft’ could also make the turn and achieve a closer time to the recorded radar data with the autopilot disengaged and manually flown (Session 5 was closest with 2 minutes 10 seconds to wings-level and 2 minutes and 40 seconds to the exit waypoint).
- d) Again, there were issues with the positioning of the entry and exit waypoints as they were not aligned (i.e. not 180° apart) leaving a short straight segment before the ‘aircraft’ intersected the exit waypoint.
e) Summary of 6 Simulator Re-enactment Sessions and Common Factors
Based on the six simulator re-enactment sessions conducted as summarised in Table 2.1G (below) and on the common factors in Table 2.1H (below), the Team concluded the following:
- i) The turn would have been carried out with the autopilot disengaged, as it was not possible to achieve a turn time of 2 minutes and 10 seconds (as suggested by recorded data) using autopilot. The manoeuvre can be performed by a single pilot. The Team also noted that the aircraft’s flight path from after the turn was consistent with the navigation being set to LNAV and/or heading mode, following published and/or manual waypoints that are not normally used with normal route (published airways between Kota Bharu and Penang).
Re-enactment Session 1 2 3 4 5 6 Ground Speed (in knots) 475 475 425 400 475 425 Autopilot Engaged ✓ ✓ ✓ ✓ x X Additional Waypoint x x X N05.15.6 E100.27.5 ✓ ✓ Bank angle (in degrees) 26o 26o 28o 23o 30-32o 35o Exit Waypoint Time Over-Shooting 2 min 45 sec 3 min 3 sec 3 min 30 sec 2 min 40 sec 2 min 28 sec Table 2.1G - Re-enactment Sessions
Common Factors 1. Fuel 41,200 kg 2. Gross Weight 215,410 kg 3. Height 35,000 ft 4. Entry Point N07.05.7o E103.47.1o Exit Point N07.12.7o E103.38.7o 5. Autothrottle Engaged Table 2.1H - Common Factors
- ii) From the data it was determined that the ‘aircraft’ was on heading mode that varied from 239o to 255o as it flew to the south of Penang where it continued westerly to Waypoint MEKAR where it finally disappeared completely at 1822:12 UTC [0222:12 MYT], about 10 nautical miles north of MEKAR.
- iii) Based on the Team’s review of the Military recorded radar display and printout, the aircraft’s flight path could not be determined, and there is no evidence of rapid altitude and/or speed changes to indicate that MH370 was evading radar.
- iv) Without further evidence, the reason for the transponder information from the aircraft ceasing could not be determined;
- v) It is determined that only the transponder signal of MH370 ceased from the ATC Controller display whilst displays from other aircraft were still available; and
- vi) There is also no evidence to suggest that the aircraft was flown by anyone other than the designated MAS pilots. However, the Team does not exclude the possibility of intervention by a third party.
8) Session 7 – Recreating the Right Turn South of Penang Island
Initial conditions
| Fuel | 36,000 kg |
| Gross weight | 210,200 kg |
| Height | 35,000 ft |
| Speed | IAS294 (525 knots groundspeed) M0.86 Note: A tailwind of 30 knots was needed to achieve this |
| Simulator setup | SPD/LNAV/VNAV, autopilot engaged, autothrottle engaged. |
| Entry waypoint | N05.15.6 E100.27.5 |
| Exit waypoint | N05.12.0 E100.01.5 |
| Additional waypoint |
Table 2.1I - Data Input for Session 7
To get the ‘aircraft’ to track correctly (Table 2.1I [above]), both the entry and exit waypoints were entered, without a track between them in the FMC. The ‘aircraft’ was flown on heading mode to turn gently to intercept the exit waypoint.
The simulation commenced before the entry waypoint. Once crossing the waypoint, a heading change to the right was initiated to achieve a bank-angle of 5°. During the turn, the bank-angle was increased to a maximum of 10°. The exit waypoint was easily intercepted at 3 minutes and 5 seconds (the recorded radar time was 3 minutes). No further simulations were done on this turn.
2.1.2 Ho Chi Minh Air Traffic Services Operations
Based on the on-site interviews and briefing from the Team’s visit to the Office of the Vietnamese Civil Aviation Authority in Ho Chi Minh City, it was noted that the radar position symbol for MH370 dropped from the radar display at 1720:59 UTC (0120:59 MYT). MH370 had not reached waypoint BITOD which is 37 nm from waypoint IGARI and based on the aircraft speed of 480 kt, it would take approximately five minutes for MH370 to travel from IGARI to BITOD.
The Direct Line Coordination Communication transcripts between KL ACC and Ho Chi Minh ACC suggested that there were uncertainties on the position of the aircraft. This could come about from the level of understanding of the English language. The HCM Duty Controller also could not communicate effectively during the interviews and an interpreter was there to assist him.
Reference:
Ho Chi Minh radar data recording, page 33 to 41 and page 51 to 61 of the Direct Line Coordination Communication KL ACC Sector 3+5 Planner (Appendix 1.18G) transcripts between Kuala Lumpur ACC and Ho Chi Minh ACC).