AA2018-5

AIRCRAFT ACCIDENT
INVESTIGATION REPORT

KOREAN AIR LINES CO., LTD.

HL7534

July 26, 2018

 The objective of the investigation conducted by the Japan Transport Safety Board in accordance
with the Act for Establishment of the Japan Transport Safety Board and with Annex 13 to the
Convention on International Civil Aviation is to determine the causes of an accident and damage
incidental to such an accident, thereby preventing future accidents and reducing damage. It is not the
purpose of the investigation to apportion blame or liability.

Kazuhiro Nakahashi
Chairman
Japan Transport Safety Board

Note:
This report is a translation of the Japanese original investigation report. The text in Japanese shall
prevail in the interpretation of the report.
 AIRCRAFT ACCIDENT
INVESTIGATION REPORT

ENGINE FIRE DURINING TAKE-OFF ROLL

KOREAN AIR LINES CO., LTD.
BOEING 777-300, HL7534

ON RUNWAY 34R
AT TOKYO INTERNATIONAL AIRPORT
AT AROUND 12:38 JST, MAY 27, 2016

July 6, 2018
Adapted by the Japan Transport Safety Board
Chairman Kazuhiro Nakahashi
Member Toru Miyashita
Member Toshiyuki Ishikawa
Member Yuichi Marui
Member Keiji Tanaka
Member Miwa Nakanishi
 SYNOPSIS

Summary of the Accident

On Friday, May 27, 2016, a Boeing 777-300, registered HL7534, operated by
Korean Airlines Co.,Ltd, as the scheduled Flight 2708 of the company, flight crew had a
rejected takeoff on runway 34R at the Tokyo International Airport during a takeoff roll
to Gimpo International Airport, because there was a warning to indicate a fire from the
No.1 (left-side) engine activated at around 12:38, the flight crew stopped the aircraft on
the runway, and conducted an emergency evacuation. There were 319 people in total on
board, consisting of the PIC, sixteen other crew members, and 302 passengers. Among
them, 40 passengers were slightly injured.

Probable Causes
It is highly probable that the causes of this accident were the fracture of the high
pressure turbine (HPT) disk of the No.1 (left-side) engine during the takeoff ground roll
of the HL7534, the penetration of the fragment through the engine case and the
occurrence of subsequent fires.
Regarding the cause for the 1st stage HPT disk to be fractured, it is probable that a
step was machined exceeding the allowable limit when machining U-shaped groove on
the aft side of the 1st stage HPT disk to manufacture the engine and from this step the
low-cycle fatigue cracks were initiated and propagated during running of engine.
Regarding why the step could not be found, it is somewhat likely that defects failed
to be detected at the time of the inspection by the manufacturer during the production
process. And as for the cracks that were not found, it is somewhat likely that those cracks
failed to be detected at non-destructive inspection on the disk by the Korean Airlines Co.,
Ltd , at the time of maintenance of the engine in use.
Regarding the fire breakout from the No.1 engine, it is probable that due to the
impact forces generated by the release of the fragment from the ruptured rim part of the
1st stage HPT disk through the engine case and the engine rundown loads generated
when the engine stopped suddenly, the cracks were developed in the outer case of the
Fuel Oil Heat Exchanger and the fuel and engine oil leaking through these cracks
contacted the hot area of engine case of the No.1 engine to be ignited.

1
Abbreviations used in this report are as follows:

AC： Advisory Circular

AD： Airworthiness Directive
APU: Auxiliary Power Unit
ARAIB： Aviation and Railway Accident Investigation Board（Korea）
ATC： Air Traffic Control
ATSM： Automatic Society for Testing and Materials
CA: Cabin Attendant
CAPT： Captain
CAT： Category
CCM： Cabin Crew Manual
CMM： Coordinate Measuring Machine
COM： Cabin Operations Manual
CSN： Cycle Since New
CVR： Cockpit Voice Recorder
ENG： Engine
EVAC： Evacuation
EFB： Electronic Flight Bag
EGT: Exhaust Gas Temperature
EICAS： Engine Indicating and Crew Alerting System
FAA： Federal Aviation Administration
FCOM： Flight Crew Operations Manual
FCTM： Flight Crew Training Manual
FDR： Flight Data Recorder
FOM： Flight Operations Manual
FPI： Fluorescent Penetrant Inspection
GND： Ground
HPC： High Pressure Compressor
HPT： High Pressure Turbine
IAS: Indicated Air Speed
ICAO： International Civil Aviation Organization
JST： Japan Standard Time
LCF: Low Cycle Fatigue
LPC： Low Pressure Compressor
LPT： Low Pressure Turbine
MIL： Military
MOLIT: Ministry Of Land Infrastructure and Transport
MRE： Material Review Engineer
NAS： National Aerospace Standard
NDI： Non Destructive Inspection
NTSB： National Transportation Safety Board (U.S.A.)
PA: Public Address
PF： Pilot Flying
PFD： Primary Flight Display
PIC： Pilot In Command
PM： Pilot Monitoring
POM： Pilot Operating Manual
QA Quality Assurance
QAR： Quick Access Recorder
QN: Quality Notification
RTO： Rejected Take Off
RWY： Runway
RPM： Revolutions Per Minute
SEM： Scanning Electron Microscope
SB： Service Bulletin
SDS： System Description Section
SPEC: Specification
TO/GA Take Off / Go Around

Unit Conversion Table

1 ft: 0.3048 m
1 kt: 1.852 km/h（0.5144 m/s）
1 nm: 1,852 m
1 in: 25.40 ㎜
1 μm： Micro meter (10-6 m)
1 psi: 6.895kPa
 CONTENT

1. PROCESS AND PROGRESS OF INVESTIGATION ............. 1

1.1 Summary of the Accident ......................................................................... 1
1.2 Outline of the Accident Investigation ...................................................... 1
1.2.1 Investigation Organization...................................................................... 1
1.2.2 Representatives of the Relevant States .................................................. 1
1.2.3 Implementation of the Investigation ...................................................... 1
1.2.4 Provision of Factual Information to the Civil Aviation Bureau............. 2
1.2.5 Comments from the Parties Relevant to the Cause of the Accident...... 2
1.2.6 Comments from the Relevant States ...................................................... 2

2. FACTUAL INFORMATION .................................................... 3

2.1 History of the Flight ................................................................................. 3
2.1.1 History of the Flight Based on ATC Communication Records, FDR
Records and CVR records ................................................................................. 3
2.1.2 History of Evacuation based on Video Records ...................................... 5
2.1.3 Statements of the Crew Members, Air Traffic Controllers and others . 5
2.2 Injuries to Persons ................................................................................. 10
2.3 Damage to the Aircraft .......................................................................... 10
2.3.1 Extent of Damage .................................................................................. 10
2.3.2 Damage to the Aircraft Components..................................................... 10
2.4 Information Relevant to Damaged Properties other than the Aircraft .. 10
2.5 Personnel Information ........................................................................... 10
2.6 Information Relevant to the Aircraft ..................................................... 11
2.6.1 Aircraft ................................................................................................... 11
2.6.2 Engine .................................................................................................... 11
2.6.3 Fuel and Lubricating Oil ....................................................................... 22
2.7 Meteorological Information .................................................................... 22
2.8 Runway Information .............................................................................. 22
2.9 FDR and other Information ................................................................... 23
2.10 Accident Site Information .................................................................... 23
2.11 Details of Damage of the Airframe and Engine ................................... 24
2.12 Information of Firefighting .................................................................. 25

Table of contents ( i )
2.12.1 Emergency System at Airport when Aircraft Accident occurred ....... 25
2.12.2 History of Firefighting ......................................................................... 26
2.13 Information on Rescue and Evacuation Guidance ............................... 26
2.13.1 Rescue and Evacuation Guidance By Airport Fire Station Staff....... 26
2.14 Information on Tests and Researches .................................................. 27
2.14.1 Investigation of the Engine ................................................................. 27
2.14.2 Emergency Evacuation Slide ............................................................... 32
2.15 Information of Organization and Management.................................... 36
2.15.1 Operation Manual and rules of the Company .................................... 36
2.15.2 Emergency Evacuation Training of Flight Crew ................................ 40
2.15.3 How to inform the emergency evacuation of the Company to all
passengers ....................................................................................................... 40
2.16 Additional Information ...................................................................... 41
2.16.1 Fire Extinguishing System of the Aircraft ....................................... 41

3. ANALYSIS ............................................................................... 42
3.1 Qualification of Personnel ...................................................................... 42
3.2 Aircraft Airworthiness Certificate ......................................................... 42
3.3 Relations to the Meteorological Conditions............................................ 42
3.4 Fracture of the 1st stage HPT Disk ........................................................ 42
3.4.1 Factor to initiate a step at U-shaped groove ........................................ 42
3.4.2 Causes to initiate Cracks ....................................................................... 43
3.4.3 Causes to fracture .................................................................................. 44
3.4.4 Causes to fail to find a step in U-shaped grooves ................................. 44
3.4.5 Causes to fail to discover cracks propagating from U-shaped grooves 45
3.5 Damage of the No.1 Engine ................................................................... 46
3.5.1 Damage of the HPT Case....................................................................... 46
3.5.2 Damage to Fuel Oil Heat Exchanger .................................................... 47
3.5.3 Damage to other parts ........................................................................... 47
3.6 Engine Fire ................................................................................................ 47
3.6.1 Progress of the Fire Breakout from No.1 Engine ................................. 47
3.6.2 Extinguishing Engine Fire .................................................................... 48
3.7 Emergency Evacuation ........................................................................ 48
3.7.1 Decision of the PIC................................................................................. 48
3.7.2 Actions Taken by Flight Crew ............................................................... 49
3.7.3 Actions Taken by CAs ............................................................................ 50

Table of contents ( ii )
3.7.4 Publicification of Emergency Evacuation to Passengers ..................... 50
3.7.5 Deployment of Slides ............................................................................. 51

4. CONCLUSIONS....................................................................... 52
4.1 Summary of Analysis ................................................................................ 52
4.2 Probable Causes ........................................................................................ 55

5. SAFETY ACTIONS .................................................................. 56

5.1 Actions Taken by FAA .............................................................................. 56
5.2 Actions Taken by Engine Manufacturer .................................................. 56
5.3 Actions Taken by the Company................................................................ 57
Appended Figure 1: FDR Record .................................................................... 59
Appended Figure 2: Estimated Route of Take-off Roll .................................. 60
Appended Figure 3: Three-view drawing of Boeing 777-300 ........................ 61
Attachment 1: Records of CVR, FDR, and Video........................................... 62
Attachment 2: FPI (SPOP-84) PROCESSING of KAL .................................. 67
Attachment 3: The No.1 Engine ..................................................................... 68
Attachment 4: Photos of Fractured Surface .................................................. 78
Attachment 5: Striations of a Test piece at the crack ................................... 79

Table of contents ( iii )

 1. PROCESS AND PROGRESS OF INVESTIGATION

1.1 Summary of the Accident

On Friday, May 27, 2016, a Boeing 777-300, registered HL7534, operated by
Korean Airlines Co.,Ltd, as the scheduled Flight 2708 of the company, flight crew had a
rejected takeoff on runway 34R at the Tokyo International Airport(Haneda) during a
takeoff roll to Gimpo International Airport, Seoul, Republic of Korea because there was
a warning to indicate a fire from the No.1 (left) engine activated at around 12:38, the
flight crew stopped the aircraft on the runway, and conducted an emergency evacuation.
There were 319 people in total on board, consisting of the PIC, sixteen other crew
members, and 302 passengers. Among them, 40 passengers in total, were slightly
injured.

1.2 Outline of the Accident Investigation

1.2.1 Investigation Organization
The Japan Transport Safety Board (JTSB) designated an investigator-in-charge
and three investigators on May 27, 2016 to investigate the accident.

1.2.2 Representatives of the Relevant States

An accredited representative and advisers of Republic of Korea, as the State of the
Registry and the Operator of the aircraft involved in the accident, and an accredited
representative and advisers of the United States of America, as the State of Design and
Manufacture of the Aircraft and Engines involved in the accident, participated in the
investigation.

1.2.3 Implementation of the Investigation

May 27 to June 1, 2016 Site investigation,
aircraft and engine examinations and interviews
June 4 to 7, 2016 Engine teardown investigation at a repair shop of
the engine manufacture
June 9, 2016 Interviews
June 13 to 16, 2016 Turbine disk Inspection at the laboratory of
NTSB of the United States of America
June 16, 2016 Interviews
June 22 to June 24, 2016 Turbine disk Inspection at a facility of the Engine

1
 manufacturer
July 19 to July 22, 2016 Investigation on a manufacturing process of the
turbine disk and interviews
August 24, 2016 Interviews
October 10 to 14, 2016 Interviews, Investigation of the Engine repair shop
and Aircraft maintenance factory

1.2.4 Provision of Factual Information to the Civil Aviation Bureau

Based on the engine teardown investigation at the repair shop of the engine
manufacturer, because of evidence that a 1st stage high pressure turbine disk was
fractured and pieces penetrated through an engine case, on June 18, 2016, to the Civil
Aviation Bureau and the users of the same type of engines, the information was provided
concerning the issuance of the recommendations to inspect the 1st stage high pressure
turbine disk. Furthermore, the later investigation revealed that during the takeoff roll,
because the 1st stage high pressure turbine disk had a partial rim separation, the case
and an engine cover were damaged and at the same time as cracks were initiated at
engine parts (fuel-oil heat exchanger) by the impact due to the partial rim separation of
the 1st stage high pressure turbine disk, it was confirmed that the fuel and engine oil
leaked from the location including the engine cover caused fire at outside of the fire zone,
the JTSB provided this information to the Civil Aviation Bureau on November 8, 2017.

1.2.5 Comments from the Parties Relevant to the Cause of the Accident
Comments were invited from parties relevant to the cause of the accident.

1.2.6 Comments from the Relevant States

Comments were invited from the relevant States.

2
 2. FACTUAL INFORMATION

2.1 History of the Flight

On Friday, May 27, 2016, an Boeing 777-300, registered HL7534 (hereinafter
referred to as "the Aircraft"), operated by Korean Air Lines Co.,Ltd (hereinafter
referred to as "the Company") was scheduled to fly to Gimpo International Airport
Seoul, Republic of Korea from Tokyo International Airport(Haneda) as a scheduled
Flight 2708 of the Company.
There were 319 people in total on board, consisting of the Pilot-in-Command
(hereinafter referred to as "the PIC"), sixteen other crew members, and 302
passengers. The PIC sat in the left seat as PM1*1 and the First Officer (hereinafter
referred to as "the FO") in the right seat as PF*1 in the cockpit.
According to the records of the flight data recorder (hereinafter referred to as
"FDR"), the air traffic control (hereinafter referred to as "ATC") communication and
the cockpit voice recorder (hereinafter referred to as "CVR"), and the statements of
crew members, a local controller of Tokyo airport traffic control tower (hereinafter
referred to as "the Tower") and others, the history of the flight up to the accident
was summarized as below.

2.1.1 History of the Flight Based on ATC Communication Records, FDR

Records and CVR records
Around 12:36:40 The Aircraft entered Runway 34R.
Around 36:53 The Tower cleared the Aircraft for takeoff.
Around 37:04 The Aircraft had its thrust levers advanced forward and
commenced a takeoff roll as N12 increased.
Around 37:34 Along with a decrease of N1 of the No.1 engine, the
warning for malfunction of the No.1 engine was activated.
Around 37:35 The Aircraft reached 119 kt.
Around 37:35 Thrust levers were moved backward.
Around 37:36 Pressure for brake of the Aircraft started to increase.
Around 37:37 The Tower reported a fire breakout from the No.1 engine
and order an emergency stop.
Around 37:38 The first FIRE BELL from the No.1 engine was activated.

1 “PF (Pilot-Flying) and PM (Pilot-Monitoring)” are the terms to identify pilots on the basis of
role sharing when operating an aircraft by two pilots: The PF is mainly in charge of aircraft
control and the PM is mainly in charge of monitoring of the aircraft in flying status, cross-
checking of PF’s operations and performing tasks other than flying.

2 “N1” indicates RPM of fan and low pressure compressor at dual-spool jet engine. In addition, N
is a symbol to indicate rpm.

3
 Around 37:39 The FO called-out “SPEED BRAKE up, the No.2 engine
reverse”.
Around 37:45 The Aircraft reported to reject a takeoff (RTO) to the
Tower.
Around 37:50 The Tower requested a dispatch of fire engines via crash-
phone at Airport office.
Around 37:51 The ground speed of the Aircraft became 0.
Around 37:59 The PIC ordered cabin attendants “Crew at the station”.
Around 38:01 The PIC and the FO commenced the checklist for the
engine stop.
Around 38:13 Cut off fuel to the No.1 engine.
Around 38:20 ENGINE FIRE BOTTLE No.1 – OPEN to extinguish the
No.1 engine fire.
Around 38:27 The fire warning from the No.1 engine was released.
Around 38:51 Two chemical fire engines were dispatched from the Fire
Department East Building.
Around 38:51 The Aircraft reported that the fire was contained.
Around 40:40 The second FIRE BELL from the No.1 engine was
sounded.
Around 40:52 The Aircraft reported to the Tower that there was a
message of fire breakout displayed on the No.1 engine. At
this time, the first two fire engines had arrived at the
scene and had commenced the fire-fighting operation.
Around 40:59 Firefighting for the No.1 engine (the second time).
Around 41:11 Fire warning for the No.1 engine was released. The FO
reported to the PIC that the fire was out, again.
Around 42:07 The third FIRE BELL from the No.1 engine was activated.
Around 42:13 The Aircraft reported to the Tower that because the
message of fire breakout from the No.1 engine was on and
the message did stayed on, the emergency evacuation from
the right hand side of the Aircraft would be required.
Around 42:37 The PIC called for the emergency evacuation checklist of
the FO.
Around 42:51 The PIC activated the emergency evacuation signal switch
ON.
Around 43:03 The PIC addressed the emergency evacuation to cabin via
PA3.
Around 43:14 The PIC called for the checklist of the FO, again.
Around 43:25 The FO commenced to perform emergency evacuation the
checklist.
Around 43:45 Cut off fuel to the No.2 engine was operated.
Around 43:48 FDR stopped to record.
Around 43:50 The Aircraft reported to the Tower to execute an
emergency evacuation.
(See Appended Figure 1. Recordings of FDR and Attachment 1.Recordings of CVR,

3
“PA” is an abbreviation of Public Address and means a broadcasting system.

4
 FDR and Video.)

2.1.2 History of Evacuation based on Video Records

Around 43:17 L1 and R1 door opened and the slide deployed normally
and took upright positions.
Around 43:22 R5 door opened and the slide deployed but did not take
upright position.
Around 43:25 R3 door opened and the slide deployed normally and took
upright position.
Around 43:27 R4 door opened and the slide deployed but did not take
upright position.
Around 43:29 R2 door opened and the slide deployed normally and took
upright position.
Around 43:50 The first passengers evacuated via R3 door.
Around 44:04 The slide at R4 door re-positioned forward.
Around 44:06 Passengers commenced the evacuation from R4 door.
Around 47:04 The last passenger evacuated via the R2 door.
It took three minutes and 47 seconds, since when L1 door had been opened
first to the time when passenger who was thought to be the last had evacuated via
R2 door. Thereafter, the PIC had evacuated via R1 as the last of all, but because the
video could not confirm this situation, the time to have completed the evacuation of
all people was not verified.
Time correction for Video was performed by matching the estimated ground
speed of the Aircraft based on the analysis of Video with the ground speed in FDR.
The time in Video may contain an error of approximately 2 seconds as maximum.
(See Attachment 1. Recordings of CVR, FDR and Video.)

2.1.3 Statements of the Crew Members, Air Traffic Controllers and

others
(1) The PIC
At around 12:20, the PIC started the engine and taxied the Aircraft to Runway
34R. During this time, the engine, instruments and others had no trouble, the PIC
did not feel any vibration, odd sound or odd smell.
The FO was PF and the PIC was PM. The Tower cleared the Aircraft for
takeoff, therefore, the PIC placed his hand over thrust levers as preparing Reject
Takeoff (hereinafter referred to as “RTO”). The PIC called “80 kt Hold”.

5
 When the airspeed was at approximately 100 kt, “bang “ sound was heard,
and because the Aircraft drifted to left slightly, the PIC decided RTO, exchanged the
roles with the FO to take a control and immediately moved the thrust levers to idle
position. The V14 of the Aircraft at that day was 122 kt, the speed when moving the
thrust levers at idle position was approximately 110 kt, therefore, there were margin
in speed by approximately 10 kt till the V1 speed.
During the time to stop the Aircraft, lights were on the fire warning and the
sound of warning was confirmed. Furthermore, the EICAS5 message of “FIRE ENG
L (fire on the left engine)” was also confirmed. As the Aircraft stopped, the PIC had
checked the left thrust lever of being at the idle position and cut off the left side fuel
control switch. At this time, the FO pulled the fire handle, but still the message of
“FIRE ENG L (fire on the left engine)” was displayed, therefore, discharged the first
fire extinguisher bottle and did the time check. After a short time, the message went
out. As the FO reported the Tower “RTO on Runway 34R”, the Tower replied that
the fire engines were dispatched to the Aircraft. After the message went out, the
PIC instructed the FO to set the parking brake and told him “Fire was gone. That’s
OK.” Considering the possibilities of emergency evacuation, the PIC made PA to
cabin attendants “Crew at the Station”, then performed the memory item6. The fire
warning went out temporarily, but because after ten seconds, the light was turned
on again, he discharged the second fire extinguisher bottle. The fire warning went
out once, but immediately the light was turned on again, therefore, he decided to
perform the emergency evacuation. When discharging the fire extinguisher bottle
twice, the PIC saw the fire engines ahead, but since it was still further away, he felt
that the emergency evacuation must be carried out in a hurry. The PIC called for
the emergency evacuation checklist of the FO to hurry, but the checklist was not
found at the specified location, he could not read out the checklist, immediately.
While the FO was looking for the checklist, the PIC thought that the
emergency evacuation should be carried out in a hurry, therefore, he performed the
engine shut down procedure from his memory.
After the PIC completed the engine shut down procedure, he made PA to
evacuate via right hand side slides. And then, the FO completed the emergency
evacuation checklist by reading out the tablet (hereinafter referred to as “the

4
“V1” is the maximum speed that an operator is able to start RTO operation at serious events on an engine or
others affecting the continuous safe flight.
5 “EICAS” is the system to display the condition of engine and aircraft in integrated manner, and various

troubles by the messages on a display

6
Performing “the Memory Item” means to carry out prompt countermeasures based on the memory in an
emergency situation without seeing checklist.

6
Tablet”) and evacuated to the cabin.
More than a half of passengers still could not be evacuated and their movement
were slow, then the PIC urged them to evacuate from the right hand side following
the cabin attendants’ instruction to stay calm in order not to be panic because the
fire were suppressed completely and it was all right.
After receiving the report from the purser that all passengers were evacuated,
first the FO evacuated, then after checking that no one remained in the Aircraft, the
PIC evacuated from R1 slide. After the evacuation, following the instruction by the
firefighter, the PIC left the vicinity of the Aircraft and moved to the direction of the
sea.
(2) The FO
When entering the runway, everything was normal. The FO was PF. The PIC
was PM and communicated with ATC (Air Traffic Control). According to the
regulation of the Company, the PIC should take the thrust levers up to the V1, after
the FO pushing the TO/GA7 switch, the PIC took over the charge of thrust levers.
Ten to fifteen seconds after hearing the call “80 kt, Hold” by the PIC, the PIC
took over the control due to the trouble and stopped the Aircraft, then I saw the
EICAS message of “FIRE ENGINE L”. The PIC stopped the No.1 engine, set the
parking brake and declared “Crew at the station”.
The PIC and the FO performed the memory item. After three to five seconds
from the discharge of the first fire extinguisher bottle, the message “FIRE ENG L”
went out. Reported to the Tower that the fire was out. Then, after five to ten seconds,
the same message was reappeared, therefore the second fire extinguisher bottle was
discharged, and three to five seconds later, again the message was out. But, again
five to ten seconds later, the same message was reappeared, therefore the PIC
decided to evacuate from the right hand side.
Because the FO was called for the emergency evacuation checklist of the QRH8,
he looked in the box at right side where QRH is normally stored but could not find,
then looking through the box at left side and the FO’s flight bag, but there were no
finding again and the FO was confused. Later on, he recalled the tablet had the
checklist and read out the emergency evacuation checklist in the tablet. The PIC
told the Tower to evacuate at Runway 34R. The PIC clearly instructed to evacuate

7 “TO/GA switch” means the switch attached to thrust lever and relating to an auto-throttle.
When pressing at the time of takeoff, it transits to “N1 mode”, the thrust lever is advanced to
takeoff thrust, when pressing during the approach, it transits to “GA” mode, and then “go
around N1” thrust is set.
8 “QRH” is a booklet in checklist styles publishing Normal Operation and Abnormal / Emergency

Operation from FCOM (Flight Crew Operation Manual).

7
from the right hand side. However, after exiting from the cockpit, the FO was
confused that the L1 slide was deployed, but the FO confirmed that no one evacuated
from there.
(3) Chief Purser and Cabin Attendants
Chief Purser felt sure that there would be emergency evacuation to be carried
out based on the report of a cabin attendant in charge of L3 who found the smoke at
outside of the Aircraft, and she judged that an evacuation from R1 slide would cause
no problem because no smoke was seen there as checking outside through the
window at Door L1 and the passengers were almost to its capacity of the cabin, prior
to the emergency evacuation signal from the PIC and the announcement to
“Evacuate form the right hand side” via PA.
Based on the emergency evacuation signal from the PIC, L1 slide deployed, but
as seeing outside through the window of L1 Door, because the fire from the No.1
engine was quite obvious and there were fire engines parking at the exit of L1 slide,
the L1 slide was not used as the result. As announcement of not carrying baggage
and evacuating from the right side were made and cabin attendants kept shouting
at the passengers that do not carry baggage and take off the high heel, but many
passengers evacuated with their carry-on baggage.
Because the personnel in charge of R5 operated the slide manually but it did
not deploy normally, she blocked R5 and guided passengers to R4 to evacuate.
Personnel in charge of R4 tried to stop passengers to evacuate because the R4 slide
was blown by the wind at first,
but as the slide re-positioned
itself, the personnel started the
evacuation using this slide.
Furthermore, because R5 could
not be used therefore passengers
gathered to R4, the passengers in
the rear were instructed to use
the left side aisle to evacuate from
the right front side door. Figure 1. Door Layout for Emergency Evacuation
(4) Air Traffic Controller
The ATC cleared the Aircraft for takeoff from Runway 34R. Until then, the
ATC did not feel any abnormality. As the Aircraft commenced the takeoff ground
roll, because a fire from the No.1 engine near Taxiway C3 was seen, the ATC
reported to the Aircraft about the fire broke out from the No.1 engine and ordered

8
to reject the takeoff.
The ATC requested fire engines to be dispatched via crash phone. Before the
full stop as the ATC thought, we received the report from the Aircraft about their
RTO. At the time, a smoke was coming out of the Aircraft but not seeing fire. Later
on, the pilot reported that the fire message was displayed, so we reported back to
them that the fire engines were already dispatched. Then, the Aircraft reported that
the fire was out. And soon, the Aircraft reported that the No.1 engine fire message
had annunciated. At this time, already the first fire engine arrived at the site and
started to fight fire.
After one to two minutes, the Aircraft reported that “the fire message had
annunciated again (third time), we require the emergency evacuation from the right
hand side”. About 10 seconds after this, the slide was deployed. Shortly, the pilot
reported that “MAYDAY, we evacuate” and at the same time we saw the passengers
starting the evacuation. After this, as trying to confirm whether the Aircraft shut
down the right engine or not, but there was no response and we could not confirm
that there were any pilot left in the cockpit.
(5) Passenger A
According to the Passenger A who sat near the rear of the right wing, the male
cabin attendant shouted in Korean “run, pronto” and the door opened suddenly.
Hearing this, the passenger were getting into panic. As soon as the door opened, the
passengers were rushing to evacuate. There were no cabin attendant at the R3 door
near the right wing. The Passenger A went R3 and had companion escape ahead,
but the companion had leaped out to the runway because there were no one to assist
at the bottom of the slide. As following, the Passenger A was evacuated as being
pushed from the behind, hit his right knee onto the runway and suffered a bruise.
(6) Airport Firefighter (Airport Security Section and Disaster Prevention and Air
Safety Foundation)
According to an airport firefighter who worked at the east fire station, when
the Aircraft departed, as confirming the sound of “bang ” to know something
abnormal event occurred, the firefighter started to prepare for the dispatch prior to
the crash phone ringing. When entering the runway, the flame from the No.1 engine
was confirmed. Within about two minutes since the dispatch, the firefighter arrived
at the site. Judging that the water could be discharged from downwind side,
deployed the fifth truck at the left wing tip and the third truck at the rear of the
No.1 engine to start to discharge water. There were smoke coming out of the No.1
engine, the flame could be seen at the engine cowling. Later on, the forth fire engine

9
from the west fire station started to discharge water from forward of the No.1 engine,
and the third and fifth truck stated to extinguish the fire by using dry chemical
which could reach the inside of the engine. After the fire engines from the Tokyo
Metropolitan Fire Department arrived, we reported the situation and then dealt
with logistics support (be on guard).
According to the station firefighter who guided the evacuation, immediately
after the initial firefighting activities were started, the emergency evacuation was
commenced, but there were no passenger to assist at the bottom of the slide, the
firefighters took the charge of assisting the passenger at the bottom of R1, R2 and
R4 slides. There were none to assist at the R3 slide. After the evacuation, as the
passengers were shooting pictures near the Aircraft and making telephone calls,
they did not respond to the guide or instruction given by the airport firefighters,
therefore, they were guided by the guide-pointing rod to the perimeter road at sea
side. Many of passengers were carrying their baggage, and some of them carried
large suitcases.

This accident occurred on Runway 34R of Tokyo International Airport

(N35º32’, E139º48’) and at around 12:28 on May 27, 2016.
(See Appended Figure 2. Estimated Taxiing Route Map)

2.2 Injuries to Persons

40 of passengers suffered minor injuries like bruises, scratches and others.

2.3 Damage to the Aircraft

2.3.1 Extent of Damage
Slightly Damaged

2.3.2 Damage to the Aircraft Components

(1) The No.1 engine: Damaged (See Attachment 3. Condition of the No.1 engine)
(2) The left wing Flaps: Penetrating marks

2.4 Information Relevant to Damaged Properties other than the Aircraft

None

2.5 Personnel Information

(1) PIC: Male, Age 49

10
 Airline Transport pilot certificate (Airplane) January 3, 2002
Type rating for Boeing 777 November 24, 2009
Class 1 aviation medical certificate
Validity July 31, 2016
Total flight time 10, 410 hours and 05 minutes
Flight time in the last 30 days 32 hours and 00 minutes
Total flight time on the type of the aircraft 3, 205 hours and 22 minutes
Flight time in the last 30 days 32 hours and 00 minutes

(2) FO: Male, Age 41

Commercial pilot certificate (Airplane) April 30,1999
Type rating for Boeing 777 July 7, 2016
Instrument flight certificate April 30, 1999
Class 1 aviation medical certificate
Validity May 13, 2017
Total flight time 5, 788 hours 16 minutes
Flight time in the last 30 days 40 hours 00 minutes
Total flight time on the type of the aircraft 2, 531 hours 18 minutes
Flight time in the last 30 days 40 hours 00 minutes

2.6 Information Relevant to the Aircraft

2.6.1 Aircraft
Type Boeing 777-300
Serial number 27950
Date of manufacture January 4, 1998
Certificate of airworthiness AS071213
Validity Since September 21, 2012 until discontinued/limited
Category of airworthiness Aircraft Transport T
Total flight time 64, 028 hours 00 minutes
Flight time since IAA (every 500 hours, performed on April 12, 2016)
412 hours 00 minutes
(See Appended Figure 3.: Three-view drawing of Boeing 777-300)

2.6.2 Engine
Location to be worked No1 (Left) No2(Right)
Type PW4090

11
 Serial number P222221 P222017
Date of manufacture October 23,2004 February 5, 1997
Total flight times 41,594 hours 70,660 hours
Total number of use time 9,832 cycles 11,059 cycles

2.6.2.1 The Structure of the Engine

The engine installed on the aircraft is a dual-spool turbofan engine consisting
of a low pressure compressor and turbine (expressed in blue at Figure 2) and high
pressure compressor and turbine (expressed in red at Figure 2). From the front of
the engine, the structure consists of a 112-inch diameter fan, six-stage low pressure
compressor (hereinafter referred to as “LPC”), 11-stage high pressure compressor
(hereinafter referred to as “HPC”), annular combustor, two-stage high pressure
turbine (hereinafter referred to as “HPT”), and seven-stage low pressure turbine
(hereinafter referred to as “LPT”). (See Figure 2. Engine Structure (Schematic))

Figure 2. Engine Structure (Schematic)

2.6.2.2 HPT (high pressure turbine)

The parts of the engine structure are distinguishing by calling the parts where
is exposed to high temperature and pressure combustion gas (Figure 2 Combustor,
Turbine and Exhaust Pipe) as hot section and calling the other parts as cold section.
Especially because HPT of the hot section operates under a high temperature, rotor
blades, stator vanes, disks and other of HPT’s components are made of a heat

12
resistant alloy having excellent heat resisting properties.
HPT is a component of engine driving HPC through a high pressure shaft as
converting the high temperature and pressure combustion gas from the combustor
to rotating movement by expanding via two-staged turbine and it consists of the 1st
stage stator vane (V1), the HPT1 rotor blade (R1), the second stage stator vane (V2)
and the HPT 2 rotor blade (R2). Each HPT rotor is a disk-shaped disk with dozens
of airfoil rotor blade installed on outer periphery of the disk.
HPT rotor is installed on a shaft (axis) and rotating in high speed between
stator vanes. During its operation, the HPT rotor receives thermal stresses caused
by the high temperature and pressure combustion gas, and centrifugal forces due to
the high speed rotation. Per the Type Certificate Data Sheet (TCDS), the maximum
exhaust gas temperature (EGT) is 675°C and the maximum HPT rpm is 10,850 RPM.
Furthermore, when starting an engine and making the engine the maximum
thrust to takeoff, the outer diameter of the disk near the blades are heated rapidly
and it expands as the result, but because inner diameter of the disk (hub) does not
become high temperature, tensile force would be generated between outer diameter
(rim) and inner diameter. As operating for some time, the heat transmits to the inner
diameter, the temperature difference would be gone. And at next, when the engine
stops, the outer diameter of disk would be shrink rapidly due to the cooling, but the
inner diameter would not be cooled so easily, therefore, the compressive stress would
be generated between the outer and inner diameter of the disk.
As these samples, the disk receives a set of tensile stress and compressive
stress per one cycle due to difference in temperature, which generated one cycle per
one flight (start – takeoff – climb – cruise – descent – landing – stop). Count this one
set of stresses as one cycle. The one cycle time varies depending on the operating
route distance, but it should be about one to 14 hours in general.
Comparing to fatigue accumulated within a short time like vibration, the metal
fatigue caused by receiving repeatedly sets of low cycle stresses as above mentioned
is called as a low cycle fatigue (hereinafter referred to as “LCF”). Almost all disks
installed in the engine have the fatigue life limit by LCF. Therefore, the engine
manufacturer set a fatigue life for the 1st stage HPT disk of the engine installed on
the Aircraft as 13,300 cycles for its use limit.

2.6.2.3 Processing and Inspecting when manufacturing the HPT Disk

The 1st stage HPT disk has the structure with machined grooves to install the
air seal (hereinafter referred to as “U-shaped grooves”) to suppress a leak of cooling

13
air or the grooves to install rotor blades on the perimeter as machined. (See Photo
1. the HPT Disk)
Furthermore, since large stress would act on the HPT disk repeatedly, if mere
scratch or step exceeding the manufacturing allowance exists, because of notch
effect9, it could be an originating point for stress to concentrate or fatigue crack, the
machining and inspecting at the manufacturing have being controlled strictly.

Photo 1. HPT Disk

U-shaped grooves are machined for cutting and processing by vertical and
horizontal movements of the tip of the machining tool while the 1st stage HPT disk
which is a workpiece is mounted and turned on the turn-table of the vertical milling
machine. The vertical milling machine is processing automatically by computer
program control. Machine Operator (worker) monitors the status of positioning of a
workpiece, fixing, setting the machining tool and automatic processing.
Because U-shaped groove are cut from both sides of groove from the outer side
and the inner side, a machined resultant step-like trace would be initiated at the
seams from both sides. The Machine Operator, in order to remove this mismatch
step at the seams when processing final finishing of U-shaped groove, inserts 0.010
inches shim stock into a clearance between the tip on the machining tool and the
bottom of U-shaped groove while checking the situation by a fingertip touches and
visual examination, controls the manual feeder installed on the vertical automatic

9 “Notch Effect” is the phenomena that the surface of object with a notch could be initiated far
bigger stress than a flat smooth surface, when an external force works.

14
lathe for a precision finish, and sets the final processing position by adjusting the
clearance between tip of machining tool and the bottom of U-shaped groove. This
adjustment is to compensate the wear occurred at the tip of machining tool, and the
value (Z axis; vertical feeding) of the final machining position would be displayed on
the manual feeder counter for precision finishes. Inputting this value from keyboard
to vertical automatic lathe, automatic machining would be executed, but the real
final machining position is programmed to send the tip of machining tool at the
0.010 inches lower than this value in order to compensate the shim stock thickness.
According to the investigation by recreating the processing as the final
processing position without using the 0.010 inches thick shim stock and
investigating by recreating the machining by setting the condition without no gap
at the bottom of U-shaped groove initiated by the tip of the machining tool had
resulted in 0.010 inches deep step at the bottom of U-shaped groove. (See Figure 3.
Processing the HPT Disk with use of a vertical milling machine.)

Figure 3. Processing the HPT Disk with use of a vertical milling machine

After the manufacturing, a machine operator confirms that the HPT disk are
within its allowable range for manufacturing by visual inspection, touch
examination and measurement by instruments (Product Inspection). Subsequently,

15
an inspector (Inspector) confirms that the product is machined according to the
manufacturing instruction, and confirms that an inspection itself is carried out
correctly based on a work order and manufacturing drawing (Quality Inspection).
Work order and manufacturing drawing which used for inspection are
managed by the computer at technical section and the latest version are constantly
delivered to the workplace through the in-house Intranet.
The inspection method used by an inspector are specified depending on the
location of inspection such as visual inspection, touch examination, measuring by
measurement instrument and CMM (Coordinate Measuring Machine; three
dimensional measuring device). Furthermore, when the inspector judges that it is
necessary for checking the machined step, making a replica using thermosetting
rubber compound to measure the mismatch between the two machined radii in the
groove by the optical comparator. Manufacturing allowable limit is shown in
manufacturing drawings. Because the locations of U-shaped groove are not specified
as the critical inspection points (blade mounting slot, installing grooves for hub shaft
and others which possess the highly critical situation), standard manufacturing
allowable limits of 0.002 inches is applied for the 1st stage HPT disk.
Furthermore, regarding the critical inspection points, the manufacturing
drawing has the note concerning the detailed inspection, in the records column in
the work order had the records of the measured values of the product inspection and
quality inspection, but other that, the manufacturing drawing does not have note
and only to pass or fail as the result of inspection are recorded in the work order. In
addition, U-shaped groove of the 1st stage HPT disk of the No.1 engine of the Aircraft
had not been processed at the repair work after the manufacture.

2.6.2.4 Records of manufacturing the HPT disk

At the engine manufacturer, the 1st stage HPT disk receives a product
inspection by machine operator at the time of manufacturing and a quality control
by inspector as described in 2.6.2.3. According to the manufacturing records
(Inspection results recorded in the work order) of the 1st stage HPT disk of the
engine, there were no entry of measurement data and likes since U-shaped groove
is not an critical inspection point for the quality control, but there was a description
that the inspection result was approved.

2.6.2.5 Quality Control System of the Engine Manufacturer

The engine manufacturer holds Production Certification from FAA of the

16
United States of America which is the State of Manufacture. On-site inspections
were carried out regarding the facilities, the equipment, the organizations, the
personnel and the quality management system relating to the manufacture of the
engine along with confirmation of the records of the manufacturing the 1st stage
HPT disk of the Engine and as interviewing the machine-operator and the inspector
who had worked to manufacture the engine, however, the facts to be the cause of
this incident was not found. When a malfunction of being out of the allowable limit
and others is found at the quality management inspection, QN (Quality Notification)
will be issued and there is a system to make up decision to rework or to reject by the
MRE (Material Review Engineer) who is an engineer of the technical section and in
charge of quality control, however, QN was not issued for the 1st stage HPT disk of
the engine.

2.6.2.6 Non-destructive Inspection of the HPT disk

Regarding the HPT disk in order to find out signs of either damage or cracking
in advance due to the LCF described in 2.6.2.2, the engine manufacturer is
requesting users to carry out Non Destructive Inspection (NDI) when disassemble
the HPT at an engine maintenance work and to carry out Fluorescent Penetration
Inspection (hereinafter referred to as “FPI”) which is one of NDI, to the 1st stage
HPT disk. It is difficult for regular visual inspection to detect a discontinuous
damage which opens up to a surface, but this FPI is an inspection to detect the
damage using penetrant containing fluoresce. Besides, regarding the requirement
to implement FPI, well-known standard (MIL SPEC, MIL STD, NAS410, ASTM-
STD-1595 and others) or aircraft manufacturers’ engineering data specify facilities,
equipment, operator and working method relating this NDI strictly and require to
carry out as following these. The Company implements FPI as following the
specified procedure by the engine manufacturer. Outline of FPI procedure to detect
cracks is as follows;
① Apply the fluorescent penetrant to the inspection surface or soak in it, then
because of capillary phenomenon the penetrant liquid penetrates into cracks.
(Penetration Processing)
② Remove an excess of penetrant remaining on a surface.
(Removal process and Pre-Rinse)
③ After in the emulsifier to remove the penetrant from the surface of the part to
allow the penetrant to remain in the crack, rinse and dry the part.
(Emulsification apply)

17
④ Spray developer and suck up the penetrant from cracks. (Developing processing)
⑤ When illuminate with Ultra-Violet Light,the cracks emit fluorescence.
(Inspection)
(See “Attachment No.2. FPI manual of the Company” for the detailed procedure of
FPI.)

Figure 4. Outlined procedure to detect surface flaws by FPI

According to the FAA advisory circular (AC43-3 “Nondestructive Testing
Aircraft”), FPI is effective to detect flaws at surfaces of non-magnetic metal,
ferromagnetic metal and non-metallic materials.

2.6.2.7 Major Operating History of the No. 1 engine and 1st stage HPT
disk
According to the maintenance records of the engine, the major history of the
No.1 engine and the 1st stage turbine disk after their manufacturing are as shown
in Table 1. The 1st stage HPT disk of the engine was manufactured on October 28,
2004 and has a life limit of 13,300 cycles. It remained with the engine since then.
The part number was 53L121-001, SN CKLBHE5552. According to the
maintenance records, the 1st stage HPT disk had accumulated 9,832 cycles and
41,594 hours since new and the time and cycles since new of the engine and 1st stage
turbine disk are the same.
FPI for the 1st stage HPT disk was implemented at the times of manufacturing

18
the engine ①, and of disassembling HPT module10 ②, ⑥ and ⑧. Besides, the 1st
stage HPT disk of the engine was installed from the time of manufacturing the
engine, and number of cycles and use hours up to the time of accident were 9,832
cycles and 41,594 hours as the same for the both engines. In addition, when
disassembling the HPT module at the delivery to the engine repair shop, FPI on the
1st stage HPT disk for each time was implemented as specified by the engine
manufacturer.

Table 1. Major History of the No.1 Engine

Equipped
Use hours / number
No Machine Reason to Delivery FPI on disk
Date of cycles since its
. Number Overhaul Destination Yes; ○, No; ×
manufacture
Location
HL7573 right ○(when inspecting the
October 28, 0 hour
① (New product as newly
2004 0 cycles
Machine) manufactured)
EGT*; Engine
August 13, ○(when disassembling 11,168 hours
② High Manufacturer
2007 HPT module) 3,043 cycles
temperature Repair Shop
March 5,
③ HL7531 right
2008
Engine
June 17, Internal ×(not disassembling of 12,691 hours
④ Manufacturer
2008 damage HPT module) 3,262 cycles
Repair Shop
November
⑤ HL7573 left
18, 2008
The
March 31, Internal ○(when disassembling 22,120 hours
⑥ Company’s
2011 corrosion HPT module) 6,050 cycles
Repair shop
September
⑦ HL7734 left
6, 2011
EGT; The
June 29, ○(when disassembling 36,825 hours
⑧ High Company’s
2014 HPT module) 8,023 cycles
Temperature Repair shop

November
⑨ HL7534 left
12, 2014

May 27, This Accident 41,594 hours

⑩
2016 May 27, 2016 9,832 cycles

2.6.2.8 The Engine Repair Shop of the Company

The Company has received the approval for the maintenance work from
MOLIT of Republic of Korean as the State of Registry. Upon implementing the on-
site investigation concerning facilities, equipment, organization, personnel, and the

10
“Module” is maintenance units which forms the engine structure parts in order to improve the
maintainability.

19
quality management system relating to the FPI at the engine repair shop of the
Company, we had checked the maintenance records at the time to carry out FPI on
the 1st stage HPT disk of the engine, but we could not find any fact to be the cause
of this accident.
The FPI at the Company was provided exactly as the method specified by the
Engine Manufacturer and the FPI is implemented by an operator and an inspector
who hold the qualification and capabilities required for FPI.
Operator carries out the inspection (NDI inspection) following the work order.
Inspector carries out QA inspection to guarantee the quality of inspected product by
confirming the inspection carried out properly by the operator following the
procedure and work process specified at the last.
On June 29, 2014 (hereinafter referred to as “at that time”), the inspector
(hereinafter referred to as “the Inspector”) who carried out FPI inspection on the
last of the 1st stage HPT disk of the engine was working as an inspector for about 25
years after working as operator at the Company for about 10 years. At that time,
the Inspector had accumulated about 27 years of experience at the Company. Then,
the Inspector was retired and is now working as cleaning aircraft parts at a company
relating to the Company. The Inspector made a statement concerning FPI at that
time of the investigation of this accident.
At that time, an operator (hereinafter referred to as “the Operator”) who
carried out the last FPI inspection on the 1st stage HPT disk of the engine was
working for cleaning aircraft parts at the Company for three years and then became
the Operator at the engine shop of the Company. At that time, the years of
experience at the Company was about 19 years. When investigating, the Operator
became an inspector working for the Company and he was the personnel who had
executed FPI in real at the time of investigating the FPI implementing system of
the Company as described later in 2.14.12, . At that time, the inspector and the
operator held the appropriate qualifications (FPI level11 Ⅱ) and passed the annual
Eyesight test provided by the Company.
According to the Operator, at first confirming the work order and executing
the visual checking a workpiece for its part number and its serial number, and then
the operator would carry out the work as following the work procedure (the

11 “FPI level” is an inspection qualification level that is specified by nation’s regulatory standard
authorized depending on an experience time and a degree of difficulty per the type of NDI and
Level 1 becomes to Level 2 then Level 3 as the highest. After the qualification authorized, in order
to maintain the competence, it is necessary to have experience times more than specified
inspecting time and pass an Eyesight test, and others. With being qualified more than level 2, one
can inspect alone.

20
procedure specified by the engine manufacturer).
According to the engine manufacturer, the inspection in a dark room, as an
inspector would take a few minutes to adjust to the darkness prior to the work, the
Inspector had been taking for about two to five minutes and the Operator for about
ten minutes.
The Operator had an experience to find out a crack at the cooling hole of the
2nd stage HPT disk of the same type engine in a past. The Inspector had no
experience to find out any crack at his inspecting turbine disk of the engine.
Furthermore, since the Operator became the Inspector of the Company at the
time of investigation who carried out FPI in real at the time of investigating the FPI
system of the Company, and made an statement described in 2.14.1.2, the following
are based on the statement of the Operator;
① When carrying out inspection, an inspector could inspect turbine disk done alone,
but depending work load, he/she would inspect it with other parts as combination.
Turbine disk inspection is done on the disk suspended from the belt traveling while
rotating the disk, and at first the whole disk is seen from the front. These are done
because the forward side of disk has higher level of critical. Time duration to inspect
a disk is about 30 minutes but it could be about one hour to inspect when taking a
longer time.
② When we differed in our opinion with the Technical section, we discuss with other
inspector and carry out the inspection all over again if required. Prior to the final
inspection done by the inspector, operator carries out inspection as following the
manual.
③ Regarding the inspection results, the differences in opinion do not occur that
many, up to now, but we had a cleaning section to re-clean because of insufficient
cleaning.
④ From 2004 to the present, there are no changes in work environment concerning
work quality or measures on malfunction.
⑤ The human factors training is carried out periodically for all member in the
engine shop as the subject. As the training is targeting for foreseeing operational
errors, restraining over-confidence and others, it is very useful for everyday
operations.
⑥ Regarding points to inspect at FPI of the engine with care, there is nothing
particular, but it is important to see the points where need cautions with care at the
inspection by confirming this according to the manual. We have not questioned the
engine manufacturer and others regarding manual, till now.

21
2.6.2.9 Records of FPI on the 1st stage HPT Disk
As described in 2.6.2.5, the 1st stage HPT disk of the engine after its
manufacturing had received FPI four times in total, which were implemented at the
time of engine manufacturing, at the engine manufacturer repair shop and the
engine repair shop of the Company, but any of these records did not contain
descriptions regarding any malfunction such as cracks and others.

2.6.3 Fuel and Lubricating Oil

The fuel and the lubricating oil used for the Aircraft were Jet A-1 and Mobil
Jet Oil II, respectively. The spontaneous ignition temperature of Jet A-1 is
approximately 210 ºC and the flash point is approximately 40 ºC.

2.7 Meteorological Information

Aeronautical weather observations for the Airport around the time of the
accident were as follows:
(1) METAR (Aerodrome routine meteorological report)
12:30；Wind direction: 060°, wind speed: 20 kt, visibility: more than 10 km
Cloud Amount: 1/8-2/8, Type Cumulus, Cloud base: 800 ft
Amount: 3/8-4/8, Type Cumulus, Cloud base: 1,300 f
Amount: 5/8-7/8, Type Cumulus, Cloud base: 1, 800 ft
Temperature 19℃; Dew point 18 ℃; Altimeter setting (QNH) 29.80 inHg
(2) SPECI (Aerodrome special meteorological report)
12:52 ; Wind direction: 060°, wind speed: 18 kt, visibility: 25 km
Cloud Amount: 1/8-2/8, Type Cumulus, Cloud base: 800 ft
Amount: 3/8-4/8, Type Cumulus, Cloud base: 2,000 f
Amount: 5/8-7/8, Type Cumulus, Cloud base: 4, 000 ft
Temperature 19℃; Dew point 17℃; Altimeter setting (QNH) 29.80 inHg

2.8 Runway Information

Tokyo International Airport has four runways as Runway A (16R/34L, 3,000m
by 60 m in width), Runway B (04/22, 2,500 m by 60 m in width), Runway C (16L/34R.
3,360 m by 60 m in width), and Runway D (05/23, 2,500 m by 60 m in width), and
this accident occurred at Runway C. (See Appended Figure 2. Estimated Taxiing
Route)
The runway had received the periodic runway inspection from 11:00 to 11:10,

22
however, no anomaly to cause obstacle to its operation was found.

2.9 FDR and other Information

The Aircraft was equipped with a FDR capable of recording for a duration of
about 25 hours and a CVR capable of recording for a duration of about two hours,
manufactured by Honeywell, U.S.A. The records concerning this accident were
retained in both recorders. The time data on the FDR and CVR were corrected by
correlating the time signals on the ATC communication records with the VHF
transmission keying signals in the FDR and ATC communication records in the
CVR.

2.10 Accident Site Information

The Aircraft entered to Runway 34R from the approach taxiway C1 and
commenced the takeoff roll. At some time later, the Aircraft rejected the takeoff and
stopped on the runway at short of Taxiway C5 with heading direction of 335 º
(magnetic direction). During this time, the Aircraft run approximately 1,350 m. tire
marks due to the heavy brakes were confirmed from the stop position to
approximately 520 m south along the runway centerline and at the both sides of the
centerline. Besides, there were many rejected fragment and parts relating to the
engine found and collected over the runway originating the point about 680 m from
inside of the runway threshold to approximately 570 m north, the taxiway and the
grass land of the perimeter road. (See Figure5.Tire Marks and Scattering range of
fractured Parts.)

Figure 5. Tire Marks and Scattering range of Fractured Parts

23
2.11 Details of Damage of the Airframe and Engine
The on-scene examination at Tokyo International Airport revealed the
following details of damage of the airframe and engine. In addition, within the
following sentence, all references to position or directions, as referenced to the clock,
will be as viewed from the rear, looking forward, unless otherwise specified. Besides,
“Photo # X” means the number of photos shown in the Attachment 3.
(1) Damage of the No1 Engine were as follows;
① The left side of the engine and the external accessories were burned and sooted.
(Photo #1)
② The rear flange of the diffuser12 case at the position from about 5:30 to 6:30
o’clock was bent radially outward. The inner diffuser case had cracks between the
aft flange and the center part of the case and at about 12 o’clock, there was a piece
missing from the case.
③ The HPT case between the position of 7:30 and 9:00 along the periphery was
bent radially outward and twisted, exposing the 1st stage turbine disk and blades
and the 2nd stage turbine stator inner support.
All of the visible 1st stage turbine blades were in place in the disk were
fractured transversely across the airfoil adjacent to the blade root platform. There
were no 2nd stage turbine stator vanes visible through the hole in the HPT case.
(See Photo #7.)
④ Air Starter valve had a hole. (Photo #8)
⑤ A piece was missing from the rim of the 1st stage HPT disk.(See Photo #10.)
The missing pieces are found among the scattered debris at grassland near
Runway C.
⑥ On the body case of fuel-oil heat exchanger13, cracks and soot due to fire
damage were confirmed.
⑦ Pieces are missing along the periphery of cowling from the left rear frame of the
translating cowl along the periphery of cowling to front. (See Photo #11.)
(See Attachment 3. Status of the No.1 engine)
(2) Outboard Flap
The outboard flap had an 8-inches (20cm) long crack in the trailing edge at 48
inches (120cm) length to outboard from the inboard edge. (See Photo 2. Outboard

12 “Diffuser” is located at the aft side of compressor and is the device convert high velocity, lower
pressure airflow from the high pressure compressor to lower velocity, higher pressure airflow
prior to entering the combustor.
13 “Fuel-oil heat exchanger” is a system to warm fuel in order to prevent freezing of water in fuel

and cool the engine oil by exchanging heat between fuel and oil.

24
flap showing the cracks.)

Photo 2. Outboard flap showing the cracks

(Note: The flap position shown in this photo is differed from the takeoff flap
position at the time of accident.)

2.12 Information of Firefighting

2.12.1 Emergency System at Airport when Aircraft Accident occurred
(1) Tokyo International Airport Emergency Plan
Tokyo Airport Office has Tokyo International Airport Emergency Plan
established in compliance with ICAO Annex 14 (Aerodrome) and stipulating the
counteractions when an Aircraft accident occurs. According to the Emergency Plan,
when the accident occurs, the Airport office shall report to the relevant authorities
using emergency contact table, and request a firefighting operation, a medical
rescue operation and others, and this plan includes the training and others, too.
(2) Dispatching Airport Firefighter and Stationed Emergency Vehicle
Five chemical fire engines, one command vehicle, one water truck, one rescue
lighting vehicle and one medical transport vehicle shall be assigned at the Airport.
At the time of this accident, eleven Airport Firefighters shall ride separately on five
chemical fire engines and one medical transport vehicle and shall be dispatched.
(3) Dispatch Situation at Tokyo Fire Department

25
 Tokyo Fire Department shall dispatch 48 fire engines in total and 14
ambulance to the site and 239 firefighters in total shall carry out firefighting
operation and emergency medical aid operations.

2.12.2 History of Firefighting

Regarding the response taken by the Airport Fire Station Staff when this
accident occurred, according to the records of the airport office, the summary was as
follows;
12:38 Fire station received the dispatch request through Crash Phone from the
Tower.
12:39 Two chemical fire engines (the third car and the fifth car) were
dispatched.
12:41 The engines arrived at the site from the rear of the Aircraft and started
to fight fire.
12:42 The forth fire engine arrived at the site from the front of the Aircraft and
started to discharge water.
13:03 Airport fire station staff finished the initial firefighting operation,
handed over the firefighting operation to Tokyo Fire Department and
dealt with logistics support.
According to the records of Tokyo Fire Department, a fire suppressed14 was at
14:21 and a confirmation of fire extinguished was at 15:09.

2.13 Information on Rescue and Evacuation Guidance

2.13.1 Rescue and Evacuation Guidance By Airport Fire Station Staff
At 12:42,the eighth Fire Engine arrived at the site from the front of the Aircraft
and dealt with the Evacuation guidance for crew and passengers. At 12:43, the ninth
Fire engine arrived at the site from the Aircraft and dealt with the Evacuation
guidance. Airport Fire Station Staff assigned at the base of the emergency
evacuation slide R1,R2 and R4(hereinafter referred to as “the Slide”) and following
the confirmation of all passengers evacuation, guided the crew and passengers to a
perimeter road at the side of Runway C.
At 12:51, the twelfth fire engine (ambulance) arrived at the site and provided
the medical care to the injured people near the edge at east side of Runway C.

14“Fire Suppressed” and “Fire Extinguished” are terms used for Firefighting, “Fire Suppressed”
means that a force of fire is lost by firefighting, and “Fire Extinguished” means that the fire is out
and the condition of no more firefighting required by the firefighter.

26
2.14 Information on Tests and Researches
2.14.1 Investigation of the Engine
2.14.1.1 Analysis of fractured surface
After disassembling the No.1 engine of the Aircraft at the engine overhaul
facility, a metallurgical examination of the damaged 1st stage HPT hub (disk) and
the ejected disk fragment from the body was carried out by the NTSB and the engine
manufacturer. The findings as the results of this investigation are as follows;
The 1st stage turbine disk was complete except for a section of the rim that
was missing and the disk’s material conformed to the requirements. The missing
piece of the disk’s rim, that was about 19.68 cm (7.75 inch) as measured along the
snap inner diameter, was recovered from along the edge of the runway. The complete
fracture surface on the disk corresponded to the complete fracture surface on the
piece of the rim. The fracture surface on the disk and the recovered piece of the rim
and elliptical-shaped patterns. There were 73 blade slots in the disk and 9 blade
slots in the recovered piece of the rim. The 1st stage turbine disk has 82 blade slots.
0.010 inches (0.25 mm) deep step was confirmed all over periphery of U-shaped
groove at the aft side rim of the 1st stage HPT disk. Maximum allowable limit for
machining mismatch when manufacturing was 0.002 in (0.05 mm) as described in
2.6.2.3. As the result of the detailed investigation around the step, it was confirmed
that several cracks were dotted along the left and right step around where the 1st
stage HPT disk was fractured. A detailed visual inspection could confirm the size of
opening were about 1 mm to 4 mm respectively. (See Photo 3-1. Fractured Rim,
Photo 3-2. U-shaped Groove and Crack 1, Photo 3-3. U-shaped Groove and Crack 2,
and Photo 4. Created Replica of the Step)
Regarding the fracture surface of the ejected fragment, cracks were originating
at the machined step in U-shaped groove and was propagating from the aft side of
the 1st stage HPT disk to the front side. The fracture surface exhibited severe
damage due to the impact and others, but the beachmark 15 of a typical
characteristic of the fatigue crack fracture surface was confirmed of its initiating
along with the propagating area on crack. The size of the three thumbnail cracks
were as follows;
① large; approximately 0.608 inches deep, approximately 2.358 inches long
(15.44mm x 59.89 mm)
② medium; approximately 0.282 inches deep, approximately 1.000 inches long

15“Beach mark” is half circle pattern like shell to be seen at a macro-observation of fatigue crack
surface.

27
 (7.163 mm x 25.40 mm)
③ small; approximately 0.088 inches deep, approximately 0.489 inches long
(2.235 mm x 12.42 mm)
( See Attachment 4. Photo of fracture face)

Photo 3-1. Fractured Rim

28
Photo 3-2. U-shaped groove and crack 1

Photo 3-3. U-shaped groove and crack 2

29
 Photo 4. Created Replica of the Step

The fractured surface of the ejected piece of the 1st stage HPT disk was
damaged, however, stable striations could be counted at a part of the section with
the scanning electron microscope (SEM). The fractured surface on the disk was
unreadable because of oxidation from the fire extinguishing agent used by the fire
department. Furthermore, open the crack which was cut out a part from the disk
which had clear cracks from the remaining aft side of the disk, and analyzed the
crack surface. (See Attachment 5. Striations at the crack on the test piece)
The striation count analysis completed on the fracture surface and crack surface
estimated 2,130 cycles and 2,868 cycles of repeated stress respectively. Why these
two numbers differed from each other were because the time period to initiate
cracks were differed and cracks initiated at the initial crack propagation stage
was damaged by corrosion, heat and others. As shown in Figure 5, at the time of
accident, the disk had 9,832 cycles as the total number of cycles since the
manufacturing the disk, subtracting 2,130 cycles from these number became
7,702 cycles and subtracting 2,868 cycles from these became 6,964 cycles. Besides,
the total number of use cycles of the disk at the time of the previous inspection
was 8,023 cycles. Based on these numbers, it is possible that the 1st stage HPT
disk had cracks when it delivered to the engine shop of the Company to have the
previous inspection (November 12, 2014). (See Figure 6. Number of Cycles of the

30
 Disk estimated from the analysis of the fracture surface of rejected fragment)
Furthermore, from the numerical analysis based on the counting at the crack
surface, values of 0.088 to 0.176 inches (2.24 mm to 4.47 mm) along the surface were
obtained on the cracks at the latest inspection at the engine repair shop of the
Company.

Figure 6. Number of Cycles of the Disk estimated from the analysis of

the fracture surface of rejected fragment

2.14.1.2 Investigation of FPI implementing system by using sister disk

at the Company
For the so-called sister disk of the 1st stage HPT disk of the same engine which
was manufactured and inspected by the same operator and the same inspector, at
the engine repair shop of the Company, the status of following inspection and others
carried out by the same operator were checked and the FPI implementing system of
the Company was investigated.
(1) Inspection of the disk based on the PW Service Bulletin (SB) PW4G-112-72-342
(issued on September 23, 2016, category 5,6 (recommend)”
After conducting an acceptance test to confirm no anomaly existing at the
sister disk, creating replicas by use of rubber compounds at four locations (positions
at 12:00, 3:00, 6:00 and 9:00) of U-shaped groove at outside of the aft side disk
following the Service Bulletin regarding the above subject, the step of U-shaped
grooves are measured and confirmed to be within the allowable limits. (See Table
2.) The technical report were issued for the same type engine users to check
manufacturing error of the step exceeding the allowable limit for U-shaped groove

31
of the 1st stage HPT disk and at the quality inspection as described in 2.6.2.3, it
instructed the measuring the step by creating the replica as required by the decision
of an inspector.
Table 2.
Step of U-shaped Groove
Limit 0.0020 in max
Location of Replica
12 o’clock 0.0015 in
3 o’clock 0.0014 in
6 o’clock 0.0014 in
9 o’clock 0.0015 in
(2) FPI based on the working procedure of the engine manufacturer
As the result of FPI verification investigation using a sister disk, the sister
disk did not have any problem, and there were no crack indications.
At the time of implementing FPI on the sister disk at the engine overhaul shop
of the Company, regarding facilities, systems, procedures, operators and inspectors
did not have any disqualification in compliance with the 2.6.2.6 requirements.

2.14.1.3 The 1st Stage HPT Disk on Other Same Type Engine
According to the investigation carried out by the engine manufacturer, the step
of U-shaped groove exceeding the allowable limits was not found from the HPT disk
of the same type engine other than the engine of this accident with the
implementation of the technical report in 2.14.1.2 (1).

2.14.2 Emergency Evacuation Slide

2.14.2.1 Emergency Evacuation System
The Aircraft has the evacuation exit doors at ten locations and each door
equipped with an emergency slide /raft. This system deploys a slide automatically
as soon as the door opened at the emergency situation and let passengers and crew
to evacuate to outside of the aircraft. (See Figure 1. Door Layout for Emergency
Evacuation.)
SDS (System Description Section) in the system relating to the manual
provided by the aircraft manufacturer has the following description regarding the
deploying time of the slide and the wind speed limit at the time of deployment;
The average time for door opening and slide/raft inflation is seven seconds. The
slide/rafts operate correctly in wind as much as 25 kt.

32
2.14.2.2 Situation at Emergency Evacuation
According to the statements of the cabin attendants, they had deployed the
doors at five locations of the right side of the Aircraft and attempted to evacuate,
but the R5 door at the most rear side (hereinafter referred to as “R5 door”) was not
fully deployed. For the door at the left side of the Aircraft, L1 door at the most front
side had only the slide deployed, but it was not used. (See Photo 5.Emergency
Evacuation Slide.) Furthermore, the situation of the emergency evacuation was
recorded by the Monitoring camera at the Airport, it was confirmed that there were
passengers with baggages to evacuate and no assistances at the bottom of the slide
when the evacuation was started. Photo 6. shows the comparison of deploying R5
slide at the normal situation and at the accident.
Photo 7. shows the deploying situations of R3, R4 and R5 slides of the Aircraft
at the time of the accident. Furthermore, the R3 slide is narrower in width and
shorter in length compared to the R4 and R5.

Photo 5. Emergency Evacuation Slides

33
 Photo 6. Comparison of R5 slide deploying status

Photo 7. Deploying situations of slides at R3, R4 and R5

2.14.2.3 Information of R5 Slide

Name; B777 Evacuation Slide/Raft
Part Number; 62774-424, Serial Number; 1192
Date of Manufacture; May, 2005
Date of Previous Inspection; March 16, 2016
(Overhaul at MNE-GMP Maintenance Center), boarding on HL7534 on April
16, 2016

2.14.2.4 The result of Investigation on R5 Slide

Approximately 5 cm long tear was confirmed at the right lower corner (at the
right corner topside of the side of ground) of R5 slide outer material. As the results
of investigating the details at the Aircraft Manufacuturer factory, R5 slide had no
other anomaly other than that. (See Photo 8. R5 slide.)
Also, according to the latest inspection records (March, 2016) of R5 slide done
by the Company, there was not air leak, malfunction to deploy and others.
Furthermore, at the time to deploy the slide (12:43:22), No.2 Engine was idling and

34
the cut off of No.2 engine was operated at 12:43:45 based on visual materials and
FDR. (See Appended Figure 1. Records of FDR.)
According to the materials concerning the engine exhaust air flow by the
Aircraft manufacturer, B777-300 Engine Exhaust verocity contours at idle thrust
are 55 km/h and reach approximately 40 m from the rear of the Aircraft and narrows
in width. (See Figure 7. Predicted Engine Exhaust Velocity Contours – Idle Thrust.)

Photo 8. R5 Slide

35
 Figure 7. Predicted Engine Exhaust Velocity Contours – Idle Thrust

2.15 Information of Organization and Management

2.15.1 Operation Manual and rules of the Company
2.15.1.1 The Documents required to be onboard of Aircraft
According to the Company, the checklist (QRH) described in 2.1.3(2) was found
in the rack at the rear side of the FO’s seat, later. The Company regulates as follows.
The Company specifies the document required to be on board of aircraft as follows;
(1) Confirmation Obligation by flight crew for documents to be onboard
B777 POM OPERATIONAL POLICY contains the following descriptions regarding
the obligation of flight crew
CREW DUTIES Reference CAP F/O PF PM
COCKPIT PREPARATION
Check A/C documents FOM 6 ● ●

(2) Documents required to be onboard of Aircraft

Regarding the documents required to be onboard of aircraft, FOM 1.3.1 has the
following descriptions (excerpts);
Operational Doc on Board
Legal Documents Documents /Manuals Location Loading Departments

36
 omission
Operations QRH Cockpit Flight Operations
Manual Technical Support
Department
(3) The location of onboard documents are specified as follows (excerpts);
No Location Name of Publication Note
④ Left Hand QRH Stowage QRH, Captain Announcement
Manual
⑤ Right Hand QRH Stowage QRH,POM

2.15.1.2 The Regulation regarding Engine Fire

Regarding the procedure when engine fire occurs, 777 Flight Crew Operations
Manual of the Company regulates as follows;
FIRE ENG L,R
Condition: Fire is detected in the engine
１ A/T ARM switch
(affected side)..............Confirm..................OFF
２ Thrust lever
(affected side)..............Confirm..................Idle
３ Fuel CONTROL switch
(affected side)..............Confirm..................CUTOFF
４ Engine fire switch
(affected side)..............Confirm..................PULL
５ If the FIRE ENG message stays shown:
Engine fire switch

37
 (affected side).........................Rotate to the stop
and hold for 1 second
If after 30 seconds, the FIRE ENG message
stay shown:
Engine fire switch
(affected side)..........................Rotate to the other stop and hold
for 1 second
－－－－－－－－－－－－－－－－－－－－－－－－－－－－－－－－
６- 11 Omission

2.15.1.3 Manuals provided for Emergency Evacuation and others

Concerning procedures and likes for an emergency evacuations of the Company
were provided in FCOM（Flight Crew Operations Manual), FOM（Flight Operations
Manual), 777POM, 777QRH, and COM（Cabin Operations Manual) of the Company
as follows (excerpts);
(1) FCOM (Evacuation Checklist)
Condition: An evacuation is needed
１ Parking brake..............................Set C
２ OUTFLOW VALVE switches
(both).....................................MAN F/O
３ OUTFLOW VALVE MANUAL
Switches (both).....................Hold in OPEN
until the outflow
Valve indications show fully
open to depressurize the airplane F/O
４ FUEL CONTROL switches
(both)....................................CUTOFF C
５ Advise the cabin to evacuate. C
６ Advise the tower. F/O
７ Engine fire switches(both).................PULL F/O
８ APU fire switch................Override and pull F/O
９ If an engine or APU fire warning occurs:
Related fire
switch....................Rotate to the stop
and hold for 1 second F/O
(2) FOM（NON-NORMAL OPERATIONS 8.2.13, Outlined Excerpts）

38
 ■ When an emergency evacuation is expected
PIC Cabin Crew
A:"Attention, crew at station" Standby at their stations preparing for
the next step.
■ When an emergency evacuation is required
PIC First Officer
"Passenger evacuation."
Conduct Passenger Evacuation Procedure in accordance with POM/QRH.
PA: Notify Tower:
"This is the Captain. Evacuate, "KE000, passenger evacuation, request
evacuate." emergency equipment."
Evacuation Command S/W:ON(then Cabin Crew
silence the cockpit warning horn) Initiate passenger evacuation
Note ) If an emergency fire or other conditions make certain exits
unusable, state the direction of egress, and evacuate on the runway, if possible.
The PIC should make a decision on the direction of exits
depending on which engine has the fire, wind direction, attitude and position of
the aircraft and the extent of aircraft damage.

(3) COM（ESCAPE THE AIRCRAFT 4.6 Outlined Excerpts)

4.6.1 EVALUATE
Cabin crew should evaluate the situation to decide evacuation when the
aircraft stops moving. First of all, captain's PA is important to evaluate the
situation. If there's not PA or other actions taken, Cabin Crew should call
for Captain's PA by using Emergency signal.

Evaluate according to PIC's PA after landing or when rejected take-off

（Excerpts）
PIC'S PA EVALUATION CABIN CREW DUTY
"Attention, crew at Emergency Be prepared for
station." Evacuation emergency evacuation at
Anticipated right position.
"This is the Captain" Emergency Evacuation Command the emergency
"Evacuate, evacuate." Required evacuation
(4) COM 4.6.5 B(c)
(c) If an Evacuation is warranted

39
 Command the evacuation as determined after communication with the
captain and other crews.
(5) 4.6.6 EVACUATION INSTRUCT
Release Seatbelts！
Get Out! Leave Everything!
(6) 4.6.8 ACTIVATE EXIT AND EVACUATION-Door Exit
・Quickly confirm armed status of exit.
・Open exit. Utilize all available exits by requesting passenger assistance when
responsible for more than one exit.
・If exit is jammed or slide/raft is not usable, attempt to open it again and
(if necessary),redirect passengers to an alternate exit using appropriate
command.
・Command the first passengers. "Stay At Bottom! Help People Off!"

2.15.2 Emergency Evacuation Training of Flight Crew

According to the records of the Company, the Company implements annual
emergency evacuation training for flight crew and the PIC, the FO, Chief Purser
and all cabin attendants had received this specified training.

2.15.3 How to inform the emergency evacuation of the Company to all

passengers
Regarding how to act at the time of emergency evacuation, the Company
explains to passengers by using demo-video and at the same time, having the safety
leaflet at each seat to ensure all passengers to know that when to evacuate, takeoff
high heels, do not take baggage, the passengers seated at the emergency exit seat
should evacuate first and assist other passengers at the bottom of slide and others.

40
2.16 Additional Information
2.16.1 Fire Extinguishing System of the Aircraft
When a fire broke out from an engine, a warning bell activates and a master
warning light and the red light of the fire handle turn on, “FIRE ENG L(R’) (engine
fire left (right))” in red characters is appeared on the display of EICAS. The Aircraft
has equipped two bottles of fire retardants behind the panels of right side wall in
the front baggage room, which operates independently as a fire extinguishing
system to an engine fire. Pulling a fire handle activates operations to close a main
cock of fuel and bleed air valve, to shut off the operating fluid of the hydraulic
system, to stop power generation of a generator, and to prepare to fight fire by
activating the fire retardant injection circuit. Turning the fire handle to the left or
right discharges the fire extinguisher bottle. If the first fire extinguisher bottle did
not extinguish the fire, turning the handle in the other direction will discharge the
second fire extinguisher bottle. If the first injection could not extinguish the fire,
turning the fire handle to other side enables to inject the second fire retardant. (See
Photo #9. Fire Extinguishing System of B777 Engine.)

Photo 9. Fire Extinguishing System of B777 Engine

41
 3. ANALYSIS

3.1 Qualification of Personnel

Both the PIC and the FO held valid airman competence certificates and valid
aviation medical certificates.

3.2 Aircraft Airworthiness Certificate

The Aircraft had a valid airworthiness certificate and had been maintained
and inspected as prescribed.

3.3 Relations to the Meteorological Conditions

As described in 2.7, when this accident occurred (at around 12:38), wind blew
at about 20 kt from the almost right abeam Runway 34R. Because the fire and
smoke accompanying the fire breakout from the No.1 engine was blown by this beam
wind to the lee side which was the left side of the Aircraft, it is probable that the fire
did not cause that much effect on the Aircraft and the cabin. Furthermore, using the
right side slide of the Aircraft which was at the up wind side for the emergency
evacuation, it is probable that there were not much effects by the fire and smoke.
However, regarding why the R5 slide at its deploying was slipped to under the
Aircraft, it is somewhat likely that as described later in 3.7.5, the strong abeam wind
from the right combined with the thrust still being produced by the #2 engine at the
ground idle setting could have the effects.

3.4 Fracture of the 1st stage HPT Disk

3.4.1 Factor to initiate a step at U-shaped groove
As described in 2.14.1.1, at the teardown investigation of the engine, the 0.010
inches (0.25 mm) high step which was exceeding the allowable limit was confirmed
in U-shaped groove at aft side of the 1st stage HPT disk of the engine. As described
in 2.6.2.3, it was specified to confirm that U-shaped groove was machined to be
within the manufacturing allowable limit as verifying based on the product
inspection by the machine operator and the quality inspection by the inspector at
the time of manufacturing a disk by the engine manufacturer after the
manufacturing. As described in 2.6.2.3,because U-shaped groove is machined from
the outer side and inner side of the groove in order to level a step which is created
on seam from the both sides, a machine operator shall adjust the clearance and
obtain the standard value to set the final machining position as inserting 0.010

42
inches thick shim stock into a clearance between a blade tip of the machining tool
and the bottom of U-shaped groove. This value will be fed into the vertical milling
machine by using a keyboard and processed automatically, but the final machining
position in real is programmed to send the tip of the machining tool to the lower
position by 0.010 inches (0.25 mm) which is corresponding to the thickness of the
shim stock. According to the verification done by the engine manufacturer,
recreating the processing in a final processing position as setting no space between
the tip for a tool and the bottom of U-shaped groove without using the 0.010 inches
thick shim stock, the bottom of U-shaped groove was cut far and resulted in being
the 0.010 inches high step.
In addition, as described in 2.6.2.3, when repairing after the manufacturing of
the disk, U-shaped groove was not machined.
Based on these, regarding the step (hereinafter referred to as “the step in U-
shaped groove”) of exceeding the allowable value in U-shaped groove at the aft side
of the 1st stage HPT disk was occurred, it is somewhat likely that the shim stock was
not used properly at the process of manufacturing the disk for the machine operator
to input a reference value to a vertical automatic lathe as a final machining position,
or it is somewhat likely that a miss-input was made at the time to input a reference
value to a vertical milling machine. In order to adjust the clearance, the work is
carried out using shim stock as checking the condition by sight and fingertip touches
and even an experienced machine operator has to use the shim stock procedurally,
otherwise because there is a possibility to mistakenly estimate a reference value or
to input erroneously a reference value, it is necessary for the engine manufacturer
to revise the inspection methods and machining method of U-shaped groove from
the view to prevent an occurrence of human errors.

3.4.2 Causes to initiate Cracks

As described in 2.14.1.1, at the engine teardown investigation, the step of
exceeding allowable limit in U-shaped groove at the aft side of the 1st stage HPT
disk was confirmed and investigating in detail around the step, cracks were revealed
along the step in U-shaped groove near the parts of the disk fracture. Furthermore,
as described in 2.6.2.3, because the repetitive stress is acting on the outer peripheral
portion of the disk per a flight, therefore, if step and likes could be there, stress could
be concentrated. Based on these, it is probable that the cracks were generated from
these parts as repetitive stress acting on the step of exceeding the allowable limit in
U-shaped groove at the aft side of the 1st stage HPT disk per a flight.

43
3.4.3 Causes to fracture
As described in 2.6.2.2, it is probable that the crack generated in U-shaped
groove was propagating by the action of repetitive stress per a flight. As described
in 2.6.2.7 and 2.6.2.8, the Company executed the FPI on the 1st stage HPT disk when
disassembling the HPT module, but the crack was not found. As described in
2.14.1.1, as the result of analyzing the fractured surface, it is somewhat likely that
the cracks in U-shaped groove of the 1st stage HPT disk may exist prior to the last
inspection conducted at the engine repair shop on November 12, 2014. Based on
these, it is probable that because the cracks were not be discovered at FPI conducted
the previous inspection and due to the flight following the inspection, the cracks
were propagating more to be fractured.

3.4.4 Causes to fail to find a step in U-shaped grooves

As described in 3.4.1, the step of exceeding the allowable limit for U-shaped
groove at the aft side of the 1st stage HPT disk was occurred when the engine
manufacturer was manufacturing the disk. However, as described in 2.6.2.4, it is
highly probable that the step was not discovered at the product inspection which
was done by the machine operator to confirm the HPT disk to be within a
manufacturing standard value, and the quality inspection which done by the
inspector to confirm the conformity to the design standard, and was shipped out.
As described in 2.14.1.1, 0.010 inches (0.25 mm) deep step was confirmed along
the whole periphery of U-shaped groove of the aft side of the 1st stage HPT groove
at the overhaul shop of the Company. It was somewhat likely that this confirmed
step could be detected by naked eyes, because the confirmed step was five times of
0.002 inches (0.05 mm) which was the allowable value for the machined step at the
time of manufacturing. As described in 2.6.2.3, it is highly probable that the machine
operator and the inspector were conducting the tests following the test method
specified in the work instruction and the manufacturing drawing like a visual
observation, a palpation, and measurement by the measuring instrument, CMM
(Coordinate Measuring machine) and others, depending on the parts to be inspected.
However, the allowable value which determines a result of inspection is specified in
the manufacturing drawing, but the parts of U-shaped groove is not specified as the
critical location to be inspected like blades mounting slots and grooves spline
machined for hub and shaft, therefore, 0.002 inches (0.05 mm) was applied for the
allowable value for the whole of the 1st stage HPT disk as a standard value.

44
Furthermore, regarding the parts specified as the critical points to be inspected,
notes regarding the detailed inspection and entries of the measurements were
required for the work instruction, but for the other parts, only the results of
inspection were recorded in the work instruction. Based on these, it is somewhat
likely that the machine operator and the inspector did not pay sufficient attentions
to U-shaped groove which is not specified as the critical inspection location and have
failed to detect a step. As described in 2.14.1.3, according to the investigation results
of the engine manufacturer after the accident, the other same type engine did not
have the step exceeding the allowable limits. It is probable that because the engine
manufacturer did not estimate a high potential to have a malfunction occurred at
U-shaped groove, it was not specified as critical inspection parts. On the other hand,
because the HPT disk is the part to receive repetitive stress, repeatedly, if there are
slight scratches or step of exceeding manufacturing allowable limit, these could be
a point to initiate the stress concentration or fatigue crack due to notch effect. It is
necessary for the engine manufacturer to call attentions of machine operators and
inspectors as a critical inspection location at the inspection process regarding the
parts where step like the one in U-shaped groove could be caused by manufacturing.

3.4.5 Causes to fail to discover cracks propagating from U-shaped

grooves
As described in 3.4.3, it is somewhat likely that the cracks propagated from the
step of exceeding the allowable limit of U-shaped groove was existing prior to the
delivery to the engine repair shop of the Company on November 12, 2014, however,
the latest FPI described in ⑧ of 2.6.2.7 did not reveal the propagated
cracks.Furthermore, as described in 2.6.2.8 and 2.6.2.9, it could not be confirmed
that there was a fact that could be a cause of this accident from the inspection
system, inspection records, and others at the repair shop of the Company.
As described in 2.6.2.8, according to the Operator and the Inspector in charge
of applying the FPI on the 1st stage HPT disk of the engine, the inspection was
conducted as following the work procedure instructed by the engine manufacturer.
First, after visually assessing the whole, then start to inspect the front side of disk
as the locations of being highly importance. In addition, the Inspector and the
Operator have never seen the crack at the U-shaped groove. As described in 2.14.1.3,
according to the investigation results by the engine manufacturer, there was no
preceding case U-shaped groove step of exceeding the allowable limit and this one
was the first case. Based on these, regarding the reason that the cracks propagating

45
from U-shaped groove could not be discovered, as described in 3.4.4, there was no
case that the area of U-shaped groove had error till now, furthermore, because U-
shaped groove is not point to be inspected with attention according to the inspection
manual specified by the engine manufacturer, the Inspector and the Operator
inspected with special emphasis on the critical inspection parts like mounting slots
for turbine blades, on the other hand, it is somewhat likely that the cracks in U-
shaped groove was failed to be detected. In addition, as shown in Photo 3 of 2.14.1.1,
it is possible that the fact attributed because cracks was existing along the
machining trace in U-shaped groove as dotting, it was difficult to find.

3.5 Damage of the No.1 Engine

3.5.1 Damage of the HPT Case
When watching the damaged condition of the No.1 engine, as described in
Attachment 3 (2) to (4), there were no large scale damage on the No.1 engine inlet,
fan blades, LPC and HPC, and traces of foreign matters were not confirmed.
Furthermore, as described in Attachment 3 (6), the HPT case wall between 5:30 and
7 o’clock was bent radially outward and twisted which was approximately 36 cm
(approximately 14 inches) long and approximately 11.4 cm (4.5 inches) wide along
the periphery. In addition, the 1st stage HPT blades were fractured across the airfoil
adjacent to the root platform and there were sectors of the 1st and 2nd stage HPT
stator vanes at the bottom of the engine that were missing. After the accident, as
described in 2.10, at the runway and the grassland in vicinity from the point of
approximately 680 m from the inside of the runway threshold as the starting point
to the north side of approximately 570 m, stator vanes including the metallic debris
shown in Attachment 3 were found, as described in Attachment 3 (6) ⑥ and
2.14.1.1, according to the teardown investigation of the No.1 engine, metallic debris
was a part of the fractured the 1st stage HPT disk rim and the fractured parts of the
1st stage HPT disk and the fractured face of the recovered rim were matched. Based
on these, it is highly probable that the damaged parts of the HPT case of the No.1
engine was caused by the penetration due to centrifugal force the fractured rim part
which were fractured and ejected out of the 1st stage HPT disk to almost 8 o’clock
direction through the HPT case. At that time, it is highly probable that the 1st stage
HPT blades and the 2nd stage HPT stator vanes were fractured and ejected due to
centrifugal force through the opening of the HPT case to outside.

46
3.5.2 Damage to Fuel Oil Heat Exchanger
Regarding why the Fuel Oil Heat Exchanger had cracks generated, as
described in Attachment 3 (10), but there were no damage impacted from outside,
and as described in Attachment 3 (5), the inner diffuser case had cracks and missing
pieces and as described in Attachment 3 (8), the tail cone mounting bolts were
missing.
Therefore it is probable that those cracks were developed, when the 1st stage
HPT disk rim had been fractured and released, and struck into the turbine case, a
strong shock force was generated. In addition, it is also probable that engine run
down loads, which were generated when the engine No.1 stopped suddenly following
the disk rupture, was the contributing factor.

3.5.3 Damage to other parts

(1) Left-side Translating Cowl of the No.1 engine
As described in Attachment 3 (11), because missing pieces and burnt damage
at the left side translating cowl of the No.1 Engine was confirmed and the 2nd stage
HPT blades stuck in the inner wall of the dropped translating cowl, regarding the
fracture of the left translating cowl, it is highly probable that as described in 3.5.1,
it was generated that the 1st stage HPT disk rim parts was fractured and the
fractured piece penetrated the HPT case to hit on the translating cowl.
(2) LPT and the Tail Cone
As described in Attachment 3 (7) and (8), it is somewhat likely that regarding
why there were no apparent damage on LPT and tail cone, the fractured rim parts
within the HPT, the HPT blades and others were ejected through the fractured
opening to outside of the Aircraft.
(3) Outboard Flap
Regarding the cracks of the outboard flap, it is probable that the fragment
flying due to the fracture occurred at the 1st stage HPT impacted on the flap and
caused the cracks.

3.6 Engine Fire

3.6.1 Progress of the Fire Breakout from No.1 Engine
Regarding the progress of fire breakout from the No.1 engine, it is highly
probable that as described in 3.5.1, the 1st stage HPT disk rim was fractured during
the takeoff roll of the Aircraft and as described in 3.5.2, due to the impact forces and
engine run down loads generated by the release of the fragment from the 1st stage

47
HPT disk through the engine case and the leaked fuel and engine oil through this
cracks contacted the high temperature engine cases of the No.1 engine to be ignited.

3.6.2 Extinguishing Engine Fire

As described in 2.1.3, as confirming the fire breakout from the No.1 engine by
the engine fire warning sound and warning message (hereinafter referred to as “the
fire warning messages”), the PIC, following the procedure provided in FCOM of the
Company described in 2.15.1.2, activated the first fire extinguisher bottle equipped
to the Aircraft described in 2.16.1. Then the fire warning messages went out once,
but five to ten seconds later, the fire warning messages reappeared. Therefore,
attempting to extinguish the fire by using the second fire extinguisher bottle, the
fire warning message went out again, and five to ten seconds later, again the fire
warning message reappeared. Concerning these, as described in Attachment 3 (11)
③, it is probable that because an opening was created due to the breakage of the left
translating cowl inner wall, the fire retardant agent were leaked from there and
could not exhibit full effects of fire retardant.

3.7 Emergency Evacuation

3.7.1 Decision of the PIC
As described in 2.1.3 (1), it is highly probable that even though the attempts
to extinguish the fire of the No.1 engine with uses of two fire extinguisher bottles
were made, because the third fire warning message was displayed, the PIC decided
to conduct the emergency evacuation. Furthermore, as described in 2.1.3 (6), at the
same time as the message was disappeared by using the second fire extinguisher
bottle, the first fire engine started to fight fire from aft of the Aircraft, but as
described in 2.1.3 (1), when third fire warning message was displayed, since the PIC
considered that there would take more time to commence the firefighting because
he saw the fire engine of coming from the front, it is probable that he had decided
that it is necessary for the emergency evacuation to be conducted in a hurry.
FOM(8.2.13) of the Company described in 2.15.1.3(2) contains descriptions like
“the Pic should make a decision on the direction of exits depending on which engine
has the fire, wind direction, attitude and position of the aircraft and the extent of
aircraft damage” and “if an emergency fire or other conditions make certain exits
unusable, state the direction of egress, and evacuate on the runway, if possible”.
It is probable that the PIC decided to evacuate to the right side because the
strong wind was blowing from the right abeam because of the No.1 engine fire.

48
3.7.2 Actions Taken by Flight Crew
(1) Instruction of Emergency Evacuation
As described in 2.1.3 (1), the PIC called for the emergency evacuation checklist
in QRH to the FO to hurry to perform, but because the QRH was not at the specified
position, the FO could not perform the checklist right away. While the FO was
looking for QRH, it is highly probable that the PIC who thought that the emergency
evacuation should be done in a hurry, implemented the procedure to stop the engine
based on his memory. According to the FCOM (Emergency Evacuation Checklist)
(B777QRH) of the Company described in 2.15.1.3 (1), after the both of fuel control
switches were cut off and the engine was stopped, an announcement of emergency
evacuation to the cabin should be made. As described in 2.1.3 (1), after cutting off
the engine, the PIC stated to announce the emergency evacuation via PA.
However, according to the analysis of FDR, CVR, QAR records and Video, the
time for the PIC to activate the emergency evacuation signal was at 12:42:51, the
time to announce to conduct emergency evacuation from the right side emergency
doors was 12:43:03, and the time for the fuel control switch of the No.2 engine to be
cut off was 12:43:45. And the first door being opened was L1 door and the time was
12:43:17. Based on these, instruction of emergency evacuation prior to halting of the
No.2 engine positioned in the evacuating direction was given, it is highly probable
that the No.2 engine was stopped about 28 seconds later after the first door was
opened. As described in 2.14.2.4, the wind velocity of the engine wake air flow is 55
km/h even at the time of idle thrust and when conducting the emergency evacuation
prior to the engine stopped, there are potential threat that the passengers could be
blown away by the engine wake air flow and others. It is necessary for the Company
to revise the education and training in order to enforce the thorough compliance to
emergency evacuation procedure.
(2) Regarding the misplacement of QRH
As described in 2.15.1.1, the Company provides manual (B777POM
OPERSTIONSL POLICY) regarding the responsibilities of pre-flight check, the
section in charge of boarding, the specified placement and the document to be on
board.According to the manual, regarding QRH checklist in paper to be used at the
time of emergency evacuation, the specified location is at the right of the FO and the
left of the PIC, the Flight Technical Support section of the Company should put there
and the PIC and the FO shall perform the checks on the loaded status respectively,
prior to the departure.

49
 As described in 2.1.3 (2), this QRH was the one that the FO took time to find
because when the PIC called for this checklist of the FO, it was not at the specified
location, as the result, the FO used the checklist in tablet. The QRH was found in a
rack at rear of the FO seat, later date, as described in 2.15.1.1. Based on these, it is
somewhat likely that the QRH was not placed at the specified location and the PIC
and the FO did not check sufficiently or did not check the documents that should be
loaded on an aircraft for sure prior to the departure. At the time of an emergency
situation to compete for a moment, if the checklist could not be found, because there
is possibilities that the action could be delayed and a recovery could not obtained,
the Company should promote thorough inspection of the documents to be carried on
an aircraft prior to a departure and it is necessary to publicize again the use of QRH
on an emergency situation for sure.

3.7.3 Actions Taken by CAs

Chief Purser opened the L1 door and deployed the slide at the emergency
evacuation, however, as described in 2.1.3 (3), because the Chief Purser was
confident to conduct an emergency evacuation due to the prior notice given by the
cabin attendant in charge of L3 door and decided by checking outside through L1
door in early stage to confirm that there were no obstacles to escape, it is somewhat
likely that when receiving the emergency evacuation signal from the PIC, she
opened the door in a flex. And instantly the Chief Purser was realizing that L1 slide
could not be used and guided passengers to other doors. As described in 2.15.1.3 (5),
4.6.6 EVACUATION INSTRUCT in COM of the Company specifies to instruct “Get
out! Leave everything” and as described in 2.1.3 (3), the chief Purser was making
announcement to evacuate from the right side without baggage and cabin
attendants shout at passengers not to take baggage and take heeled shoes off, but
as described in 2.1.3 (6), many of passengers took the baggage and like to evacuate.

3.7.4 Publicification of Emergency Evacuation to Passengers

The Company is promoting the attention to how to cope with an emergency
situation by demonstration video or an explanation by cabin attendants prior to a
departure. However, when evacuating for a real emergency, as considering the case
that passengers could be agitated and might not follow the instruction given by cabin
attendants, it is desirable that air carriers and Civil Aviation Bureau should plan to
promote the wide general public including passengers to have full knowledge
regarding the safety information at emergency evacuation with the backup reasons

50
like possible threat of heeled shoes and baggage to damage a slide to be unusable
and others, and study how to promote more solid understanding and recognition.
Furthermore, in order to make prompt and safe emergency evacuation,
remarks or safety information like leaving baggage as air carriers publicizing, how
to use evacuation slide properly and leaving an aircraft side as soon as possible after
the evacuation, should be checked in full and it is desirable even for passengers to
act as understanding the criticality to follow instruction given by flight crew and
cabin attendants when conducting and emergency evacuation in order to safe life of
oneself and others.

3.7.5 Deployment of Slides

As described in 2.14.2.2, according to the video recording, the slides R3, R4 and
R5 were blown toward the rear of the Aircraft right after the deployment, but it is
probable that R3 slide fully deployed in about 6 seconds on the runway and R4 slide
fully deployed in about 38 seconds on the runway. As described in 2.14.2.4, it is
probable that the R5 slide was properly maintained and the gas filling status for
deploying the slide was normal. Furthermore, it is probable that the tear at the top
right corner of the ground end was caused because of the friction with the runway
surface when it was fanned by the cross wind and the thrust from the #2 engine at
its ground idle setting.
As described in 3.7.2, when deploying R5 slide, the No.2 engine was at idle
thrust. As described in 2.14.2.4, according to the materials (Figure 7 Predicted Jet
Engine Exhaust Velocity Contours – idle thrust) of the manufacturer of the Aircraft,
the predicted jet engine exhaust velocity contours at idle thrust is 7 m wide at the
rear end of the Aircraft, extends to the approximately 40 m ahead while being
tapered at the level of approximate 6 m high and the wind velocity is about 30 kt.
Calculating the resultant wind from the wind of the jet engine exhaust blow at
setting of the heading at 335º as described in 2.10 with the wind direction 060ºand
velocity about 20 kt at the time of the accident described in 2.7, was resulting in the
wind direction 007ºand the velocity about 37.5 kt. The velocity is 1.5 times of the
allowable limit which is 25 kt for the velocity to make a slide have normal
deployment as described in 2.14.2.1. Because of this resultant wind’s effect, it is
somewhat likely that R5 slide which is at most rear end and the spacing to the
ground is the longest, slipped under the rear end of the Aircraft by folded, deployed
with its top caught by the runway therefore could not return to a normal standing
position spontaneously as the effect of wind gone.

51
 4. CONCLUSIONS

4.1 Summary of Analysis

(1) The PIC and the FO held both valid airman competence certificates and valid
aviation medical certificates. Besides, the Aircraft had valid airworthiness
certificate and had been maintained and inspected as prescribed. (3.1, 3.2)16
(2) Cause to generate step in U-shaped groove
It is somewhat likely that the step of exceeding allowable limit at the aft side of
the 1st stage HPT disk could be generated due to the incorrect use of shim stock by
the machine operator at the process to input the a reference value for final
machining position to vertical automatic lathe when machining the disk or because
errors could be input when inputting the a reference value into the vertical
automatic lathe. For the engine manufacturer, from the views of preventing the
human error, it is necessary to study the revision of the inspection method and the
machining method of the U-shaped grove. (3.4.1)
(3) Cause of the crack generation
It is probable that a mis-match exceeding the allowable limit in the U-shaped
groove at the aft side of the 1st stage HPT disk had received repetitive stresses at
every cycle and the crack was generated originating from the step. (3.4.2)
(4) Cause of the 1st stage HPT disk fracture
It is probable that because the generation of cracks as originating from the step
of exceeding the allowable limit in U-shaped grooves at the aft side of the 1st stage
HPT disk were propagated by the repetitive force per a cycle, however, the FPI
conducted during the process could not reveal the cracks and the cracks were
propagated furthermore to be resulted in the fracture. (3.4.3)
(5) Cause to miss the step in U-shaped groove
It is somewhat likely that the machine operator and inspector did not pay full
attention to U-shaped groove part which was not specified as a critical inspection
point and had failed to detect the step. Since the heavy stress could repeatedly act
on the HPT disk, slight scar or step of exceeding the allowable limit could be
originating point of fatigue crack and stress concentrating point due to the notch
effects. Therefore, it is necessary for the engine manufacturer to call the attentions

16 The number listed in each sentence end of this clause, indicates a main clause number of “#3, Analysis”

concerning the description of the each sentence.

52
 of the machine operator or inspector to the parts where a step could be generated
like U-shaped groove when machining, by specifying the parts as a critical
inspection location. (3.4.4)
(6) Cause to miss the crack propagating from U-shaped groove
It is somewhat likely that the crack propagating from U-shaped groove could be
existing prior to the delivery to the engine factory of the Company, but the crack
could not be discovered at the latest FPI. Regarding this, an operator and an
inspector inspected with an emphasis on the critical points like turbine blade
mounting parts, on other hand, it is somewhat likely that the crack for U-shaped
groove failed to be detected. Adding more, it is somewhat likely that the cracks
which were dotted along the machining trace in U-shaped groove was difficult to
find because of assimilating into the machining trace. (3.4.5)
(7) Damage of the HPT case
It is highly probable that the damage of HPT case at the No.1 engine was due
to penetrating of the rim debris of the 1st stage HPT disk to the 8 o’clock direction
through the HPT case because of a centrifugal force. (3.5.1)
(8) Damage of Fuel Oil Heat Exchanger
Regarding why the Fuel Oil Heat Exchanger got cracked, it is probable that
when the fractured rim of the 1st stage HPT disk had been fractured and released,
and struck into the turbine case, a strong shock force was generated, in addition,
engine run down loads were generated when the No.1 engine stopped suddenly
following the disk rupture. (3.5.2)
(9) Damage and others on other parts
Regarding the breakage of the left translating cowl, it is somewhat likely that
it was created by the hitting of fractured rim debris of the 1st stage HPT disk.
Regarding why LPT and the tail cone did not have almost no damage, the
fractured piece of the rim of the HPT was ejected through the opening caused by
this ejection.
Regarding the cracks at the outboard flap, it is probable that it was caused by
the hit by the ejected debris due to the fracture of the 1st stage HPT disk rim. (3.5.3)
(10) Process of the fire breakout from the engine
Regarding the progress of fire breakout from the No.1 engine, it is highly
probable that due to the impact forces generated by the release of the fragment
from the ruptured rim part of the 1st stage HPT disk through the engine case and
the engine rundown loads generated when the engine stopped suddenly, the cracks
were developed in the Fuel Oil Heat Exchanger and the fuel and engine oil leaking

53
 through these cracks contacted the hot area of engine cases of the No.1 engine to
be ignited.(3.6.1)
(11) Extinguishing the engine fire
Regarding that the PIC attempted to extinguish the fire using two fire
extinguisher bottles equipped on the Aircraft and once the fire warning message
was gone, but it reappeared, again, it is probable that because the opening was
created due to the damage to the inner wall of the left translating cowl, the fire
extinguisher could not fully be effective. (3.6.2)
(12) Decision taken by the PIC
Even the PIC attempted to extinguish the fire using two fire extinguisher
bottles equipped on the Aircraft, because the third fire warning message
reappeared, it is highly probable that the PIC decided to conduct emergency
evacuation and it is probable that because of the fire at the No.1 engine and strong
abeam wind from the right, he decided to evacuate to the right side. (3.7.1)
(13) Response (action) taken by the flight crews PIC’s
It is highly probable that the emergency evacuation was instructed prior to the
No.2 engine stop in the direction of evacuation and it took about 28 seconds from the
first door open and the No.2 engine stop. Regarding the instruction of emergency
evacuation was given before the No.2 engine stop, it is somewhat likely that the PIC
decided the emergency evacuation and then called for the emergency evacuation
checklist of the FO and turned the emergency evacuation switch on at the same
time, however, the FO could not find out the emergency evacuation checklist of QRH
(paper), and he took some time to read out the emergency evacuation checklist in a
tablet.
Regarding the reason that the FO could not find QRH, it is somewhat likely that
the QRH was not placed at the specified location and the PIC and the FO checked
insufficiently or did not check the documents that should be loaded on the aircraft
for sure prior to the departure. The Company promotes thorough inspection of the
documents to be carried on an aircraft prior to a departure and it is necessary to
publicize again the use of QRH on an emergency evacuation. (3.7.2)
(14) Response (action) taken by cabin attendants
Regarding the fact that the chief purser opened the L1 door and let the slide
deploy, as she judged there would be no problem to evacuate by looking outside
through a window of L1 door in advance, it is somewhat likely that when she
received the signal of emergency evacuation from the PIC, automatically she
opened the L1 door. Instantly, as the chief purser recognized that it is not possible

54
 to evacuate through L1 door, she guided passengers to other door. The chief purser
and cabin attendants announced that do not carry baggage to evacuate, but many
of passengers had carried their baggage to evacuate. (3.7.3)
(15) Disseminating the knowledge regarding the emergency evacuation to the
passenger;
Cabin attendants had instructed passengers to evacuate without baggage
through the right side slide, but it is probable that many of passengers did not
follow the instruction not to carry the baggage. It is desirable that air carriers and
Civil Aviation Bureau should plan to promote the wide general public including
passengers to have full knowledge regarding the safety information at an
emergency evacuation with backup reasons like possible threat for heeled shoe and
baggage to damage slide and cause the slide unusable and others, and study how
to promote more solid understanding and recognition.(3.7.4)
(16) Deploying Slide
R5 slide slipped, bent in the rear under the fuselage and fully deployed with
the tip caught on the runway due to the effect of the resultant wind in wind
direction 007ºand 37.5 kt of wind velocity and 20 kt of engine exhaust flow, it is
somewhat likely that even after the engine exhaust flow of the No.2 engine were
stopped, it was unable to return to a normal standing position.(3.7.5)

4.2 Probable Causes

It is highly probable that the causes of this accident were the fracture of the
high pressure turbine (HPT) disk of the No.1 (left-side) engine during the takeoff
ground roll, the penetration of the fragment through the engine case and the
occurrence of subsequent fires.
Regarding the cause for the 1st stage HPT disk to be fractured, it is probable
that a step was machined exceeding the allowable limit when machining U-shaped
groove on the aft side of the 1st stage HPT disk to manufacture the engine and from
this step the low-cycle fatigue crack was initiated and propagated during running of
engine.
Regarding why the step could not be found, it is somewhat likely that defects
failed to be detected at the time of the inspection by the manufacturer during the
production process. And as for the cracks that were not found, it is somewhat likely
that those cracks failed to be detected at non-destructive inspection on the disk by
the Company at the time of maintenance of the engine in use.
Regarding the fire breakout from the No.1 engine, it is probable that due to the

55
impact forces generated by the release of the fragment from the ruptured rim part
of the 1st stage HPT disk through the engine case and the engine rundown loads
generated when the engine stopped suddenly, the cracks were developed in the outer
case of the Fuel Oil Heat Exchanger and the fuel and engine oil leaking through
these cracks contacted the hot area of engine cases of the No.1 engine to be ignited.

5. SAFETY ACTIONS

5.1 Actions Taken by FAA

On March 9, 2017, FAA (Federal Aviation Administration of the United States
of America) issued Airworthiness Directive (AD) concerning a turbofan engine
manufactured by Pratt & Whitney. The contents, based on this accident, concerning
PW4074, PW4074D, PW4077, PW4077D, PW4084D, PW4090 and PW4090-3
turbofan engine manufactured by Pratt & Whitney, requested an implementation of
an inspection to evaluate the condition of an U-shaped grooves (web, rim, fillet) at
back of the 1st stage HPT hub and not using the hub where defects were found at
the inspection. The Airworthiness Directive means to improve the unsafe condition
of these products and was effective on April 13, 2017.

5.2 Actions Taken by Engine Manufacturer

(1) Provisions of information and issuances of technical information for the engine
users;
① On June 17, 2016, the engine manufacturer issued All Operator Wire Net Case
5 No.CAS-30064-Q7K9L2 to instruct users of the PW4000 series engines to apply
FPI on all of the 1st stage HPT disk had been removed from the HPT module.
② On June 24, 2016, the engine manufacturer had issued PW4000-112A11
Operator Communication to instruct that all of the 1st stage HPT disk and the 1st
stage rotor assemblies currently removed from HPT module of PW4000-112 engine
to be disassembled and to inspect the aft web area for any indication to be a step at
the area (web, rim and fillet) including U-shaped groove, sufficiently at the
moment.
③ On September 23, 2016, the manufacturer issued P&W SB PW4G-112-72-342
(category 5 and 6) to instruct to apply the surface inspection and measure by
creating replicas of the area including U-shaped groove at the aft side of the 1st HPT
disk.

56
(2) Change of the 1st stage HPT disk manufacturing process;
The engine manufacturer changed the final finishing process of U-shaped
groove at the 1st stage HPT disk to have the process by a machine only by canceling
the process to have manual operation by a machine operator, in order to have no
processing error.
(3) Change of the product inspecting process
The engine manufacturer added the inspecting process to records step of U-
shaped groove at the inspection of the 1st stage HPT after manufacturing and add
U-shaped groove inspection (creating replica of step at U-shaped groove and
confirmation based on the use of Shadowgraph Machine) at the outside of the aft
side of the disk based on the Service Bulletin described in 2.14.1.2 to the inspection
of U-shaped groove.

5.3 Actions Taken by the Company

(1) The Company is carrying out the inspection on the 1st stage HPT disk based on
the Service Bulletin issued by the engine manufacturer shown in 5.2.(1), for the
same type engines used by the Company.
(2) The Company noticed all flight crew regarding the pre-flight check to check the
on-boarding condition of the documents which should be on board on March 17, 2017.
(3) The Company reflected the training procedure for flight crew to carry out
evacuation after the engine stops.
(4) Prohibition for passengers to carry baggage at the time of emergency evacuation
The Company added the illustration to call attention not to carry the baggage at
the time of emergency evacuation into the demonstration video for aviation safety
to show at all aircraft belonging to the Company.
(5) The Company revised the Evacuation Part of the Cabin Operation Manual to
add the item for the cabin attendants to have a control of passengers, which was
applied since October 17, 2016. The main contents are as follows;
① Describe the instruction to leave the baggage to evacuate and the request to
assist the following passenger to the passenger sitting at emergency seat, in Korean
and English as before, in Japanese and Chinese to be added.
② At the time of unexpected emergency landing, it is critical to grasp the situation,
therefore inform the information of the situation in Korean and English to
passengers. Considering the nationalities of passengers, if possible, the cabin
attendant could use other languages to instruct additional information.
③ When the flow of evacuation halts or the speed decreases, check the cabin for any

57
passenger remained. When the assigned area is empty, the cabin crew can pre
evacuation than some other passengers and control the passengers outside the
aircraft.

58
Appended Figure 1.: FDR Record

59
Appended Figure 2.: Estimated Route of Take-off Roll

60
Appended Figure 3.: Three-view drawing of Boeing 777-300

61
Attachment 1.: Records of CVR, FDR, and Video

JST Voice Content

(omitted)
12:36:53 TWR Korean Air 2708, wind 070 at 21, runway34R, cleared for
take-off.
F/O Cleared for take-off runway34R, Korean Air 2708.
CAPT Cleared for take-off. You have control, I have ATC. Take-
off time set.
37:17 F/O TO/GA
37:18 CAPT Thrust reference -take-off thrust set. I have thrust.
F/O You have thrust.
37:30 CAPT 80kt hold.
37:35 “ Bang” (Engine Explosive Sound)
37:36 “ENG FAIL” (Voice Alert)
37:37 TWR Korean Air 2708, your No1 engine fired. Stop immediately,
Stop immediately.
37:38 1st Fire Bell Sound
37:39 F/O Speed Brakes - UP, No.2 reverse
37:45 F/O Korean Air 2708, reject take-off on runway34R.
TWR Understood. We saw fire from your No1 engine.
37:51 VIDEO/ GS=0
FDR
CAPT Engine fire.
F/O Fire engine left.
Parking break - set
37:57 CAPT Crew at the station, crew at the station.

38:00 CAPT Fire engine left. Memory Items. (Order)

~38:08
CAPT Fire engine left. Memory Items.
(F/O) A/T ARM S/W L/H - off (Confirm L), L/H Thrust lever - idle
(Confirm L), Fuel control S/W L/H - cut off (Confirm)
CAPT Cut off
38:10 FDR FUEL CUT OFF, FF.VALVE CLOSE, N1=0, N2=0
38:11 CBN Ladies and gentlemen, we are unable to take-off. Further

62
~38:54 information will be informed shortly. (in Korean, English
) We are waiting for the take-off clearance. Please wait at
your seat. (Japanese)
38:18 F/O Engine fire switch - pull, still fire, rotate
38:20 FDR ENGINE FIRE BOTTLE No.1 - OPEN
38:25 TWR Korean Air 2708 fire trucks are going to you.
~38:29
F/O I'm sorry. Say again.
TWR Fire vehicle going to around you.
F/O Thank you.
38:31 CAPT Fire is gone.
F/O Yes, fire is out.
CAPT Fire is gone?
F/O Yes sir, fire is gone.
CAPT Contact again and inform them fire is gone.
38:51 F/O Korean Air 2708, Fire is gone.
L3 CA IPN Call to CAPT
TWR Thank you, stand by, hold position.
F/O Holding present position, Korean Air 2708.
39:00 CAPT Hello … Hello.
L3 CA CAPT. There is smoke from engine on L3 side.
CAPT Extinguished fire. Fire truck has just reached and in
preparation. Please wait.
L3 CA Yes, sir.
39:12 CAPT It seems fire is gone.
39:57 PURS It seems fire was occurred at the back. Now, fire truck
arrived and will extinguish fire.
CAPT Hello … Hello.
PURS Yes, captain.
CAPT We have left engine fire. We extinguished, and fire was
gone. And fire truck arrived and extinguished fire. No
evacuation needed. Please wait. If it possible, please
inform them what I told you.
PURS Fire was extinguished, now?
CAPT Yes, fire has gone.

63
 PURS If so, we will return to gate?
CAPT Yes.
PURS Yes, sir.
CAPT For now, situation what I said to you...
PURS Do I need to tell all?
CAPT Just… Ahh… Evacuation checklist
40:40 FDR ENGINE FIRE ALARM (2nd time)
2nd Fire Bell Sound
40:41 F/O Fire engine comes again…
40:44 PURS Do I need to tell technical problem?
CAPT Hold on please.
40:52 CAPT Contact Tower again.
F/O Korean Air 2708, We got a Fire engine L/H message again.
TWR Roger.
CAPT Fire.
40:59 FDR ENGINE FIRE BOTTLE No.2 - OPEN
41:01 TWR Korean Air 2708, right now fire engine reaching your No.1
engine.
F/O Roger thank you, Korean Air 2708.
F/O Fire truck arrived.
41:11 CAPT Bottle discharged again…
F/O Fire is gone again.
CAPT OK.
CAPT Left side ----
41:38 F/O 2 fire trucks are coming forward.
42:06 PA Ladies and gentlemen,
42:07 3rd Fire Bell
42:08 CBN We just rejected take-off for technical problem. Further
~42:38 information will be informed shortly. (in Korean, English
and Japanese)
42:09 F/O Fire engine left message.
CAPT Need to Evacuation
42:13 F/O Tower Korean Air 2708, we got a fire engine L/H message
again
TWR Roger.

64
 42:22 CAPT Tower Korean 2708, we still got the message. We need
evacuation to the right hand side.
TWR Understand.
42:30 CAPT Evacuation.
F/O Roger, Korean 2708.
42:37 CAPT Evacuation Checklist. (Order)
42:51 CAPT Evacuation signal sound
42:52 F/O Evacuation Checklist. (reply)
43:03 CAPT EVAC, EVAC, EVAC to the right hand side.
43:14 CAPT Evacuation Checklist. (Order)
43:17 F/O Stand by.
43:17 VIDEO L1, R1 DOOR OPEN
43:22 VIDEO R5 DOOR OPEN
43:22 CAPT You can find in this…
43:26 F/O Evacuation Checklist. (reply)
43:27 F/O Parking Brake - set
Outflow Valve S/Ws (both) - manual
43:27 VIDEO R4 SLIDE deployment bounce to the fuselage
43:29 VIDEO R2 DOOR OPEN
43:36 F/O Outflow Valve Manual S/Ws (both) - hold and open
position
43:41 CAPT Fully open completed
43:43 F/O Fuel Control S/Ws (both) - cut off
43:45 CAPT Both Cut off
43:45 QAR FUEL CUT OFF RH（5m35s from the LH engine cut off)
F/O Advise the CBN to evacuate
F/O Advise the Tower
43:48 FDR FDR recording terminated. (03:43:52)
43:50 CAPT TWR, Korean Air 2708, mayday, evacuate, evacuate on the
34R.
TWR Understood.
43:50 VIDEO Passengers commence to evacuate via R3 door
44:04 VIDEO R4 slide fully deployed.
44:06 VIDEO Passengers commence to evacuate via R4 door
Time (JST) has been proved by a time signal recorded in ATC communication.

65
Legend; TWR Tokyo Tower 124.35 MHz
CPT; Cockpit CPT; Captain
F/O; First Officer PURS; Purser
CA; Cabin Attendant CBN; Cabin
IPN; Intercom PA; Passenger Address System

Time Correction for FDR, CVR and QAR were set by making the ATC
communication recorded in CVR and VHF radio transmission signal recorded in
QAR and FDR with a time signal recorded in ATC communication.

66
Attachment 2.: FPI (SPOP-84) PROCESSING of KAL

67
Attachment 3.: The No.1 Engine

The No.1 engine of the Aircraft was

as follows. Furthermore, the following
sentence uses the clockwise numbering
shown on the periphery of the rotating axis
of the engine as its center as looking from
aft of the Aircraft to forward. The left and
right of the engine are indicated as seeing
from aft of the Aircraft to forward. (See
Photo #1.)
(1) General
The left side of the No.1 engine and the external components were burnt and
sooted. (See Photo #2. The left side of the No.1 engine.) The right side of the engine
did not have any sooting or thermal damage. The insulation blanket over the fuel
and hydraulic lines and the covering on the lines was undamaged. (See Photo #3.
the right side of the No.1 engine.) The main gearbox deoiler vent tube was severed
in line with the hole in the HPT case.

Photo #2. The left side of the No.1 engine

68
 Photo #3. The right side of the No.1 engine
(2) Inlet and fan blade
The inlet and fan blade did not have any apparent damage. There were no
trace of sucking objects like birds from outside. When the fan was rotated, the LPT
rotated concurrently. (See Photo #4. Inlet and Fan Blade.)

Photo#4. Inlet and Fan Blades

(3) LPC
The LPC inlet did not have any apparent damage. The stator vane leading
edges did not have any corresponding damage.
(4) HPC
The 15th stage compressor blades at the most rear HPC were all in place.
The tip of the blades were bent slightly opposite the direction of rotation. The
previous stage stator vane were all in place and had no damage.

69
(5) Diffuser Case
The outer diffuser case did not have any indications of a rupture or thermal
distress. The diffuser case rear flange between about 5:30 and 6:30 was bent
radially outward. The inner diffuser case had five cracks between the aft flange
and the center part of the case that varied in length from 1.5 (3.81) to 10.5(26.67
cm) inches. At about 12 o’clock, there was an approximately 2 inch (5.08 cm) by 1
inch (2.54 cm) piece missing from the case. (See Photo #5 Inner Diffuser Case)

Photo #5. Inner diffuser case

(6) The HPT
① The HPT case wall between 5:30 and 7 o’clock was bent radially outward and
twisted which was approximately 36 cm (approximately 14 in) long and
approximately 11.4 cm (4.5 in) wide along the periphery. (See Photo #6-1. The
damage of the HPT case and Photo #6-2. The damage of the HPT case.)

70
Photo #6-1. The damage1 of the HPT case

Photo #6-2. The damage2 of the HPT case

71
 It was confirmed that the 1st stage HPT blades and the 2nd stage HPT
stator vanes were missing from the roots. (See Photo #7. Fractured condition of the
1st stage HPT blades and the 2nd stage HPT stator vanes.)

Photo #7. Fractured condition of the 1st stage HPT blades and
the 2nd stage HPT stator vanes

② There was a raised edge on the edge of the tab that corresponded to a notch in
the edge of the hole in the starter air duct valve. (See Photo #8. A Hole opened in
Air Starter Valve.)

Photo #8. A Hole opened in Air Starter Valve

72
③ The HPT case’s forward flange was missing bolts between 3:30 and 8:30 and the
aft flange was missing bolts between 5:30 and 7:30.
④ Investigating the runway and an adjacent area after the accident, stator vanes
and others of the 1st stage HPT and the 2nd stage HPT were recovered. (See Photo
#9. Recovered Stator Vanes and others.)

Photo #9. Recovered Stator Vanes and others

⑤ A piece of the 1st stage turbine disk rim was recovered from amidst the debris
that was recovered from Runway 34R. The piece was about 7.8 inches (20 cm) long
and weighed 1.875 kg (4.134 lb) with seven blade slots. (See Photo #10. Piece of the
1st stage HPT disk rim.)

Photo #10. Piece of the 1st stage HPT disk rim

73
 There was a ratchet mark17 on the fracture surface at the 1st stage HPT disk
rim. (See Photo #11. Ratchet Marks.)

Photo #11. Ratchet Marks

⑥ As conducting a teardown inspection on the No.1 engine, approximately 20 cm (7

in) long piece from the rim of the 1st stage HPT disk was missing. The missing piece
surface and the fractured face of the fragment of the rim part shown in Photo #7
were matched.
(7) LPT
The LPT blade of the last stage LPT (the ninth stage) were all in place and did
not have any apparent impact damage. There were several ninth stage LPT blades
that had spatter on.
(8) Tail Cone (Exhaust Pipe)
Tail cone was still attaching to an exhaust pipe case. Nine tail cone mounting
bolts at the positions (the lower side) from 4:00 to 8:00 were missing. (See Photo
#12 .Tail Cone)

17
“Ratchet marks” are the step-like junctions between adjacent fatigue cracks that propagate
and link up.

74
 Photo #12. Tail Cone
(9) Lubricating Oil System
The lubricating oil was confirmed to be in the oil tank. The magnetic chip
detectors (MCD) for the accessory gearbox were pulled for examination, but the
MCDs did not have any debris or fuzz on the magnetic tips. The main oil filter and
the oil had no debris nor an acid odor.
(10) Fuel Oil Heat Exchanger
The main body case of Fuel Oil Heat Exchanger located at 8 o’clock of the
HPT case had cracks and soot by thermal damage, but there were no damage
impacted from outside. Three cracks were confirmed at the rear of the main body,
and the longest crack was approximately 34 cm (13.4 in) long. (See Photo #13-1
Fuel Oil Heat Exchanger and Photo #13-2 Cracks of Fuel Oil Heat Exchanger.)

75
 Photo #13-1. Fuel Oil Heat Exchanger

Photo #13-2. Cracks on the Fuel Oil Heat Exchanger

(11) Engine Cowl
① Inlet cowl
Impacting marks or damage was not confirmed on the right and left inlet
cowl.
② Fan cowl
No damage was confirmed on the right and left fan cowl.
③ Translating cowl18
The right and left translating cowls were in place and the latches were locked.
The aft edge of the left translating cowl had a large section, that was approximately
107 cm (42 in) wide axially at it widest by about 279 cm (110 in) long

18 “Translating Cowl” is the cowling of having a role of nozzle injecting air from the fan and a
role to inject the air from fan forward as providing a clearance between the fan cowl and sliding
back position of this cowl.

76
circumferentially missing from the aft edge between 6:30 to 9 o’clock that exposed
the core cowl. (See Photo # 14.Damage of Translating Cowl)
It was confirmed that the paint on the exterior of the aft side of the left
translating cowl below the missing section was blistered or burned away. The left
translating cowl (thrust reverser) inner wall had an approximately 94 cm (37 in)
long circumferentially by 79 cm (31 in) wide axially, an there was another hole that
was burned away. The interior of the left fan duct and translating cowl (thrust
reverser) inner wall were sooted. There were two pieces of the 2nd stage HPT vane
embedded in the inner surface of the translating cowl of the missing section.
④ The right side translating cowl
The right side translating cowl was not damaged, sooted or thermally
stressed. The inner wall of translating cowl and associated fan duct did not have
soot.
⑤ Aft cowl
The aft cowl pressure relief doors at 2, 4, 8 and 10 o’clock were all open.

Photo # 14. Damage of Translating Cowl

77
Attachment 4.: Photos of Fractured Surface

Photo ① and ②: Beach mark (Seashell pattern) from the originating point was
confirmed at the fatigue crack propagation area at the fractured surface of fractured
and ejected debris of the 1st stage HPT disk rim part.
Photo ③： Crack propagates from originating point at the aft side of the disk to
the forward side. The fractured surface exhibited significant mechanical damage.
The fatigue progression exhibited a bluish heat tint.
Photos ⑤ and ⑥ was the pictures enlarged and photographed parts enclosed by
yellow lines shown in Photo ④ via Scanning Electron Microscope (SEM) and
striation which is a characteristics of fatigue fracture surface was confirmed.

78
Attachment 5.: Striations of a Test piece at the crack

79