\
PB92-910405
NTSB/AAR-92/04
NATIONAL
TRANSPORTATION
SAFETY
BOARD
WASHINGTON, D.C. 20594
AIRCRAFI’ ACCIDENT REPORT
BRITT’ AIRWAYS, INC., d/b/a
CONTINENTAL EXPRESS FLIGHT 2574
IN-FLIGHT STRUCTURAL BREAKUP
EMB12ORT, N3370 1
EAGLE LAKE, TEXAS
SEPTEMBER 11,199l
The National Transportation Safety Board is an independent Federal agency dedicated to
promoting aviation, railroad, highway, marine, pipeline, and hazardous materials safety.
Established in 1967, the agency is mandated by Congress through the Independent Safety
Board Act of 1974 to investigate transportation accidents, determine the probable causes of
the accidents, issue safety recommendations, study transportation safety issues, and evaluate
the safety effectiveness of government agencies involved in transportation. The Safety
Board makes public its actions and decisions through accident reports, safety studies, special
investigation reports, safety recommendations, and statistical reviews.
Information about available publications may be obtained by contacting:
National Transportation Safety Board
Public Inquiries Section, RE-51
490 L’Enfant Plaza, S.W.
Washington, D.C. 20594
(202)382-6735
Safety Board publications may be purchased, by individual copy or by subscription, from:
National Technical Information Service
5285 Port Royal Road
Springfield, Virginia 22161
(703)487-4600
PB92-910405
NTSBIAAR-92104
NATIONAL TRANSPORTATION
SAFETY BOARD
WASHINGTON, D.C. 20594
AIRCRAFT ACCIDENT REPO.RT
BRITT AIRWAYS, INC., d/b/a
CONTINENTAL EXPRESS FLIGHT 2574
IN-FLIGHT STRUCTURAL BREAKUP
EMB-120RT, N33701
EAGLE LAKE, TEXAS
SEPTEMBER 11,199l
Adopted: July 21,1992
Notation 5612B
Abstract: This report explains the structural breakup in flight and crash of Continental Express
Flight 2574, an Embraer 120, in a cornfield near Eagle Lake, Texas. The safety issues discussed
in this report include the feasibility of developing a means to advise flightcrews of recent
maintenance work on aircraft and the need for reviewing regulations, policies and practices for
establishing required inspection items (RIIs) with a view toward developing more specific
identification of RIIs. Safety recommendations concerning these issues were made to the
Federal Aviation Administration.
Y
CONTENTS
EXECUTIVE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1.
1.1
1.2
1.3
1.4
1.5
15.1
15.2
15.3
1.6
1.6.1
1.6.2
1.7
1.8
1.9
1.10
1.11
1.12
1.13
1.13.1
1.13.2
1.13.3
1.14
1.15
1.16
1.16.1
1.17
1.17.1
1.17.2
1.17.3
1.17.4
1.17.5
1.17.6
FACTUAL INFORMATION
History of Flight ....................................................................................
Injuries to Persons .................................................................................
Damage to Aircraft ...............................................................................
Other Damage .......................................................................................
Personnel Information ...........................................................................
The Captain ..........................................................................................
The First Officer ...................................................................................
Management and Maintenance Personnel .............................................
Aircraft Information ..............................................................................
The Airplane .........................................................................................
Maintenance Information ......................................................................
Meteorological Information ..................................................................
Aids to Navigation ................................................................................
Communications ...................................................................................
Aerodrome Information .........................................................................
Flight Recorders ....................................................................................
Wreckage and Impact Information ........................................................
Medical and Pathological Information ..................................................
Flightcrew and Passengers ....................................................................
Maintenance Personnel .........................................................................
Air Traffic Control Personnel ................................................................
Fire .......................................................................................................
Survival Aspects ...................................................................................
Tests and Research ...............................................................................
Airplane Performance ...........................................................................
Additional Information ..........................................................................
Maintenance Records Review--General ................................................
General Maintenance Manual (GMM) ..................................................
Horizontal Stabilizer Maintenance ........................................................
Required Inspection Items (RIIs) ..........................................................
Right Elevator Replacement ..................................................................
No. 1 Engine Over-torque ......................................................................
...
111
V
1
4
4
4
5
5
5
6
8
8
9
9
9
10
10
11
11
15
15
16
16
16
17
17
17
23
23
23
24
30
31
32
2.
2.1
2.2
2.3
2.4
2.5
2.6
2.7
ANALYSIS
General . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Aerodynamic and Structural Failure Aspects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Flightcrew Preflight Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Maintenance Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Required Inspection Items (RIls) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Senior Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Regulatory Oversight . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
33
34
37
38
41
43
44
3.
3.1
3.2
Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ...f......
..
Probable Cause . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
48
50
4.
RECOMMENDATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
50
5.
APPENDIXES
CONCLUSIONS
Appendix A--Investigation and Hearing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Appendix B--Personnel Information . . . . . . . . . . . . . . . . . . . . . .
Appendix C--Summary of Maintenance Personnel Interviews ..............
Appendix D--Airplane Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..*................
Appendix E--Cockpit Voice Recorder Transcript . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Appendix F--Flight Data Recorder Information . . . . . ...............................
Appendix G--Letter from Embraer Aircraft Corporation .......................
Appendix H--Plots of Dynamic Flight Simulation Results ....................
Appendix I-- Safety Board Response to FAA ......................................
...............................
iv
55
56
60
64
65
77
79
81
83
EXECUTIVE SUMMARY
On September 11, 1991, about 1003 Central Daylight Time,
Continental Express Flight 2574, an Embraer 120, operating under Title 14 of the
Code of Federal Regulations, Part 135, experienced a structural breakup in flight
and crashed in a cornfield near Eagle Lake, Texas. The 2 flight crewmembers,
1 cabin crewmember and 11 passengers aboard the airplane were fatally injured.
The National Transportation Safety Board determines that the
probable cause of this accident was the failure of Continental Express maintenance
and inspection personnel to adhere to proper maintenance and quality assurance
procedures for the airplane’s horizontal stabilizer deice boots that led to the sudden
in-flight loss of the partially secured left horizontal stabilizer leading edge and the
immediate severe nose-down pitchover and breakup of the airplane. Contributing
to the cause of the accident was the failure of the Continental Express management
to ensure compliance with the approved maintenance procedures, and the failure of
FAA surveillance to detect and verify compliance with approved procedures.
The issues in this investigation focused on:
1.
The responsibilities of the Federal Aviation Administration
and aircraft manufacturers and operators to determine the
critical items and inspection levels of aircraft systems.
2.
The procedures for relaying and standardizing maintenance
shift turnover information.
As a result of this investigation, the Safety Board issued safety
recommendations to the Federal Aviation Administration on the feasibility of
developing a means to advise flightcrews of recent maintenance work on aircraft
and the need for reviewing regulations, policies and practices for establishing
required inspection items with a view toward developing more specific
identification of such items. Also, as a result of this investigation, on February 28,
1992, the Safety Board issued safety recommendations to the Federal Aviation
Administration that would enhance both flight standards surveillance of
Continental Express and flight standards Program Guidelines, including the
National Aviation Safety Inspection Program.
V
A
a
NATIONAL TRANSPORTATION SAFETY BOARD
WASHINGTON, D.C. 20594
AIRCRAFT ACCIDENT REPORT
BRITT AIRWAYS, INC., d/b/a
CONTINENTAL EXPRESS FLIGHT 2574
IN-FLIGHT STRUCTURAL BREAKUP
EMB-120RT, N33701
EAGLE LAKE, TEXAS
SEPTEMBER 11,199l
1. FACTUAL INFORMATION
1.1
History of the Flight
On September 11, 1991, about 1003, Central Daylight Time (CDT),
Continental Express Flight 2574, an Embraer 120, operating under Title 14 of the
Code of Federal Regulations, Part 135 (14 CFR 135), experienced a structural
breakup in flight and crashed in a cornfield near Eagle Lake, Texas.1
The flight, with call sign “Jetlink 2574,” departed Laredo International
Airport, Texas (LRD), about 0909, en route to Houston Intercontinental Airport
(IAH). Following takeoff, the flight was assigned a cruise altitude of flight
level 250 (FL250). The flightcrew was later instructed to descend to FL240.
After receiving a radar handoff, the flightcrew made initial radio
contact with Houston Air Route Traffic Control Center (Houston ARTCC) radar
controllers for the Eagle Lake sector at approximately 0948:43. At 0954:14,
Houston ARTCC instructed the flight to ‘I...cross five five miles southwest of
Intercontinental [IAH] at and maintain niner thousand.” At 0954:20, the flightcrew
responded, “OK fifty-five miles southwest of Intercontinental at niner thousand,
we’re out of flight level two four zero....”
At 0959:51, Houston ARTCC instructed the flight, “Jetlink twentyfive seventy-four, roger, fly heading zero three zero, join the Humble two three
four radial GLAND, rest of route unchanged.” The flightcrew responded at
lUnless otherwise noted, all times listed are local, CDT, based on the 24-hour
clock.
2
095957, “Zero three zero, join the GLAND six arrival, twenty-five seventy-four.”
The response was the last radio transmission from the flight.
Just prior to losing radio communications with the flight, the two
Houston ARTCC controllers for the Eagle Lake Sector were relieved by another
controller. During the position relief briefing, all three controllers noticed the loss
of the airplane radar beacon return for Jetlink 2574. At 1004:53, the radar
controller, who had assumed duty, initiated the first of four attempts to contact the
flight. The flightcrew did not respond. The radar controller then advised his
supervisor that radio and radar contact had been lost.
The cockpit voice recorder (CVR) revealed normal conversation
during the descent from FL240. Appendix E contains the CVR transcript.
Following the last radio transmission, at 0959:57, the CVR recorded the flightcrew
receiving automated terminal information service (ATIS) “Golf’ on radio No. 2,
about 1000:03.
At 1003:07, the cockpit area microphone (CAM), as recorded on the
CVR, picked up sounds of objects being upset in the cockpit. These sounds were
followed immediately by one that was comparable to a human “grunt.”
The remaining sounds were warnings produced by the airplane’s aural
warning systems, as well as mechanical sounds indicating breakup of an aircraft in
flight. The sound of wind was picked up by the CAM, beginning at 1003:13. The
CVR tape stopped at 1003:40, about 33 seconds after the onset of the sound of
objects being upset in the cockpit. The entire CVR recording lasted for 31 minutes
and 6 seconds.
Radar data and a readout of the airplane’s flight data recorder (FDR)
showed the airplane in descent, passing through about 11,800 feet mean sea level
(msl), when a sudden pitchover occurred. The FDR data showed that there was
then a sudden negative vertical acceleration of at least 3 l/2 negative g, as well as
roll and yaw moments, heading changes, and sudden changes in engine
parameters.2
21 g equals 1 times the force of gravity. The FDR was limited in negative
vertical acceleration readout to 3.375 negative g. The data reached that point on several
occasions, following the initial negative acceleration. Appendix F contains relevant FDR data
plots.
3
Prior to the pitching over of the airplane, the engines were operating
normally. At the start of the sudden pitchover, FDR data revealed a sudden
oscillation in propeller speed, recorded in percentage of standard revolutions per
minute (rpm). Propeller rpm initially decreased from what had been a constant
85 percent for both engines. However, within 2 seconds, the r-pm for both engines
increased. No. 2 engine rpm decreased again, but then increased to well over
100 percent until the data ended.
All the eyewitnesses who were interviewed observed the occurrence
from the ground. A total of eight witnesses reported that they saw the airplane for
at least part of the time after they realized it was in distress, until impact.
The following describes various eyewitness observations of the
airplane:
flying normally
wings level, slightly nose down3
suddenly consumed by fireball
wingtips and part of tail protruding from fireball
a bright flash
orange or red-orange flames at time of flash or immediately
thereafter
sputtering engines, followed by three pops
a revving sound
a flat spin to the left until impact
left wing dangling from blown out area
right wing missing
3Manufacturer-provided data indicate that a negative lo-degree pitch angle
(before the sudden attitude and other changes) is normal and could be expected for the
conditions of the accident flight.
4
.
flying parts during downward spiral
After impact, the airplane was upright, in a wings-level attitude,
partially imbedded in the ground and burning.
Local fire and rescue personnel responded to the crash and
extinguished the fire. All persons aboard the airplane were fatally injured, and the
bodies of two of the occupants were lying outside of the airplane. Both pilots were
still strapped in their seats.
The accident occurred in visual meteorological conditions (VMC), $
daylight. The main wreckav, including the cockpit and cabin, came to rest at 29
30’ 98” north latitude and 96 23’ 21” west longitude.
1.2
Injuries to Persons
Flightcrew Cabincmw
Fatal
Serious
Minor
None
Total
1.3
2
0
0
4
2
1
0
0
n
1
Passengea Other Total
11
0
0
-Q
11
0
0
0
-0
14
0
0
4
14
Damage to Aircraft
The airplane was destroyed in the crash and fire. The airplane was
valued at around $7.75 million.
1.4
Other Damage
There was no claim for damage to the harvested cornfield and pasture
land into which the main wreckage and other portions of the airplane fell.
I
1.5
Personnel Information
1.51
The Captain
The captain, age 29, was born on April 20, 1962. He was hired by
Continental Express Airlines on October 10, 1987. He held airline transport pilot
certificate No. 565336474, with ratings for the EMB-120 and Airplane Multiengine
Land. His most recent Federal Aviation Administration (FAA) first-class medical
certificate was issued on July 18, 1991, with the limitation: “Holder shall wear
correcting lenses while exercising the privileges of his airman certificate.”
Company records indicate that at the time of the accident the captain had
accumulated approximately 4,243 total flying hours, of which 2,468 were in the
EMB-120.
The captain received his initial ground school and proficiency check
in the EMB-120 as a first officer, completing the training on October 29, 1988. He
completed upgrade ground school training on September 21, 1989, and received a
type rating in the EMB-120 on September 29, 1989. He completed his initial
operating experience and received a line check on October 2, 1989. His last
proficiency check was on March 9, 1991. His last recurrent training was
completed on May 29, 1991, and his last line check was accomplished on
August 8,199l.
1.5.2
The First Officer
The first officer, age 43, was born on November 9, 1947. He was
hired by Continental Express Airlines on March 12, 1990. He held airline transport
pilot certificate No. 1963386, with ratings for the EMB-120 and Airplane
Multiengine Land. His most recent FAA first-class medical certificate was issued
on August 30, 1991, with no limitations. Company records indicate that, at the
time of the accident, the first officer had accumulated approximately 11,543 total
flying hours, of which 10,300 were obtained prior to his employment with
Continental Express. He had a total of 1,066 hours in the EMB-120.
The first officer completed initial ground school in the EMB-120 on
March 30, 1990. He completed flight training on April 19, 1990. His initial
operating experience and line check were completed on April 24, 1990. He was
subsequently upgraded to captain on the EMB-120, completing that training and
initial operating experience on May 14, 1990. Although he no longer held a
regular captains bid number, the airline allowed the first officer to retain his
6
currency as a captain. He received proficiency checks on October 29, 1990, and
April 11,199l.
1.5.3
Management and Maintenance Personnel
The president, age 51, was hired in July 1990, as President of the
Commuter Division of Continental Airlines, Inc. Continental Express is a wholly
owned subsidiary of Continental Airlines. He had worked previously for Eastern
Airlines (owned by the same parent company as Continental and Continental
Express), from January 1987 to July 1990, in several successive positions: Staff
Vice President and Counsel for Regulatory Compliance; Vice President for Base
Maintenance; Special Assignment; and Vice President for Administration. Prior to
his employment with Eastern Airlines, he had worked for New York Air (19801986) and had served as its Vice President for Operations. He holds a commercial
pilot certificate with ratings and limitations of airplane single engine land with
instrument privileges. He also holds a private pilot certificate with ratings and
limitations of airplane multiengine land.
The Senior Director of Maintenance and Engineering, age 48, was
hired in August 1990. He had worked previously for Eastern Airlines, from
September 1989 to August 1990, as Manager of Special Projects. From June 1987
to June 1989, he worked for Aloha Airlines, first as Director of Quality Control
and then as Director of Maintenance. His Airframe and Powerplant License was
issued on April 10, 1968.
The Senior Director of Quality Assurance and Control, age 46, was
hired in February 199 1. He had worked for Eastern Airlines from 1969 to 1991
and had served as Manager of Aircraft Inspection. His Airframe and Powerplant
certificate was issued on June 26, 1979.
The second shift supervisor, age 28, who was in charge of N33701,
was hired by Continental Express on April 9, 1988, as a mechanic. He was
promoted to shift supervisor on January 19, 1990. His previous employment
included service with the U. S. Army from 1982 to 1985. His Airframe and
Powerplant certificate, number 383749034, was issued on December 19,1987.
The second shift inspector, age 25, who removed the attaching screws
from the tops of the left and right horizontal stabilizer leading edge assemblies,
was hired on July 11, 1989, as a mechanic. He was promoted to inspector on
October 24, 1990. His previous employment included service as an aircraft
7
electrician in the U. S. Navy. His Airframe and Powerplant certificate, number
456456725, was issued on February 5,1989. The inspector had received company
discipline on two occasions that related to inspections. In August 1991, he
received a warning for having “missed a crack...in inspection of engine exhaust
stack.” He received a second warning that month because he “did not finish all
paperwork required...missed 15 task cards on the accountability sheet.”
The company had a written policy for disciplinary action that included
the following forms of progressive discipline: verbal counseling or reprimand;
formal counseling and written warning; probation; suspension; dismissals; and
immediate dismissal without notice. According to the written policy, “there is no
precise formula for applying discipline” so no specific action would be taken after
a specific number of warnings.
The second shift mechanic, age 43, was hired on July 2, 1990, as a
mechanic. His previous employment included work as an aircraft mechanic with
Continental Air Micronesia (1989-1990), and flight line mechanic and inspection
dock chief with the U. S. Air Force (1986-1989). He holds Airframe and
Powerplant certificate number 45 1760789 issued on March 7,199O.
The second shift supervisor, age 29, (who was not responsible for
N33701), was hired on October 25, 1987, as a mechanic. He was promoted to
inspector in 1989 and to shift supervisor on January 19, 1990. He was previously
employed as an airplane mechanic for two fixed-based operators (1987) while he
completed school. His Airframe and Powerplant certificate, number 451396613,
was issued on January 26,1988.
The third shift supervisor, age 26, was hired by Britt Airways, Inc.,
(later merged into Continental Express), on June 8, 1987, as a mechanic at the air
carrier’s Cleveland base. He was promoted to an inspector on November 27, 1989,
transferred to the Houston base as a mechanic on March 16, 1990, and was
promoted to shift supervisor on August 17, 1990. His previous employment
included work as a helicopter mechanic and crew chief in the U. S. Army
(1984-86), and as a jet engine mechanic in the U. S. Air Force Reserves (1986-87).
His Airframe and Powerplant certificate, number 312767386, was issued on
June 16,1989.
The third shift inspector, age 36, was hired by Britt Airways, Inc.,
(later merged into Continental Express), as a maintenance helper at the
Bloomington, Indiana, base on September 1, 1982. He was promoted to aircraft
8
mechanic in 1986. In 1989, he spent 9 months at the Houston base where he was
promoted to inspector. He returned to the Houston base as an inspector on May 1,
1991. His Airframe and Powerplant certificate, number 347508432, was issued on
April 26,1986.
The hangar workers, consisting of mechanics, inspectors, and
supervisors, who were directly involved in work on the tail structure of the
airplane, represented about 23 percent of the second shift workers and 21 percent
of the third shift workers employed by Continental Express at the time of the
accident. Together, they represented about 15 percent of the entire hangar
workforce from all shifts.
1.6
Aircraft Information
1.6.1
The Airplane
The airplane, U.S. registration N33701, was an Embraer EMB-120,
manufactured in Brazil. The serial number was 120-L77. Continental Express
Airlines acquired the airplane on April 15, 1988. Records showed that the airplane
had accumulated 7,229.8 hours and 10,009 cycles as of September 10, 1991. The
airplane was configured with 10 rows of double passenger seats on the right side of
the cabin and 10 rows of single passenger seats on the left side of the cabin.
The gross takeoff weight for the airplane, upon departure from LRD
on the accident flight, was calculated by the flightcrew as 22,272 pounds, including
1,815 pounds for passengers, 259 pounds for cargo, and 3,100 pounds for takeoff
fuel. The calculated weight for the takeoff from LRD was 3,081 pounds below the
maximum allowable takeoff weight of 25,353 pounds.
The airline’s EMB-120 Aircraft Operations Manual stated, “The
balance of the aircraft is controlled by the load in the aft cargo hold. To keep
aircraft CG [center of gravity] within allowable limits, there are minimum and
maximum loads for the aft cargo hold which vary as the passenger load varies.”
A table provided in the airline’s Alert Bulletin 91-17, dated
September 3, 1991, established a minimum weight of 78 pounds and a maximum
weight of 794 pounds for a passenger load of 11 persons. The documented load of
259 pounds in the aft cargo hold was within CG limits.
9
1.6.2
Maintenance Information
The procedures for maintaining the airplane were contained in the
airline’s General Maintenance Manual (GMM), which was approved by the FAA
(See section 1.17.2). A review of the maintenance records for N33701 was
conducted, and personnel responsible for the maintenance and inspection of
N33701 the night before the accident were interviewed (See section 1.17.1).
1.7
Meteorological Information
There were no significant meteorological information (SIGMET)
advisories or center weather advisories (CWAs) in effect for the area before or
after the time of the accident.
The weather conditions reported by the National Weather Service for
Palacios, Texas, which was the nearest reporting station to the accident site, were:
0950 (about 15 minutes prior to the accident):
Estimated ceiling 3,000 feet broken, 10,000 feet broken,
25,000 feet overcast, visibility 6 miles, haze, temperature
83 degrees, dewpoint 74, wind 070 degrees at 7 knots, altimeter
30.08.
At 1050, about 45 minutes after the accident, the reported weather
at Palacios, Texas, was:
Estimated ceiling 3,000 feet broken, 10,000 feet broken,
25,000 feet broken, visibility 7 miles, haze, temperature
86 degrees, dewpoint 74, wind 070 degrees at 7 knots, altimeter
30.03.
1.8
Aids to Navigation
At 095951, Houston ARTCC directed the airplane to “...fly heading
zero three zero, join the Humble two three four radial GLAND, rest of route
unchanged.” This radio transmission was the last one that the flight acknowledged.
At the time of the accident, the airplane was in a descent under
positive radar control by Houston ARTCC, Eagle Lake Sector, and had been
10
instructed to intercept the radial. There were no difficulties regarding aids to
navigation or air traffic control (ATC) reported in this accident.
1.9
Communications
Houston ARTCC’s communications with the flight took place for
approximately 11 minutes, beginning at 0948:43, when the flight reported in,
“Houston Center Jetlink twenty-five seventy-four flight level two four zero.” The
last transmission from the flight occurred at 0959:57, with Jetlink 2754
acknowledging Houston ARTCC’s instructions to “...join the GLAND six
arrival....” (See appendix E).
Neither the CVR nor ATC tapes indicate any communication
difficulties between the crewmembers nor between the flight and air traffic
controllers until after communications in the airplane and from the airplane were
lost. From the beginning of the CVR recording, at 0933:36, until the sound of
objects moving in the cockpit, at 1003:07, there is no difficulty indicated in any of
the communications or background sounds. The first officer, however, remarked at
0936:29, “Do you smell something like paint thinner?” and the captain replied, “A
little bit, yeah.”
The first indication that there might have been some difficulty was the
lack of response to three calls from the Houston ARTCC Eagle Lake Sector
controller to “Jetlink twenty-five seventy-four, say altitude,” at 1004:53, 1005:12,
and 1005:32. All three controllers for the Eagle Lake Sector (two outgoing and
one incoming) noted about the time of the change to the relief controller that the
radar return for Jetlink 2574 had disappeared from the screen.
1.10
Aerodrome Information
The flight was inbound to IAH. The airport elevation is 98 feet msl.
The airport is operated continuously. There are four primary nonintersecting
runways, the longest of which, 14L/32R, is 12,000 feet long by 150 feet wide.
There were no difficulties reported regarding any aerodrome in this
accident.
11
1.11
Flight Recorders
The CVR and PDR were recovered from their installed positions in
the aft portion of the airframe. There was minor damage to the recorder cases from
impact forces. The recorders showed no evidence of having been subjected to fire.
The CVR recording was clear and showed no evidence of loss in quality as a result
of crash damage. The PDR recording was also of good quality.
1.12
Wreckage and Impact Information
Separated parts of the airplane, including all eight propeller blades,
were within about a 1.5 nautical mile radius of the main wreckage.
The horizontal stabilizer, or top of the T-type tail, had separated from
the airplane before impact and was lying about 650 feet west-southwest of the main
wreckage. Some of the structure and skin from approximately the upper third of
the vertical stabilizer were still attached to the horizontal stabilizer. The lower two
thirds of the vertical stabilizer remained attached to the tail cone in the main
wreckage. The leading edge/deice boot assembly for the left side was missing
from the horizontal stabilizer. The left side leading edge/deice boot was later
found by investigators in a small corral about 3/4 mile west of the main impact site.
The left engine and propeller assembly, minus the four propeller
blades, was lying approximately 370 feet south-southeast of the main wreckage.
The left wing was in the wreckage, still attached to the fuselage by the lower
attachment points, but it was folded under the fuselage and the inboard portion of
the right wing. The right wing was in its proper position, still attached to the main
fuselage. Part of the right wing tip was found about l/5 mile west of the main
impact site. The right engine remained attached to the right wing, and the four
propeller blades were separated from the propeller hub assembly.
Both engines and propeller systems, including the eight separated
propeller blades, were sent to the facilities of the engine manufacturer for
disassembly and inspection, under the supervision of the Safety Board. The
disassembly and inspection determined that the right engine had over-sped and
overtorqued before impact. The left engine had no evidence of over-speed or
overtorque. The eight propeller blades that had separated from their attaching
points to the hubs, and the hub side attaching points, were fractured. There was no
evidence of a defect or anomaly in either engine or propeller assembly prior to the
unusual attitudes and in-flight breakup of the airplane. The damage to the engines
12
and propellers was compatible with the results of extreme changes in airplane
attitudes, and, in the case of the left engine, separation from the airplane before
ground impact.
The Colorado River, flowing approximately north to south, ran about
1.2 miles west of the main crash site. An agricultural pilot, who flew over the
crash site shortly after impact, reported seeing a piece of airplane wreckage
floating down the river. However, investigators did not find any wreckage in the
river.
During the Safety Board’s examination of the wreckage, none of the
47 screws that would have attached the upper surface of the leading edge assembly
for the left side of the horizontal stabilizer was found. There was no evidence of
distress in the upper attachment holes for the left side leading edge assembly or
indication that the attaching screws were installed when the left side leading edge
assembly separated from the horizontal stabilizer. In addition, a “lip” was formed
on the forwardmost frame on the left lower side of the horizontal stabilizer spar
cap. That frame (spar cap), with receptor holes for the lower attaching screws, was
the area into which the screws mounted the underside of the left side leading edge
assembly to the stabilizer. This lower frame area showed signs of distress. Figures
la through Id show the condition of the left horizontal stabilizer leading edge.
The lower attachment screws remained installed, but the leading edge
assembly had separated from the stabilizer, with the exception of a small portion of
composite structure remaining below the two farthest inboard screw heads. The
spar cap on the lower left side of the horizontal stabilizer showed evidence of being
pulled down so that it would project into the wind stream along with the leading
edge. This pulling damage is consistent with the left side leading edge assembly
having been ripped down and away from the lower attaching screws as it separated
from the horizontal stabilizer. This evidence was consistent with screws missing
on the top side of the left leading edge assembly, and the lower attaching screws
holding fast, pulling down the frame (spar cap) on the lower side of the stabilizer,
and thereby forming the lip.
The main portion of the airplane came to rest upright and partially
imbedded in the cornfield on a heading of about 360 degrees. There was no
indication that the main wreckage moved after initial ground impact.
The crash site was approximately 3 miles south-southwest of the town
of Eagle Lake, Texas, and 60 nautical miles west-southwest of IAH.
13
Figure 1 a.--View of horizontal stabilizer from underside.
Figure lb.--Leading edge assembly and outboard portion
of left side of horizontal stabilizer.
Figure lc.--Front frame of left side of horizontal stabilizer
with lower screw attaching area angled downward.
Figure ld.--View of interior leading edge assembly for left side
of horizontal stabilizer.
Note that upper screw attachment holes (lower) show no signs of distress.
15
The nose section and the bottom surface of the forward section of the
fuselage were crushed. The extreme aft section of the fuselage, including the still
attached upper 2/3 of the vertical stabilizer, had compression impact damage.
The fuselage cargo door that was found 18 feet from the main
wreckage had deep grooves and scratches in the outer skin. Instantaneous
overloading was apparent on the bayonet fittings and roller attachments at the
forward and aft cargo door frames. The lower half of the cabin boarding door
remained attached to the fuselage; and the door operating handle was in the stowed
position. The main landing gears and the nose gear were in their stowed positions.
The nose landing gear was displaced upward by impact forces.
1.13
Medical and Pathological Information
1.13.1
Flightcrew and Passengers
Autopsies performed on the 3 crewmembers and 11 passengers by the
Harris County Coroner’s Office, Texas, determined that all occupants sustained
fatal traumatic injuries consistent with sudden impact. Two occupants were
ejected from the aircraft at impact. Most persons who were found inside the
airplane were subjected to the postimpact fire. No evidence of preimpact fire
injuries or smoke inhalation by occupants was found.
Toxicological analyses were completed on specimens of the captain’s
blood and urine and on other tissues of the first officer and flight attendant because
samples of their blood and urine were not available. The captain’s test results were
negative for carbon monoxide, hydrogen cyanide, alcohol, and other licit and illicit
drugs. Test results of the first officer and flight attendant were negative for licit
and illicit drugs except alcohol. A liver sample from the first officer tested positive
for alcohol at a level of .06 percent, and a bile sample from the flight attendant
tested positive for alcohol at a level of .07 percent. Evidence of heat coagulation
was noted in all tissue samples of the first officer and flight attendant that were
examined. Heat exposure can accelerate putrefaction and post-mortem production
of alcohol. A second testing of samples by another laboratory found higher levels
of alcohol, but the laboratory director noted that putrefaction of the samples had
occurred prior to their arrival at the laboratory.
16
1.13.2
Maintenance Personnel
During the on-site portion of the investigation, a request was made by
the Safety Board for urine and blood samples from the 12 persons who had been
involved in the maintenance of the airplane on the evening and midnight shifts on
September 10 and 11, 1991. They included two mechanics, two supervisors, and
an inspector from the second or evening shift; and four mechanics, two supervisors
and one inspector from the third or night shift.
Blood and urine samples were obtained by the airline’s office of
Human Resources and Drug Abatement. The samples were obtained for 11 of the
individuals during the night work period of September 14 through 15, and from the
remaining person the following morning, September 15, 1991. The samples were
provided to the Safety Board and were tested. All test results were negative for
alcohol and drugs of abuse.
1.13.3
Air Traffic Control Personnel
About 1300, September 11, 1991, 2 hours after the accident, the
Safety Board asked the FAA for urine and blood samples from the air traffic
controllers at the Houston ARTCC. Samples were voluntarily provided by the two
controllers who last spoke to the flightcrew. Also, samples were provided by the
controllers’two supervisors. The samples were submitted and retained under
Safety Board authority.
Because there was no evidence of air traffic controller involvement in
the accident, the samples obtained from the two controllers and their supervisors
were not analyzed. The samples were subsequently returned to these individuals.
1.14
Fire
There was a fire in flight, as well as after ground impact. This was
confirmed by eyewitnesses and wreckage examination.
The horizontal stabilizer and about l/3 of the upper vertical stabilizer
had separated from the airplane before ground impact. The horizontal stabilizer,
with about 3 feet of the uppermost vertical stabilizer still attached,‘contained some
light soot deposits. A broken edge of composite material that spanned the upper
surface of the horizontal stabilizer, along the center line of the horizontal stabilizer,
showed a small burned area. Although there were bits of molten aluminum
splattered on the lower two thirds of the vertical stabilizer, there was no evidence
17
of molten aluminum splatters on the upper portion of the vertical stabilizer or the
horizontal stabilizer.
The lower two thirds of the vertical stabilizer that remained attached
to the fuselage was found in place in the main wreckage. Bits of molten aluminum
were found splattered on the left surface of this lower portion of the vertical
stabilizer.
Approximately the lower half of the primary and secondary rudder
control surfaces that remained attached to the lower portion of the vertical
stabilizer showed heat damage, including molten aluminum splatters. The upper
half of the rudder control surfaces, which was found as a unit in a field
approximately 4/10 mile west of the main wreckage, showed no evidence of smoke
deposits or fire damage. The upper and lower sections of the rudder control
surfaces were placed together, and a clear demarcation line was seen where the
rudder surfaces had broken.
1.15
Survival Aspects
The accident was not survivable.
The police chief learned about the accident about 1010, and the first of
two 350 gallon, four-wheel drive mini-pumper fire trucks arrived at the accident
site around 1020. The fire was nearly extinguished when the first truck arrived,
and limited effort was required to extinguish the remaining flames. In total, about
12 volunteer firemen and 6 ambulances responded to the crash.
1.16
Tests and Research
1.16.1
Airplane Performance
The airplane was flying to the northeast on a 44-degree heading at the
time radar contact was lost at 1003:06 CDT. Figure 2 shows the radar-derived
ground track of flight 2574, selected sounds from the CVR, and the wreckage
distribution.
Figure 3 provides a closeup view of part of the ground track and
wreckage distribution. The piece of airplane structure farthest from the main
wreckage was the left side leading edge (LE) of the horizontal stabilizer. The LE
I
18
CONTINENTAL EXPRESS FLIGHT 2574
0.75 1
+
i
.i . . . f&4Q?.0&-FI’Gs. . . . .
v
-1.25
z
'
s
= -3.25
1882:14 - huh huh.
1662:11- push'n this descent making like
-4.25
-5.25
-5.75
I
-4.75
I
I
I
-3.75
I
-2.75
1
I
- 1.75
I
1
-0.75
0.25
EAST RANGE - NM1
‘,Figure 2.--Radar-derived ground track, CVR sounds, and wreckage distribution.
k
19
CONTINENTAL EXPRESS FLIGHT 2574
7
0.5
. . . . .+.- WRECKAGE
RADAR RET"RN . . . . . . . . . . . . . . . . . . +..
+
INSULATION
z
y-
+
INSULATION
0
RIGHT
WINGTIP
MAIN
]i
..;..
WRECKAGE
rc'
+- EEE
\
+
E
iY
s?
HORIZ STAB
t
HORIZ STAB
LE RADIUS
-0.5
‘?
WIND DIREC
-1
-1.5
L
-1
-0.5
0
EAST RANGE - NM1
Figure 3.--Closeup view of part of ground track and wreckage distribution.
20
was the first piece of structure along the flight’s northeasterly ground track,
preceding the next piece by roughly l/2 of a nautical mile.
Figure 4 shows the radar-derived descent profile. The last radar
contact occurred as flight 2574 was descending through 11,800 feet. The radarderived rate of descent during the final minute was approximately 4,000 feet-perminute, which is consistent with FDR data from the airplane.
The Safety Board used F’DR data, CVR data, and engineering
calculations from Embraer to study the motion and breakup of the airplane during
the accident sequence. The flight dynamics of the accident were simulated by
Embraer at the request of the Safety Board. Flight parameters at the time of the inflight upset, including airspeed, altitude, acceleration, and airplane attitude, were
examined. The leading edge separation from the left horizontal stabilizer was
examined, as was the separation of the entire horizontal stabilizer from the
airframe. The known flight characteristics of the airplane, before the sudden inflight changes, were used to examine the events during the accident sequence.
The FDR data show that the airplane was descending through
11,500 feet (pressure altitude) at 260 knots indicated airspeed (KIAS) when it
abruptly pitched down and entered a steep dive. The airplane was 12 knots below
the upper limit (272 KIAS) of the EMB-120 airspeed envelope when the upset
occurred. The FDR data showed that a negative load factor of at least -3.375 g was
reached about 1 second after the upset, with a corresponding decrease in airplane
pitch attitude. The peak negative acceleration is unknown because the FDR’s
recording limit of -3.375 g was reached. The normal acceleration then fluctuated
between about -0.6 and -2.4 g until the lower recording limit was reached again,
6 l/2 seconds after the upset began. At that point, the data show the airplane
descending through 9,500 feet at 280 KIAS.
During the first 6 l/2 seconds after the upset began, the data showed a
roll of 10 to 15 degrees right wing down and a nose-left heading move from 52 to
33 degrees. During the same period, the lateral acceleration was as much as l/2 g.
After the 6 l/2 second period, the airplane abruptly rolled to the right more than
160 degrees in 1 second. During that 1 second, the airplane pitch attitude reached
the minimum recorded value of -86 degrees, and then it began increasing. Normal
acceleration went from about -0.5 to +2 g. Lateral acceleration went from about .05 g to the recorded positive limit of +1 g, stayed at that limit for
21
CONTINENTAL EXPRESS FLIGHT 2574
-
30000
I
25000
1
.............. ..............
+
*
d?
20000
!!?zzh-
15000
........................
. x.0Q. $+. . . . . . .
3 . . . . . . . . . . . . fp. & .
%h, .
..............
-16
.
/
......
5000 b ..__..........._..................... f. _....._.._._.._._._..................
-21
.
MAIN
WRECKAGE
-11
-6
-1
RANGE - NM1
Figure 4.--Radar-derived descent profile.
4
22
several seconds, and then went to the negative limit of -1 g before the F’DR ceased
operation.
According to Embraer EMB- 120 engineering data, the horizontal
stabilizer angle of attack in steady-state flight at 260 KIAS is -2 degrees. An
aerodynamic stall (loss) of the left side horizontal stabilizer reduces the downward
lift vector (downforce) at the tail (which is needed to maintain steady flight), and a
large nose-down pitching moment is produced that leads to a negative wing stall.
Calculations show that the wing stall occurs within 1 l/2 seconds of the tail stall,
with a peak negative acceleration of about -5 g.
Two dynamic flight simulations were conducted in an attempt to
determine whether the data obtained from the FDR would match the circumstances
of a sudden loss of the left stabilizer leading edge or a sudden loss of the entire
horizontal stabilizer. The results of the two simulations are shown in Appendix H.
Because of limitations in available FDR data and the highly dynamic motion, the
flight dynamic predictions could only examine the first 1 l/2 seconds of the flight
after the upset.
Neither simulation could precisely duplicate the performance of the
accident airplane as recorded by the F’DR. The first simulation (that assumed a
sudden loss of the left horizontal stabilizer leading edge) showed a less severe
pitch down and negative load factor, while the second simulation (that assumed
loss of the entire horizontal stabilizer) was more severe than the FDR data recorded
for the accident flight. The first simulation incorporated a loss of downforce and
increase in drag that are estimated and consistent with standard aerodynamic
practices. Because lift from the horizontal stabilizer was directed downward, a
transient rise in normal acceleration occurred after the leading edge detached and
the downward lift was lost. In the first simulation, this “g” increase was equivalent
to l/2 of the downforce produced by the horizontal stabilizer. A transient rise in
acceleration also exists in FDR data. However, lift and drag forces would change
significantly during the dynamic motion of the airplane and would be virtually
impossible to duplicate exactly in an engineering simulator.
Calculations made by Embraer of the lift required from the horizontal
stabilizer during both postmaintenance flights show that the peak stabilizer
downforce occurred at the time of upset on the accident flight. The maximum
downforce produced by the horizontal stabilizer during the previous flight was at
least 30 percent lower than that achieved just prior to the accident.
23
Embraer was asked to provide the Safety Board with a structural
analysis report to evaluate the effects of airloads on the airplane structure after
separation of the left horizontal stabilizer leading edge, The calculations showed
that the predicted airloads on the horizontal stabilizer and vertical stabilizer
structure for a loss of the leading edge did not exceed the maximum allowable load
for the vertical and horizontal stabilizers.
The airspeed experienced at the time of the in-flight breakup,
260 KIAS, while below the manufacturer’s maximum allowable airspeed of
272 KIAS, was the highest airspeed experienced on either flight following the
maintenance. The highest airspeed recorded on the FDR on the first flight of the
day, from IAH to LRD, was 216 KIAS.
1.17
Additional Information
1.17.1
Maintenance Records Review--General
The Safety Board examined in detail the recent maintenance history of
N33701 related to the work conducted on the horizontal stabilizer. This effort
involved a review of paperwork and procedures and extensive interviews of
maintenance personnel associated with the work on the airplane the night before
the accident. Additionally, the past full year of records were examined for items
related to airworthiness directive (AD) actions and actions related to
engine/propeller and flight control discrepancies. Further, the Safety Board
reviewed Continental Express’ FAA-approved General Maintenance Manual
(GMM) and its required inspection item (RII) program.
No discrepancies were noted with AD compliance.
Certain
discrepancies were noted with respect ‘to actions taken during a past replacement
(April 24, 1991) of the right elevator on N33701, and as the result of an overtorque
on the No. 1 engine (September 24, 1990). See sections 1.17.5 and 1.17.6 for
additional details.
1.17.2
General Maintenance Manual (GMM)
The Continental Express GMM had FAA-approved procedures.
GMM 1, Section 1, Paragraphs l-6, states that “personnel performing maintenance
will follow and be familiar with the instructions as outlined herein....Instructions
and information, contained herein, bring Continental Express into compliance with
24
the appropriate Federal Aviation Regulations. For this reason, it is essential that
the contents be followed.”
GMM 1, Section 3, Paragraph 10, specifies that it is imperative for
maintenance/inspection forms to be completed to ensure that no work item is
overlooked. Such work includes the completion of maintenance/inspection shift
turnover forms, so that oncoming supervisory personnel can be made aware of
complete/incomplete work, and the documentation of incomplete work that the
mechanic can note on the reverse side of the M-602 work cards. GMM 1,
Section 5, Paragraph 7, specifically addresses several methods to ensure proper
turnover during shift changes. These methods include briefings by mechanics to
supervisors and briefings by outgoing supervisors to incoming supervisors.
The GMM contained provisions for a lead mechanic position in the
organizational structure of the maintenance department. That position was not
filled at the IAH maintenance base. According to the FAA maintenance inspector
responsible for oversight of the Continental Express maintenance facilities, the lead
mechanic position was identified in the organizational structure of one of the
merger airlines. That position did not exist at the other merger airline. Instead, the
supervisor was assigned to perform the functions assigned to the lead mechanic.
Therefore, the lead mechanic position did not exist at the IAH maintenance base
and, according to the FAA inspector, would not be considered a deviation from or
violation of the provisions of the GMM for the Houston base.
1.17.3
Horizontal Stabilizer Maintenance
The review of the maintenance records for N33701 revealed that on
August 26, 1991, during the airline’s fleet-wide campaign to examine aircraft deice
boots for winter operation, a quality control inspector had noted both leading edge
deice boots as “watch list” items on M-602 work cards because of “dry rotted pin
holes entire length” [of the boots]. On September 10, 1991, the night before the
accident, Continental Express’ Maintenance Control office scheduled both
horizontal stabilizer leading edge deice boots on N33701 for replacement.
A series of interviews was conducted from September 13 through 16,
1991, and from October 22 through 24, 199 1, with airline maintenance personnel,
inspectors, and supervisors who were working the night before the accident. These
personnel worked on the airplane on the second or “evening” shift and third or
“midnight” shift. During the first series of interviews, seven mechanics, four
maintenance supervisors, and three quality control inspectors were interviewed.
During the second series, one mechanic, one inspector, and two supervisors were
reinterviewed; and two senior directors and two FAA principal maintenance
inspectors were interviewed for the first time.
The interviews revealed that the night before the accident, the airplane
was pulled into the Continental Express hangar at IAH during the second shift at
about 2130 hours for scheduled maintenance. The scheduled maintenance included
the removal and replacement of both the left and right horizontal stabilizer deice
boots.
A change of either the left or right deice boot required that the leading
edge/deice boot assembly for that side of the horizontal stabilizer be removed from
the stabilizer. Normally, while still attached to the stabilizer, the old deice boot
would be stripped from the composite structure of the leading edge, the deice fluid
lines would be disconnected, and the leading edge would be removed and a new
deice boot bonded on. Then, the leading edge/deice boot assembly would be
reinstalled on the horizontal stabilizer by means of approximately 47 attaching
screws for each of the top and bottom sides of the assembly.
Two second shift mechanics, with the assistance of an inspector,
gained access to the T-tail, which was about 20 feet above the ground, by means of
a hydraulic lift work platform. The work was assigned by the second shift
supervisor who took charge of N33701. The two mechanics removed most of the
screws on the bottom side of the right leading edge and partially removed the deice
boot bonded to the front of the right side leading edge.
The inspector who had climbed on top of the T-tail had removed the
attaching screws on the top of the right side leading edge and then walked across
the T-tail and removed the attaching screws from the top of the left side leading
edge. The bottom screws that continued holding the horizontal stabilizer leading
edge assembly in place were not removed. The top sets of attaching screws for
both the left and right horizontal stabilizer leading edge assemblies were not visible
from the ground.
The right leading edge assembly was removed from the horizontal
stabilizer following a shift change by third shift mechanics. A new deice boot was
bonded to the front of the leading edge at a work bench inside the hangar. During
the third shift, the accident airplane was pushed out of the hanger to make room for
work on another airplane. There was no direct light placed on the airplane as it sat
outside the hangar. Work on the horizontal stabilizer was resumed outside. The
third shift mechanics reinstalled the right side leading edge assembly. They used
26
new and used screws to attach the top and bottom of the assembly to the right
horizontal stabilizer.
The second shift work on N33701 was indicated on the second shift
inspector’s written turnover sheet; however, the incoming third shift inspector
reviewed the sheet before the entry was made. The third shift maintenance
supervisor and mechanics were not verbally informed of the removal of the upper
SCRWS on the left side leading edge. The M-602 work cards had originally been
assigned to the third shift for completion, but the second shift supervisor, who was
assigned to N33701, elected to start work on the deice boots to assist the third shift
with the workload. In addition, he did not issue the M-602 work cards to the
second shift mechanics because they were in a package assigned to the third shift.
As a result, no entries were made on the reverse sides of the M-602 work cards that
would have informed the third shift supervisor and third shift mechanics that work
had been started by the second shift on both the left and right horizontal stabilizer
deice boots.
A third shift inspector later reported that he had gained access to the
top of the horizontal stabilizer to assist with the installation and inspection of the
deice lines on the right side of the horizontal stabilizer. He stated that he was not
aware of the removal of the screws from the top of the left leading edge assembly
of the horizontal stabilizer. In the dark outside the hangar, he did not see that the
screws were missing from the top of the left side leading edge assembly for the
horizontal stabilizer.
Based on information gathered from interviews and statements, the
following significant maintenance events took place the night before the accident:
2000: The second shift supervisor, who was in charge of a “C”
check on another airplane, and another supervisor normally
assigned to the flight line but who was to supervise the
work on N33701, discussed bringing N33701 into the
hangar. [There were two supervisors on the second shift.
One supervisor was normally assigned to the flight line,
but he took charge of the maintenance on N33701. The
second supervisor was in charge of a C check on another
airplane.]
2100: The supervisor who took charge of N33701 told a second
shift mechanic to remove both deice boots from N33701.
27
2130: N33701 was brought into the hangar by the second shift
supervisor, who was responsible for the C check on
another airplane. A second shift inspector informed the
other second shift supervisor, who was now responsible for
N33701, that he would volunteer to assist mechanics with
the boot changes.
2145: A third shift flight line supervisor arrived at the hangar and
noted that the third shift hangar supervisor was already
there.
2200: The second shift supervisor responsible for N33701
observed two mechanics and the second shift inspector
kneeling on the right stabilizer removing the right boot.
The third shift hangar supervisor observed the second shift
inspector lying on the left stabilizer and observed two mechanics
removing the right deice boot.
The third shift supervisor, who was working the hangar, asked the
second shift supervisor (who was responsible for the C check on
another airplane) if work had started on the left stabilizer. The
third shift supervisor observed the supervisor look up at the tail of
N33701 and state “No.”
The third shift supervisor, who was working the hangar, told the
second shift supervisor (who was responsible for the C check on
another airplane) that he would be able to change the right deice
boot that evening, that the left deice boot change could be made on
another night, and that he would return the left replacement boot to
stock. The second shift supervisor took the right replacement boot
and placed it on a work bench.
2205: The third shift inspector arrived early for work and saw
that the majority of the right deice boot had been removed.
He reviewed the inspector’s turnover form and found no
writeup on N33701 because the second shift inspector,
who had removed the upper screws, had not yet made his
log entries.
28
2215:
A third shift mechanic clocked in and went to the break
room to chat with friends until the start of his shift at 2230.
Shift Chanq
2230: The second shift inspector, who removed the upper screws
from the leading edges of both stabilizers on N33701,
filled out the inspector’s turnover form with the entry,
“helped the mechanic remove the deice boots.” He then
clocked out, and left for home. The inspector later stated
that he placed the screws that he removed from the top row
of the left and right sides of the horizontal stabilizer in a
bag and that he left the bag on the manlift.
One of the two mechanics, who was helping with the boot change
on N33701, stopped working and returned to airplane 724 to finish
work that he had started earlier in the shift.
A third shift mechanic was informed by the third shift supervisor
that he was assigned to do the line check on N33701, and that he
needed to reposition N33701 outside the hangar. N33701 was then
moved outside the hanger.
The second shift mechanic, who had been removing the deice boot
on N33701, gave a verbal turnover to the second shift supervisor
(who was responsible for the C check on another airplane). The
mechanic was instructed by the supervisor to give his turnover to a
third shift mechanic. After giving a turnover to a third shift
mechanic, the second shift mechanic locked up his tools and
clocked out.
The third shift mechanic, who received the turnover from the
second shift mechanic, was not assigned later to N33701. He later
stated that he recalled seeing the bag of removed screws on the
manlift. The third shift mechanic gave a verbal turnover to another
third shift mechanic, who later did not recall receiving a turnover
and stated that he did not see any bagged screws.
29
Another third shift mechanic arrived at the hangar and was
informed by the third shift supervisor, who was working the
hangar, that he was assigned to N33701’s boot replacement and
that he should talk to the second shift supervisor to find out what
had been accomplished. There was no discussion regarding which
of the two second shift supervisors that the third shift mechanic
should talk to. The mechanic talked to the second shift supervisor
in charge of the C check on another airplane.
The third shift mechanic then asked the second shift supervisor
(who was responsible for the C check on another airplane) what
had been done on N33701 during the second shift. The mechanic
observed the supervisor point to the tail of N33701 and say that a
few stripped screws had prevented the second shift mechanics from
removing the right leading edge. The mechanic then asked if any
work had been performed on the left deice boot. The supervisor
informed him that he did not think he would have time to change
the left deice boot that evening.
2245: The third shift line supervisor left the hangar to work at the
gate and had no involvement with N33701.
2300: The second shift supervisor responsible for N33701 left
work about this time. He had not talked to the other
second shift supervisor, the third shift supervisor, who was
working the hangar, or the third shift supervisor in charge
of line checks before he left for home.
2330: The second shift mechanic who helped with the removal of
the right boot clocked out and left for the evening.
Subsequently, the airplane was cleared for flight. The first flight was
a passenger flight from IAH to LRD at 0700. There is no evidence from the
morning’s preflight that the flightcrew knew of any of the work performed on the
horizontal stabilizer. Moreover, the FARs and airlines did not require them to be
informed of such work.
The flight from IAH to LRD was without incident. Shortly after the
accident, a passenger, who had been on the flight from IAH to LRD, informed
Safety Board investigators that he was awakened on the flight to LRD by
I
30
vibrations that rattled his beverage can on the meal tray in front of him.
Accordingly, he asked the flight attendant if he could move to another seat. The
passenger did not inform the flight attendant or any other crewmembers about the
vibrations. Others passengers on that flight, some of whom had flown on that
model airplane previously, did not recall unusual vibrations. The accident took
place on the return trip from LRD to IAH.
1.17.4
Required Inspection Items (RIIs)
Continental Express’ GMM 1 Section 5, states that “Continental
Express has established a list of items that requires a concentrated inspection (RII)
on any work performed on those items. This list includes items that could result in
a failure or malfunction that could endanger the safe operation of the aircraft, if not
properly installed or if improper parts or materials are used.” On page 5-5,
Paragraph 2, “Designated [required inspection] Items” the item “Stabilizers” is
listed. Also, 14 CFR 135.427 states “A designation of the items of maintenance
and alteration that must be inspected (required inspections) including at least those
that could result in a failure, malfunction, or defect endangering the safe operation
of the aircraft, if not performed properly or if improper parts or materials are used.”
Continental Express’ management and quality control inspectors stated
that the removal and replacement of the horizontal stabilizer leading edge deice
boots were not RIIs. RIIs are required to be inspected by a quality assurance
inspector. However, the M-602 maintenance work order cards, used the night
before the accident to assign the work to change both the left and right horizontal
stabilizer deice boots, had the RI1 “Yes” block circled. Further, the completion of
the deice boot change, the removal of the used deice boot, and the bonding of a
new boot to the right side leading edge assembly were signed off by a quality
control inspector on the third shift. However, the inspector stated that he knew that
the boot was not an RI1 and therefore conducted only a cursory walk around the tail
without inspecting the final installation of the leading edge/deice boot.
Embraer stated that the deice boots and leading edges, as assemblies,
were RIIs and were part of the larger stabilizer assembly, listed in the FAAapproved operator’s GMM as an RII. The manufacturer noted by letter (See
appendix G) that the subject assembly met the operational requirement of the FAA
for a RII, in accordance with 14 CFR 135.427(b)(2).
31
Continental Express’ management maintained that the leading
edge/deice boot assembly was a separate assembly and that if the manufacturer or
FAA had wanted the assembly treated as an RI1 or critical item they should have
made that clear.
1.17.5
Right Elevator Replacement
The maintenance records for N33701 revealed that on April 24, 1991,
the right elevator was removed from airplane 708 because of damage from a
lightning strike. Airplane 708 was subsequently returned to service following the
installation of a replacement right elevator. The damaged elevator was repaired on
April 27, 1991, and was installed on N33701 on May 2, 1991. The elevator had
been repaired using approved technical information supplied by Embraer’s
Structural Repair Manual (SRM), section 55-20-01. The SRM referred the
mechanic to section 51-62-01 of the SRM, which contained procedures for
statically balancing the elevator, after the repair had been made. The mechanic
who balanced the elevator following its repair stated that he had read the balancing
procedures contained in the SRM.
SRM sections entitled “Control Surface Static Balancing” and
“Equipment and Consumable Material for Balancing” had complete descriptions of
control surface static balancing, a table of equipment used for control surface
balancing, a balancing stand with an adjustable support, and included the Ground
Service Equipment (GSE) Number 094 and a diagram of the necessary equipment.
The investigation revealed that the approved balancing equipment was
available but apparently misplaced and was not used for the balancing of the
elevator that was eventually installed on N33701. The mechanic stated that he
used “homemade” balancing blocks on a level table and visually confirmed the
balance of the elevator. Embraer stated that it recommends the use of the
equipment listed in the SRM for balancing control surfaces; however, in
emergency situations, jack assemblies could be used, provided that the rotational
axis of the control surface is horizontal. The FAA’s Principal Maintenance
Inspector (PMI) assigned to Embraer was asked by accident investigators if the
procedure used by the mechanic was approved by the FAA, and he replied “No.”
Embraer was asked what effects an unbalanced elevator would have
on the airplane. Embraer replied that the repair to the right elevator on the accident
airplane would “represent [a] less than 1% out of balance condition, which could
be regarded as a negligible effect.”
32
1.17.6
No. 1 Engine Overtorque
On September 25, 1990, the left engine and propeller on N33701
experienced an overtorque to 141 percent. After performing the required initial
inspection of the engine, per the Pratt & Whitney Maintenance Manual 72-00-00,
Revision 6, the airplane was issued a ferry permit to return to Houston for further
detailed inspection. As a result of the overtorque, the left propeller was changed
on September 28, 1990, per the Hamilton Standard Maintenance Manual. The
engine was inspected in accordance with Pratt & Whitney Canada Service
Information Letter PW- 123, issued on March 9, 1990. On September 28, 1990, the
airplane was returned to service.
The Pratt & Whitney maintenance manual required, in addition to the
initial inspection, the following: repetitive inspections of the chip detector/filter
element after approximately 10 hours or 1 day of operation, and thereafter at
approximately 25 hours, 50 hours, and 100 hours, respectively, with the last check
at approximately 250 hours or at the next A check. If no ferrous material was
found after these checks, the engine could remain in service without further special
maintenance action and subject to local airworthiness authority approval.
The review of the maintenance records revealed that certain
procedures recommended by the Pratt & Whitney maintenance manual were not
followed. For example, there was no record that the required repetitive chip
detector inspections were performed. Continental Express stated that it had
performed a continuity check of the chip detector circuit at every line check, which
occurs less frequently than every 2 days (about 175 times in the past. year).
Continental Express added that the line check method would have detected the
presence of metal in the detector. There were no reports of chip detector problems
during that period. Continental Express also stated that it had performed eight A
checks during the same time period, in which the engine scavenge and main filters
are checked. There was no record of metal particle contamination.
Additionally, the required engine log book entry regarding the
overtorque event was not found. Also, there was no record that the PM1 had been
requested to provide or had granted the required approval for the engine to remain
in service, although Continental Express had notified the PM1 of the event.
33
2. ANALYSIS
2.1
General
Weather was not a factor in the accident. ATC services were properly
conducted and were not a factor in the accident. The flightcrew was properly
qualified and certified to conduct the flight. The performance of the flightcrew was
not a factor in the accident. The accident was nonsurvivable because of the severe
impact forces.
The examination of the wreckage confirmed that the airplane had
experienced an in-flight fire that occurred after, not before, the in-flight breakup.
Evidence to support this conclusion includes the fire damage pattern on the
empennage pieces. The fire pattern shows that the leading edge of the horizontal
stabilizer and the upper portion of the vertical stabilizer with the upper rudder
surfaces attached separated from the airframe before the in-flight fire occurred.
Although the horizontal stabilizer experienced minor soot deposits and heat
damage, it separated from the airframe before the fire damage became more
significant. The lower portions of the vertical stabilizer and lower rudder surfaces
that remained attached to the airframe until ground impact experienced significant
fire damage. Further, the lack of fire damage on the left engine suggests that this
engine separated early in the breakup sequence when the left wing failed. The
failure of the left wing released fuel that probably led to the in-flight fire.
The passenger seat that was ejected from the cabin at ground impact
suggests that the fire did not progress into the cabin area before impact. This
conclusion is supported by the absence of soot deposits in the respiratory tract of
the occupants, and the absence of elevated carboxyhemoglobin in the tissues of the
occupants.
The FDR data and examination of the wreckage revealed that the
flight control systems, engines, and propellers were operating normally before the
extreme attitude changes of the airplane. Consequently, engine and propeller
malfunctions were not a factor in the accident.
The Safety Board’s analysis of this accident included an examination
of the circumstances that led to the loss of the left stabilizer leading edge,
including: flightcrew performance related to the accident; the maintenance and
inspection conducted by Continental Express the night before the accident; the
management of the Continental Express maintenance department; the FAA
34
approval and oversight of the Continental Express maintenance program; and the
procedures for establishing RIIs by the aircraft manufacturer, the airline, and the
FAA.
The Safety Board’s analysis also exarnined the aerodynamic and
structural failure aspects related to the dynamics of the airplane after it lost the left
stabilizer leading edge.
2.2
Aerodynamic and Structural Failure Aspects
The Safety Board believes that the airplane experienced the following
sequence of events during the final moments of flight. The airplane was
descending at 260 KIAS, which was well within its operating envelope, the wings
were level, both engines were operating normally, and the pitch attitude was
10 degrees nose down. As the airplane descended through 11,500 feet, the leading
edge of the left horizontal stabilizer separated from the airframe. The left
horizontal stabilizer leading edge was the first piece of wreckage found along the
wreckage path, preceding the next piece by almost l/2 mile. This indicates that it
was the first piece to separate from the airplane. The loss of the leading edge
exposed the front spar of the left side of the horizontal stabilizer to the airstream,
and an aerodynamic stall occurred that greatly reduced the downforce produced by
the horizontal stabilizer. The reduction in downforce created a large nose-down
pitching moment, and the airplane pitched down immediately. A peak load factor
of approximately -5 g was reached at the end of only 1 second.
The airframe remained intact (minus the leading edge), and the load
factor fluctuated around -2 g, for approximately 6 l/2 seconds. The airplane pitch
attitude decreased to 68 degrees nose down, airplane heading moved 20 degrees
nose left, and a 15 degree right roll attitude was reached at the end of this period.
The airplane’s altitude was 9,500 feet, and it was flying at an airspeed of
280 KIAS. A second peak in negative load factor was then experienced, and the
Safety Board believes that the left wing failed and the right wing tip detached at
this point.
The airplane then rolled to the right at a roll rate exceeding
160 degrees per second. The Safety Board believes that the lift produced by the
intact right wing produced the extreme roll. The high airspeed and roll rate created
large airloads on the airplane’s structure. The Safety Board believes that excessive
airloads induced by the high airspeeds and/or roll rate caused the horizontal
stabilizer and left engine to separate from the airframe. The airplane then entered a
35
spin to the right, fell uncontrollably toward impact, its pitch attitude oscillating
between approximately -40 degrees and +40 degrees.
To recreate this sequence, the Safety Board relied on the substantial
amount of evidence obtained from the wreckage, CVR, and FDR. Flight dynamics
and structural simulations by Embraer provided additional data for use in the
investigation.
The Safety Board analyzed the airloads that were applied to the
partially secured leading edge on the accident airplane. The atmosphere was calm;
therefore, gust loads were probably not a factor in the separation of the leading
edge. Aerodynamic lift and drag both produce loads on the horizontal stabilizer
structure. In general, aerodynamic loads are significantly greater at higher
airspeeds since the dynamic pressure of the airstream varies directly with the
square of airplane velocity.
Aerodynamic drag exerts a force on the airplane that is opposite to the
direction of motion and parallel to the relative wind. Therefore, aerodynamic drag
created an aft load on the horizontal stabilizer structure of the accident airplane.
This force compressed the partially secured leading edge against the front spar of
the stabilizer structure and helped to keep the leading edge in place. However,
aerodynamic lift is also an important factor in the determination of airloads acting
on the stabilizer.
The horizontal stabilizer in this case provides negative, or downward,
lift to balance the pitching moment of the wings, engines, and fuselage. Airplane
nose pitch attitude is controlled up or down by deflecting the elevator attached to
the rear of the horizontal stabilizer. The lift force required at the horizontal
stabilizer to establish trimmed flight is a function of many factors, such as the
center of gravity, engine thrust, airspeed, and airplane configuration. The airplane
is described as being “trimmed” in pitch if the sum of the pitching moments created
by these factors is equal to zero. Calculations that defined the horizontal stabilizer
lift required (downward) for the two postmaintenance flights on N33701 showed
that the peak download occurred at the time of the accident flight upset. The
calculations also showed that the maximum downforce produced during the first
flight was at least 30 percent lower than that achieved at the time of the upset.
It is apparent that the airloads did not appreciably deflect the leading
edge during the first postmaintenance flight. However, the aerodynamic download
at 260 KIAS during the accident flight probably deflected the partially secured
36
leading edge downward to the point where the combination of down and aft loads
tore the leading edge from the airplane. At the time of the upset, the airspeed was
high but within normal operating limits. The FDR data show an initial transient
small increase in load factor that is consistent with the loss of l/2 of the downward
force immediately before the large nose-down pitching occurred.
Despite the limitations of flight dynamics predictions, airplane
movements during the first 6 l/2 seconds after the upset were consistent with the
forces expected following a loss of the left side horizontal stabilizer leading edge.
The FDR recorded a nose-left heading change and lateral acceleration that are
consistent with an airplane sideslip caused by a drag imbalance between the left
(higher drag) and right sides of the airplane. The higher drag on the left side of the
airplane was most likely because of the missing leading edge of the left horizontal
stabilizer spar. The airplane also reached and maintained 10 degrees to 15 degrees
of right roll during the 6 l/2 second interval after upset. The flight dynamics
simulation produced a right roll after the upset because of the unbalanced
downforce produced by the asymmetrical horizontal stabilizer.
The evidence shows that the airframe remained intact and that the load
factor fluctuated around -2 g for about 6 l/2 seconds before a second peak in
negative load factor was evident, at which point the left wing structure failed
negatively, and the airplane rolled violently to the right at a roll rate exceeding
160 degrees per second. The roll was caused by the lift of the intact right wing.
This maneuver created additional extreme airloads on the entire airplane structure,
resulting in failure and separation of the horizontal stabilizer and left engine.
The airloads on the horizontal and vertical stabilizers prior to the wing
failure, as calculated by Embraer, did not exceed the maximum allowable. This
information is consistent with the FDR data and physical evidence that the
empennage did not fail until after the failure of the left wing. Witnesses reported
seeing the airplane in a left spin prior to ground impact. Although the FDR data
show the airplane in a right spin following wing failure, the recording ends about
13 seconds before impact.
In summary, the Safety Board concludes that the FDR data,
engineering simulation, and examination of the wreckage confirm that the accident
sequence was initiated by the loss of the left leading edge of the horizontal
stabilizer.
37
Flightcrew Preflight Performance
The Safety Board found no evidence that the two pilots were informed
of the work that had been performed on the horizontal stabilizer the night before
the accident. Of course, if the pilots had wanted to review the maintenance records
for the airplane, the records would undoubtedly have been made available to them.
However, there was no indication of any work on the stabilizer leading edges in the
pilot’s airplane log book, and no indication has been found that the flightcrew was
informed of any of this work, even though the work was on a critical assembly of
the airplane--the horizontal stabilizer leading edges.
The Safety Board is aware that the work performed on the horizontal
stabilizers was considered scheduled maintenance and was not normally noted in
the pilot’s airplane log book. Further, there are no regulatory provisions for pilots
to be made aware of routine maintenance work, regardless of its complexity.
However, the Safety Board believes that a study should be undertaken on the
feasibility of developing a means to advise flightcrews about recent maintenance
actions, both routine and nonroutine, of the airplanes they are about to fly, so that
they have the opportunity to be alert to discrepancies during preflight inspections
and possibly to make an additional inspection of critical items, such as RIIs, that
may affect the safety of flight. In this case, if the flightcrew had been informed of
the previous night’s work on the airplane, they might have, with the advantage of
morning daylight, lent a crucial hand in checking the work.
The top of the horizontal stabilizer on the airplane’s “T-tail” is about
20 feet above the ground. Therefore, the flightcrew could not have seen the area of
the missing screws on top of the leading edge/deice boot during their normal
preflight inspection. However, if they had been informed of the maintenance, they
might have discussed the work with maintenance personnel and requested them to
conduct a visual inspection of the stabilizer’s upper surface. Because the
flightcrew was unaware of the previous night’s work on the airplane, the possibility
of having another set of eyes observe the work was eliminated.
The Safety Board believes that the FAA should require airlines to
establish procedures to inform pilots of all significant maintenance on airplanes
before flight. Such information would allow pilots to be more alert to potential
unsafe conditions when they conduct preflight inspections. The redundancy
provided by such a requirement would be important for many critical maintenance
items.
38
2.4
Maintenance Factors
The evidence is clear that the events during the maintenance and
inspection of N33701 the night before the accident were directly causal to the
accident. Several errors were made by the individuals responsible for the
airworthiness of the airplane. The Safety Board believes that the reasons for the
errors and the overall failure of the maintenance program are complex and are not
simply related to a single failure by any single individual. Consequently, the
Safety Board’s analysis of the maintenance and inspection program concentrated on
the systemic reasons for the accident, as well as the specific errors made by the
individuals concerned.
The Continental Express GMM had FAA-approved procedures for
shift turnovers. These procedures included briefings by mechanics to supervisors,
briefings by outgoing supervisors to incoming supervisors, completion of
maintenance and inspection shift turnover forms (so that oncoming personnel
would be aware of incomplete work), and the documentation of incomplete work
that would be noted by the mechanic on the reverse sides of M-602 work cards. In
fact, the Safety Board found no specific deficiencies in the GMM, other than the
fact that the GMM did not delineate or identify specifically the horizontal stabilizer
leading edge deice boots as an RII. Only the major structural items were listed.
However, this deficiency alone did not cause the accident, and it is not unique to
Continental Express. This issue is discussed further in section 2.5. The Safety
Board concludes that the GMM contained clear procedures, which, if followed,
could have prevented the accident.
The Safety Board concludes that the upper row of screws that had
been removed from the leading edge of the left horizontal stabilizer was undetected
because the approved procedures in the GMM were not followed by the
maintenance, supervisory and quality control personnel directly charged with
evaluating the airworthiness of N33701 before it was returned to service. The
following are examples of substandard practices and procedures and oversights by
individuals, who had an opportunity to prevent the accident:
The second shift supervisor responsible for N33701 failed to solicit an
end-of-shift verbal report (shift turnover) from the two mechanics he assigned to
remove both horizontal stabilizer deice boots. Moreover, he failed to give a
turnover to the oncoming third shift supervisor and to complete the
39
maintenance/inspection shift turnover form. He also failed to give the M-602 work
cards to the mechanics so that they could record the work that had been started, but
not completed, by the end of their shift. The Safety Board believes that the
accident would most likely not have occurred if this supervisor had solicited a
verbal shift turnover from the two mechanics he had assigned to remove the deice
boots, had passed that information to the third shift supervisor, had completed the
maintenance shift turnover form, and had ensured that the mechanics who had
worked on the deice boots had filled out the M-602 work cards so that the third
shift supervisor could have reviewed them.
.
.
.
Second Shift Swervlsor not ReSgonslble for N337U
The other second shift supervisor, who was not responsible for
N33701 but was in charge of a C check on another airplane, assigned two
mechanics to the second shift supervisor responsible for N33701. He received a
verbal shift turnover from one of the mechanics he had assigned to the other
supervisor. However, this turnover came after he had already given a verbal shift
turnover to the oncoming third shift supervisor, informing him that no work had
been done on the left stabilizer. The Safety Board found that when he did receive
the verbal turnover from the mechanic, he failed to fill out a maintenance shift
turnover form and failed to inform the oncoming third shift supervisor. Also, he
did not direct the mechanic to give his verbal shift turnover to the second shift
supervisor who was responsible for N33701 or to the oncoming third shift
supervisor. Instead, he instructed the mechanic to seek out a third shift mechanic
and report to him the work that had been accomplished.
The Safety Board believes that because the second shift supervisor
who was not responsible for N33701 did give a verbal turnover to the oncoming
third shift supervisor, and because he did accept the verbal turnover from the
second shift mechanic, he had, in effect, assumed responsibility for N33701.
Further, if the second shift supervisor had instructed the mechanic to seek out the
second shift supervisor (responsible for N33701), who had assigned him the job, or
to seek out the oncoming third shift supervisor with his verbal shift turnover
information, and had instructed the mechanic to complete the M-602 work cards,
the accident would most likely not have occurred.
The second shift quality control inspector, who had assisted the two
mechanics with the removal of the upper screws on both horizontal stabilizers,
40
signed out on the inspectors’ turnover sheet and went home. An oncoming third
shift quality control inspector arrived at work early, reviewed the inspector’s
turnover sheet, and recalled no entry. Unfortunately, the oncoming inspector
reviewed the shift turnover sheet before the second shift inspector wrote on it
“helped mechanic pull boots.” In addition, the second shift inspector failed to give
a verbal shift turnover to the oncoming third shift inspector. The Safety Board
believes that if the second shift quality control inspector had given a verbal shift
turnover to the oncoming third shift inspector and reported any work initiated
regarding removal of the upper leading edge screws on both stabilizers, the
accident would most likely not have occurred. In addition, as an inspector, he was
a “second set of eyes,” overseeing the work of mechanics. By effectively
becoming another mechanic, he removed himself from functioning as an inspector.
The second shift inspector had reportedly demonstrated substandard
performance in the recent past for which he had been disciplined. Specifically, in
August 1991, he was given a warning because he had “missed a crack...in
inspection of engine exhaust stack.” During that same month, he was given a
second warning because he “did not finish all paperwork required...missed 15 task
cards on the accountability sheet.” These examples, and his actions the night
before the accident, suggest a pattern of substandard performance on the part of
this employee.
One of the mechanics, who had assumed responsibility for the work
accomplished on N33701 during the second shift, failed to give a verbal shift
turnover, per the airline’s GMM, to the second shift supervisor (responsible for
N33701), who had assigned him to remove the deice boots. In addition, he failed
to solicit and fill out the M-602 work cards from the second shift supervisor before
leaving at the end of his shift. The Safety Board believes that, if the mechanic had
given a verbal shift turnover to the second shift supervisor responsible for N33701
or if he had given his turnover to the oncoming third shift supervisor, who was
working the hangar, directly and solicited the M-602 work cards from the second
shift supervisor, the accident would most likely not have occurred.
In summary, the Safety Board concludes that the GMM contained
adequate instructions for performing a shift turnover. However, there was a
general lack of compliance with the procedures of the FAA-approved GMM during
the maintenance and inspection of N33701 the night before the accident, and this
lack of adherence to procedures was directly causal to the accident. Mechanics,
41
inspectors, and supervisors were involved in the events that led to this accident,
and many of them apparently participated in practices and procedures that were not
approved. This evidence indicates that the management personnel of Continental
Express failed to ensure the adherence to FAA-approved procedures in the
maintenance department, a situation that resulted in the airplane being dispatched
in an unairworthy condition. Accordingly, the Safety Board believes that there was
inadequate oversight by the airline’s supervisors and managers, as well as
insufficient surveillance by the FAA.
The Safety Board believes that the discrepancies noted during the
investigation regarding the work performed on the accident airplane’s right
elevator, and the procedures required after an engine overtorque, reflect adversely
on the quality of the Continental Express maintenance department. Although these
discrepancies are not related to the cause of the accident, and may be considered
“minor” oversights, they do suggest a lack of attention to established requirements
for performing maintenance and quality control in accordance with the GMM.
2.5
Required Inspection Items (RIIs)
It was disputed whether the maintenance on the deice boot or deice
boot/stabilizer leading edge assemblies should fall under the category of RIts or
under a less critical standard maintenance item. Continental Express management,
supervisory, and maintenance personnel stated that the leading edge/deice boot
assembly was a calendar inspection item, not yet due for inspection, not integral to
the structure of the horizontal stabilizer, and not within the requirements of RIIs as
set forth in the FARs. Furthermore, they believed that if this “non-structural”
member was so critical to flight, including its loss resulting in in-flight destruction
of the airplane, it should have been identified as an item by Embraer under the
requirements of the FARs.
Prior to the accident, it may not have been immediately obvious that if
a leading edge of the horizontal stabilizer separated from the airplane in flight, at
any airspeed, the airplane would have experienced so severe a negative pitching
moment that a breakup would occur. However, aerodynamic and structural
analyses could have predicted such an event, and the accident events prove the
critical nature of this component.
As noted in appendix G, Embraer contends that the deice boot or deice
boot/leading edge assembly was clearly part of the entire stabilizer assembly,
42
thereby falling within the requirements of the FARs and the specific definition of
“stabilizer” as an RII.
With regard to the aircraft components that should be categorized as
RIIs, 14 Code of Federal Regulations (CFR) 135.427, paragraph (b) states:
Each certificate holder shall put in its manual the programs
required by paragraph 135.425 that must be followed in performing
maintenance, preventive maintenance, and alterations of that
certificate holder’s aircraft, including airframes, aircraft engines,
propellers, rotors, appliances, emergency equipment, and parts, and
must include the following:
(2) A designation of the items of maintenance and alteration that
must be inspected (required inspections) including at least those
that could result in a failure, malfunction, or defect endangering the
safe operation of the aircraft, if not performed properly or if
improper parts of materials are used.
In the above reference, under either the category of “airframes” or
“those that could result in a failure, malfunction, or defect endangering the safe
operation of the aircraft,” the leading edge/deicer boot assemblies fall within the
category of RIIs. Furthermore, the operator’s GMM 1, section 5, paragraph E,
identified “stabilizer” as an RII.
The Safety Board believes that the Continental Express maintenance
and quality assurance personnel erred in not considering the removal and
replacement of the horizontal stabilizer leading edge deice boot as an RII. The
Safety Board is aware that the deice boot in and of itself is bonded to the leading
edge of the horizontal stabilizer and by itself would not constitute an RII.
However, because the leading edge of the stabilizer must be removed to remove
and replace the deice boot, the Safety Board concludes that the process of changing
the horizontal stabilizer deice boots should have been designated an RI1 so that
there could have been a more rigorous treatment of this component during
maintenance.
In view of the confusion that existed in this case, and based on the
potential for similar confusion by airlines in designating RIIs, the Safety Board
believes that the FAA should conduct a thorough review of the regulations, policies
43
and practices for establishing RIIs. Such a review should include manufacturers
and airlines in order to develop more specific requirements.
2.6
Senior Management
A major concern in this case is whether the problems noted
represented aberrations related to individual maintenance personnel (there were
several) or rather reflected systemic issues related to company policy. The
influence of senior managers is often less tangible than that of line employees.
However, the effects of management policy can be profound, and pervasive,
affecting the company at all levels. For accident prevention purposes, it is
important to determine at what level of the company structure--from the hangar
floor to the highest executive--that attention should be focused to correct the
problems that were discovered in this investigation.
The Safety Board does not believe that the maintenance issues were
related solely to the actions of individual employees who were in the hangar the
night before the accident. There was no indication of drug problems, unusual
background, or behavioral issues related to individuals. The failure to follow
proper turnover procedures--the most dramatic failure in the accident--involved
mechanics, supervisors, and inspectors from two shifts and noncompliance with
GMM procedures. Other problems noted include the definition of work on the
horizontal stabilizer leading edge as a non-RII, and the failure to follow
manufacturer-published procedures for an elevator balance and an engine
overtorque event not associated with the accident. These items suggest a general
disregard for following established procedures on the part of maintenance
department personnel.
Two safety specialists at the Boeing Commercial Airplane Company
have recently reported on a survey that examined air carrier policies and their
relation to accident history.4 A small group of operators of Boeing aircraft that
displayed exceptional safety records over a lo-year period was interviewed. This
survey was conducted to obtain information on safety techniques that could be
brought to the attention of all operators of Boeing aircraft. They found that:
These operators characterize safety as beginning at the top of the
organization with a strong emphasis on safety and this permeates
the entire operation. Flight operations and training managers
4Lautman, L.G., and Gallimore, P. L., “Control of Crew-caused Accidents,”
Flight Safety Digest, Flight Safety Foundation, October 1989.
44
recognize their responsibility to flight safety and are dedicated to
creating and enforcing safety-oriented policies. The presence or
absence of a safety organization did not alter the total involvement
of these managers. However, a majority of the operators did
maintain an identifiable flight safety focal point. There is an acute
awareness of the factors that result in accidents, and management
reviews accidents and incidents in their own airline and in other
airlines and alters their policies and procedures to best guard
against recurrence.... This management attitude, while somewhat
difficult to describe, is a dynamic force that sets the stage for
standardization and discipline in the cockpit brought about and
reinforced by a training program oriented to safety issues.
Several research papers have recently examined the activities of upper
management that can predispose an organization to having accidents.5 They
concluded that such activities need to be addressed for meaningful accident
investigation and prevention. In this accident, the Safety Board was confronted
with a situation in which established company procedures were not being followed
by personnel in the hangar. Inspectors, who were responsible for assuring the
quality of work in accordance with established procedures, were among the worst
offenders. The Safety Board concludes that if Continental Express had had an
effective quality assurance program, the company would have detected the
procedural deficiencies noted during this investigation. The investigation revealed
that the maintenance department personnel were generally aware of the correct
procedures. Consequently, the lax attitude of personnel in the hangar suggests that
management did not establish an effective safety orientation for its employees. In
fact, the failure of management to ensure compliance with air carrier policy must
be considered a factor in the cause of the accident.
2.7
Regulatory Oversight
FAA oversight of the airline failed to fiid safety problems, such as
those found during the Safety Boards investigation. This oversight included
routine monitoring by a principal maintenance inspector (PMI) and a special
National Aviation Safety Inspection Program (NASIP) team inspection following
the accident.
5Maurino, D., “Corporate Culture Imposes Significant Influence on Safety,”
International Civil Aviation Organization Journal, April 1992.
45
In the case of the routine inspection, the former PM1 indicated that he
was subjected to a tremendous workload that limited the effectiveness of his safety
monitoring. During the time he served as PMI, from February 1989 to June 199 1,
Continental Express expanded significantly. For example, it began as Britt
Airways, Inc., with a fleet of about 45 airplanes, merged with Rocky Mountain
Airways (1989) and acquired major assets of Bar Harbor Airways (1990). At the
time of the accident, the company operated a fleet of 101 airplanes (44 Part 121
and 57 Part 135), which the PM1 characterized as the largest number of airplanes
on a single commuter Air Carrier Certificate in the United States. The former PM1
indicated that he reviewed and approved four different GMMs during this
expansion period, including an 1%volume GMM used at the end of his tenure. He
stated that he operated for about 1 year as the sole inspector at the airline’s Houston
headquarters, during which time he had additional certificate responsibilities. He
was later provided an assistant (whom he trained), and his other certificate
responsibilities were removed. The entry of the airline into bankruptcy protection,
however, required additional surveillance, and no additional personnel were
provided to assist him. He indicated that the workload considerably limited his
time for on-site inspection. He stated that he could keep up with the number of
required inspections but that the depth and quality of these inspections were limited
by a lack of time.
The PMI, who assumed responsibilities one week before the accident,
characterized his workload as “extremely full.” He stated that he worked evenings
and weekends to fulfill all his responsibilities. Maintenance personnel at
Continental Express indicated that they saw FAA personnel in the hangar
infrequently, providing estimates of “perhaps a couple times per month at
maximum...once every 2 months...every 2 or 3 months...once every 3 months,
and...the last visit might have been 6 or 7 months before.” A supervisor on the
second shift said that FAA visits were always announced with usually 1 day’s
notice in advance.
It is clear to the Safety Board that the PMI’s limited visits to the
hangar floor would make observations of deviations from GMM procedures
difficult, forcing the PM1 to rely exclusively on paperwork records that might not
have reflected actual conditions. In this accident, the mechanics failed to provide a
written indication of a turnover on the M-602 work order cards, an oversight that
was a major factor in the accident sequence. However, after the work was
completed and signed off, any future inspector would have missed this fact.
46
Shortly after the accident, a NASIP team completed an inspection of
the Continental Express maintenance program. A letter of November 18, 1991, to
the airline management from then FAA Administrator James B. Busey stated,
“During our inspection, the team favorably noted that Britt Airways [doing
business as Continental Express] has implemented an internal evaluation program.
The inspection revealed very few safety deficiencies, a fact we attribute, in part, to
the success of your internal evaluation system.”
The Safety Board is concerned that the limited scope of the NASIP
inspection might have failed to uncover areas relevant to the accident. For
example, the NASIP inspection did not fiid deficiencies in shift turnover
procedures. It is known that after the accident Continental Express took some
action to ensure compliance with the procedures required in the GMM. However,
the Safety Board believes that a thorough review of previous shift turnover records
might have revealed some paperwork deficiencies. An inspection for both the
completion of the proper paperwork, and following the paperwork trail for
randomly selected open items, from inception to completion, as well as a hands-on
inspection of aircraft and an observation of work performance and turnover
procedures during all shifts, might have deepened the level of observation during
the postaccident NASIP inspection.
In summary, the Safety Board concludes that FAA surveillance of
Continental Express was inadequate because it failed to identify and correct
deficient management actions and oversight of the airline’s maintenance
department, as well as to identify practices in the maintenance program that were
contrary to the GMM.
As the result of information obtained during the investigation about
the adequacy of maintenance practices at Continental Express, on February 28,
1992, the Safety Board issued two safety recommendations to the FAA to:
A-92-6
Enhance flight standards surveillance of Continental Express, to
include sufficient direct observation of actual maintenance shop
practices, to ensure that such practices conform to the Continental
Express General Maintenance Manual and applicable Federal
Aviation Regulations.
47
A-92-7
Enhance flight standards Program Guidelines, including the
National Aviation Safety Inspection Program, to emphasize handson inspection of equipment and procedures, unannounced spot
inspections, and the observation of quality assurance and internal
audit functions, in order to evaluate the effectiveness of air carrier
maintenance programs related to aircraft condition, the adherence
to approved and prescribed procedures, and the ability of air
carriers to identify and correct problems from within.
The FAA responded to these two safety recommendations in a letter
dated May 15, 1992, indicating its agreement with the needs expressed in the
recommendations. The Safety Board’s response to this letter, as well as to other
FAA letters that address open safety recommendations about FAA surveillance of
air carrier operations and maintenance practices, is attached as Appendix I.
48
3. CONCLUSIONS
3.1
Findings
1.
All crewmembers and air traffic controllers were properly
certified to perform their duties.
2.
There was no evidence of flightcrew activities during the
preflight inspection or during the accident flight that were
causal to this accident.
3.
There was no evidence of air traffic controller activity that was
causal to this accident .
4.
Weather was not a factor in the accident.
5.
There was no evidence of engine or flight control malfunctions.
6.
The accident was precipitated by the loss of the left horizontal
stabilizer leading edge when the airplane was in a descent
12 knots below its maximum safe operating speed, within its
operating envelope.
7.
The airplane pitched severely nose down upon the loss of the
left horizontal stabilizer leading edge, and the wings stalled
negatively.
8.
The violent motion of the airplane and the extreme airloads that
resulted from the loss of the left horizontal stabilizer leading
edge caused the airplane to break up in flight.
9.
An in flight fire occurred during the structural breakup.
10.
The left horizontal stabilizer leading edge separated from the
airplane because the upper row of screw fasteners (47) was not
in place. The airloads during the descent caused the surface to
bend downward and separate.
11.
The upper row of fasteners for the left horizontal stabilizer
leading edge had been removed during scheduled maintenance
49
the night before the accident, and a breakdown in procedures
failed to detect that the work was incomplete.
12.
The Continental Express FAA-approved General Maintenance
Manual (GMM) contained adequate procedures for
maintenance and quality control.
13.
There was a lack of compliance with the GMM procedures by
the mechanics, inspectors, and supervisors responsible for
ensuring the airworthiness of N33701 the night before the
accident.
14.
The lack of compliance with the GMM procedures by the
Continental Express maintenance department led to the return
of an unairworthy airplane to scheduled passenger service.
15.
The replacement of the horizontal stabilizer deice boots, which
required removal of the leading edges, should have been treated
as a required inspection item (RII). This would have required
the proper quality control of work performed on this critical
aerodynamic surface.
16.
Continental Express failed to follow established requirements
for performing maintenance during repair of the right elevator
and following an engine overtorque on N33701, although these
oversights were not causal to the accident.
17.
The deficiencies noted in the maintenance department at
Continental Express indicate that the airline’s management did
not instill an adequate safety orientation in its maintenance
personnel by emphasizing the importance of adhering to
procedures.
18.
The routine surveillance of the Continental Express
maintenance department by the FAA was inadequate and did
not detect deficiencies, such as those that led to the accident
involving N33701.
19.
The accident was nonsurvivable.
50
3.2
Probable Cause
The National Transportation Safety Board determines that the
probable cause of this accident was the failure of Continental Express maintenance
and inspection personnel to adhere to proper maintenance and quality assurance
procedures for the airplane’s horizontal stabilizer deice boots that led to the sudden
in-flight loss of the partially secured left horizontal stabilizer leading edge and the
immediate severe nose-down pitchover and breakup of the airplane. Contributing
to the cause of the accident was the failure of the Continental Express management
to ensure compliance with the approved maintenance procedures, and the failure of
FAA surveillance to detect and verify compliance with approved procedures.
4. RECOMMENDATIONS
As a result of its investigation of this accident, the National
Transportation Safety Board makes the following recommendations to the Federal
Aviation Administration:
In cooperation with aircraft manufacturers and airlines, conduct a
review of the regulations, policies, and practices related to
establishing required inspection items (RIIs) for airline
maintenance departments with the view toward developing more
specific identification of RIIs. (Class II, Priority Action) (A-92-79)
Require that airlines operating under 14 CFR Parts 135 and 121
study the feasibility of developing a means to advise flightcrews
about recent maintenance, both routine and nonroutine, on the
airplanes that they are about to fly, so that they have the
opportunity to be alert to discrepancies during preflight inspections
and possibly to make an additional inspection of critical items,
such as required inspection items (RIIs), that may affect the safety
of flight. (Class II, Priority Action) (A-92-80)
Also, as a result of this investigation, on February 28, 1992, the Safety
Board issued two safety recommendations to the FAA to:
A-92-6
Enhance flight standards surveillance of Continental Express, to
include sufficient direct observation of actual maintenance shop
51
practices, to ensure that such practices conform to the Continental
Express General Maintenance Manual and applicable Federal
Aviation Regulations.
b-92-7
Enhance flight standards Program Guidelines, including the
National Aviation Safety Inspection Program, to emphasize handson inspection of equipment and procedures, unannounced spot
inspections, and the observation of quality assurance and internal
audit functions, in order to evaluate the effectiveness of air carrier
maintenance programs related to aircraft condition, the adherence
to approved and prescribed procedures, and the ability of air
carriers to identify and correct problems from within.
The FAA responded to these two recommendations in a letter dated
May 15, 1992. The Safety Board’s response to that letter, and to other letters from
the FAA about open safety recommendations on FAA surveillance of air carrier
operations and maintenance practices, is attached as Appendix I.
BY THE NATIONAL TRANSPORTATION SAFETY BOARD
.
Coughlln
Vice Chairman
K. Lauber
Member
her A. m
Member
.
John I-Iamnvschmldt
Member
Chairman Vogt did not participate.
John K. Lauber, Member, filed the following dissenting statement:
52
I am perplexed by the majority decision that the actions of Continental
Express senior management were not causal in this accident. The report identifies
“substandard practices and procedures and oversights” by numerous individuals
each of whom could have prevented the accident. Included are mechanics, quality
assurance inspectors, and supervisors, all of whom demonstrated a “general lack of
compliance” with the approved procedures. Departures from approved procedures
included failures to solicit and give proper shift-change turnover reports, failures to
use maintenance work cards as approved, failures to complete required
maintenance/inspection shift turnover forms, and a breach in the integrity of the
quality control function by virtue of an inspector serving as a mechanic’s assistant
during the early stages of the repair work performed on the accident aircraft.
Furthermore, Safety Board investigators discovered two previous
maintenance actions taken on the accident aircraft, each of which departed from the
approved procedures, and each of which involved employees different from those
engaged in the deicing boot replacement. The first event was the replacement of
an elevator without use of manufacturer-specified and required balancing tools.
The second was a failure to follow specified procedures and logging requirements
in response to an engine overtorque. Although these events were in no way related
to the accident, the report indicates that they “suggest a lack of attention to
established requirements for performing maintenance and quality control inaccordance with the GMM.” That these were the only other instances noted in this
investigation cannot be taken to mean that these were the only such instances
extant--the Safety Board’s investigation of maintenance records was curtailed, as I
understand it, to accommodate the needs of the FAA’s NASIP team, and thus, this
record is not complete.
Another factor to be considered here was the failure of Continental
Express maintenance and quality assurance personnel to treat the deicing boot
replacement, which requires removal of the leading edge of the horizontal
stabilizer, as a Required Inspection Item (RII). By doing so, a separate inspection
by quality control inspectors would have been required of the work performed that
night. Even though regulations clearly establish that the horizontal stabilizer is an
RII, Continental Express maintains that the deicer boot/leading edge assembly was
a “non-structural” item, and therefore not subject to the more rigorous inspection
requirements. I find it very disturbing that senior personnel responsible for aircraft
maintenance apparently do not understand that the leading edge of any airfoil is a
critical determinant of the aerodynamic characteristics of that airfoil, and thus that
improper repair work could seriously compromise the safety of an aircraft.
53
Still another factor that I believe to be highly relevant here was the
absence of a Lead Mechanic and a Lead Inspector as specified in the GMM.
Senior management’s failure to fill these positions in effect diffused and diluted the
chain of authority and accountability among maintenance and inspection personnel
at Continental Express. A detailed examination of the organizational aspects of the
maintenance activities the night before the accident reveals a melange of crossed
lines of supervision, communications and control. This situation, more than any
other single factor, was directly causal to this accident.
The multitude of lapses and failures committed by many employees of
Continental Express discovered in this investigation is not consistent with the
notion that the accident resulted from isolated, as opposed to systemic, factors. It
is clear based on this record alone, that the series of failures which led directly to
the accident were not the result of an aberration, but rather resulted from the
normal, accepted way of doing business at Continental Express. The conclusions
in our report note the “failure of management to ensure compliance with air carrier
policy” and its failure to “establish an effective safety orientation for its
employees.” Line management of an airline has the regulatory responsibility for
not only providing an adequate maintenance plan (and we conclude that the GMM
was, in most respects, an adequate plan) but for implementing the provisions of
that plan as well. By permitting, whether implicitly or explicitly, such deviations
to occur on a continuing basis, senior management created a work environment in
which a string of failures, such as occurred the night before the accident, became
probable. Accordingly, their role must be considered causal in this accident.
Finally, I note for the record my concerns about the way certain
factual background information regarding senior management personnel has been
handled in this report. As discussed in our Board meeting, but not in the report,
two senior managers at Continental Express previously held positions of key
responsibility at two other airlines, one airline of which was the subject of both
civil and criminal litigation for maintenance-related practices, and the other airline
of which experienced a major accident which this Board determined to be, in part,
due to failures and deficiencies in that airline’s maintenance program and in the
management thereof. Both people were in line management positions within their
maintenance organizations during the time of the deficient practices, all of which
involved deviation of actual practices from those specified in relevant, official, and
approved documents. I am in no better position than anyone else to determine how
directly
H o relevant
w eto the
v present
e raccident
,
this
s information
i n c is. e
i t
i s
factual information of the kind we routinely collect in any accident investigation,
and is already in the public record, and since it is clearly not inconsistent with the
54
management practices noted in this investigation, I believe it is relevant to this
discussion, and thus deserves explicit mention here. To do otherwise is to make a
de facto decision that this information is clearly not relevant, a decision which I am
unwilling to support.
I believe the probable cause should read as follows:
The National Transportation Safety Board determines that the
probable causes of this accident were (1) the failure of Continental Express
management to establish a corporate culture which encouraged and enforced
adherence to approved maintenance and quality assurance procedures, and (2) the
consequent string of failures by Continental Express maintenance and inspection
personnel to follow approved procedures for the replacement of the horizontal
stabilizer deice boots. Contributing to the accident was the inadequate surveillance
by the FAA of the Continental Express maintenance and quality assurance
programs.
Member
July 21,1992
55
5. APPENDIXES
APPENDIX A
INVESTIGATION AND HEARING
1.
Investigation
The Safety Board was initially notified of this accident about 1230
eastern daylight time. An investigation team was dispatched from Washington,
D.C., and it arrived in Houston, Texas, about 1930 central daylight time.
Investigative groups were established for airplane performance, structures,
systems, human performance, air traffic control, cockpit voice recorder, flight data
recorder, fire, maintenance records, operations, survival factors, and witnesses.
Former Chairman Kolstad was the Safety Board Member who accompanied the
investigative team.
Parties to the investigation included the Bureau of Alcohol, Tobacco
and Firearms, Continental Express, Embraer Aircraft Corporation, the Federal
Aviation Administration, the Federal Bureau of Investigation, Hamilton Standard,
and Pratt & Whitney. The Government of Brazil was an accredited representative
to the investigation.
2.
Public Hearing
A public hearing was not conducted.
56
APPENDIX B
PERSONNEL INFORMATION
The Captain
The captain, Bradley Max Patridge, age 29, was born on April 20,
1962. He was hired by Continental Express Airlines, on October 10, 1987. He
held airline transport pilot certificate No. 565336474, with ratings for the EMB-120
and Airplane Multiengine Land. His most recent FAA first-class medical
certificate was issued on July 18, 1991, with the limitation: “Holder shall wear
correcting lenses while exercising the privileges of his airman certificate.”
Company records indicate that at the time of the accident the captain had
accumulated approximately 4,243 total flying hours, of which 2,468 were in the
EMB-120.
The captain received his initial ground school and proficiency check
in the EMB-120 as a first officer, completing the training on October 29, 1988. He
completed upgrade ground school training on September 21, 1989, and received a
type rating in the EMB-120 on September 29, 1989. He completed his initial
operating experience and received a line check on October 2, 1989. His last
proficiency check was on March 9, 1991. His last recurrent training was
completed on May 29, 1991, and his last line check was accomplished on
August 8,1991.
The First Officer
The first officer, Clinton Michael Rodosovich, age 43, was born on
November 9, 1947. He was hired by Continental Express Airlines on March 12,
1990. He held airline transport pilot certificate No. 1963386, with ratings for the
EMB-120 and Airplane Multiengine Land. His most recent FAA first-class
medical certificate was issued on August 30, 1991, with no limitations. Company
records indicate that, at the time of the accident, the first officer had accumulated
approximately 11,543 total flying hours, of which 1,066 hours were in the
EMB-120.
The first officer completed initial ground school in the EMB-120 on
March 30, 1990. He completed flight training on April 19, 1990. His initial
operating experience and line check were completed on April 24, 1990. He was
subsequently upgraded to captain on the. EMB-120, completing that training and
57
initial operating experience on May 14, 1990. Although he no longer held a
regular captain’s bid number, the airline allowed the first officer to retain his
currency as a captain.
In accordance with company procedures, the first officer received two
training periods, and was given a recheck on April 19, 1991. Both the retraining
check airman and the check airman who administered the April 19 recheck stated
to Safety Board investigators that there had been no problems in the retraining and
recheck.
The April 19, 1991, recheck, in which he gained an “up,” was the last
proficiency check the first officer received. His last recurrent training was
completed on May 17, 1991. His last line check was accomplished on June 4,
1991.
President Kolski
Stephen J. Kolski, age 51, was hired in July 1990, as President of the
Commuter Division of Continental Airlines, Inc., and served as President of
Continental Express. He had worked previously for Eastern Airlines (owned by
the same parent company as Continental and Continental Express) from January
1987 to July 1990, in several successive positions: Staff Vice President and
Counsel for Regulatory Compliance; Vice President for Base Maintenance; Special
Assignment; and Vice President for Administration. Prior to employment with
Eastern Airlines, he had worked for New York Air (1980-1986) and had served as
its Vice President for Operations. Kolski holds a commercial pilot certificate with
ratings and limitations of airplane single engine land with instrument privileges.
He also holds a private pilot certificate with ratings and limitations of airplane
multiengine land.
Director of Maintenance Wade
Senior Director of Maintenance and Engineering John Wade, age 48,
was hired in August 1990. He had worked previously for Eastern Airlines from
September 1989 to August 1990, as Manager of Special Projects. Prior to
employment with Eastern Airlines, he had worked for Aloha Airlines from June
1987 to June 1989, first as Director of Quality Control and then as Director of
Maintenance. He received an Airframe and Powerplant License on April 10, 1968.
He holds the Airframe and Powerplant certificate issued on April 10, 1968.
58
Director of Quality Assurance Fox
Senior Director of Quality Assurance and Control Ray Fox, age 46,
was hired in February 199 1. He had worked previously for Eastern Airlines from
1969 to 199 1 and served as Manager of Aircraft Inspection. He holds the Airframe
and Powerplant certificate issued on June 26, 1979.
Supervisor Massey
Second Shift Supervisor Charles Massey, 28, was hired by the airline
on April 9, 1988, as a mechanic. He was promoted to shift supervisor on January
19, 1990. His previous employment included service with the U. S. Army from
1982 to 1985. He holds Airframe and Powerplant certificate number 383749034
issued on December 19 1987.
Inspector Erlanson
Second Shift Inspector Wayne Erlanson, 25, was hired on July 11,
1989, as a mechanic. He was promoted to inspector on October 24, 1990. His
previous employment included service as an aircraft electrician in the U. S. Navy.
He holds Airframe and Powerplant certificate number 456456725 issued on
February 5, 1989. Erlanson received company discipline on three occasions. In
December 1990, he received a warning for having “written derogatory remarks to
mechanics in posted notes.” In August 1991, he received a warning because he
“missed a crack...in inspection of engine exhaust stack.” He received a second
warning that month because he “did not finish all paperwork required...missed 15
task cards on the accountability sheet.‘,’
Mechanic Beck
Second shift mechanic Robert Beck, 43, was hired on July 2, 1990, as
a mechanic. His previous employment included work as an aircraft mechanic with
Continental Air Micronesia (1989-1990), and flight line mechanic and inspection
dock chief with the U. S. Air Force (1986-1989). Beck holds Airframe and
Powerplant certificate number 451760789 issued on March 7,199O.
Supervisor Larivee
Second shift supervisor Sean Larivee, 29, was hired on October 25,
1987, as a mechanic. He was promoted to inspector in 1989 and to shift supervisor
on January 19, 1990. His previous employment included work as an airplane
59
mechanic for two fixed-based operators (1987) while he completed school.
Larivee holds Airframe and Powerplant certificate number 451396613 issued on
January 26,1988.
Supervisor Denham
Third shift supervisor Allen Denham, 26, was hired by Britt Airways,
Inc., (later merged into Continental Express) on June 6, 1987, as a mechanic at the
Cleveland base. He was promoted to an inspector on November 27, 1989,
transferred to the Houston base as a mechanic on March 16, 1990, and was
promoted to shift supervisor on August 17, 1990. His previous employment
included work as a helicopter mechanic and crew chief in the U. S. Army
(1984-86), and as a jet engine mechanic in the U. S. Air Force Reserves (1986-87).
He holds Airframe and Powerplant certificate number 312767386 issued on
June 19,1989.
Inspector Snyder
Third shift inspector Karl Snyder, 36, was hired by Britt Airways,
Inc., (later merged into Continental Express) as a maintenance helper at the
Bloomington, Indiana, base on September 1, 1982. He was promoted to aircraft
mechanic in 1986, and spent 9 months in 1989 at the Houston base during which
he was promoted to inspector. He returned to the Houston base as an inspector on
May 1, 1991. He holds Airframe and Powerplant certificate 347508432 issued on
April 26,1986.
60
APPENDIX C
SUMMARY OF MAINTENANCE PERSONNEL INTERVIEWS
During the interviews with maintenance personnel, it was stated the
accident airplane (N33701) arrived at the Continental Express maintenance hangar
at IAH, during the evening shift or second shift on September 10, 199 1, between
2000 and 2100. One of the second shift supervisors (Sl) stated that when the
airplane came into the hangar he assigned two maintenance personnel (Ml and
M2) to assist another evening shift supervisor (S2) with the replacement of the
deice boots that were bonded to the leading edge assemblies for the left and right
sides of the horizontal stabilizer. Sl stated that S2 was in charge of the
maintenance on N33701. S 1 stated that after he assigned the two mechanics to
assist S2, he went back to work on a “C” check that he had been working on for
another airplane.
S 1 stated that about 2215 to 2230, on the evening of September 10,
1991, he walked back to N33701 and saw that the mechanics had removed the
right leading edge/deice boot assembly. He said that a midnight shift (third shift)
supervisor, S3, walked into the hangar and asked what work had been
accomplished on N33701. Sl informed S3 that two mechanics were in the process
of removing the horizontal stabilizer leading edge/deice boots. Sl stated that S3
was concerned about whether his shift could do the left and right boots that
evening. Sl stated that he asked both mechanics how far they had progressed on
the removal of the deice boots. Sl stated that the mechanics told him that they had
removed the majority of the screws attaching the leading edge to the right side of
the horizontal stabilizer. Some were stripped out and needed to be drilled out. Sl
stated that he instructed both mechanics not to remove the left side stabilizer boot.
He further stated that they said nothing about performing any work on the left side
of the stabilizer. Sl stated that about this time a third shift mechanic (M3) came to
him and asked the status of the deice boots. Sl said that he told him that one boot
had been taken off of the stabilizer and the screws were out, except for a few
screws that were stripped. Sl said that M3 then went over to mechanics Ml and
M2 to talk to them further about the work which needed to be done. Sl then stated
that he went back to the “C” check inspection.
S2 stated that he assigned both mechanics to begin removing the
leading edge/deice boot assemblies from both the left and right sides of the
horizontal stabilizer. He said that at some time after 2200, he walked out to
N33701 and saw one of the three second shift quality control inspectors (11)
61
kneeling on the right side of the horizontal stabilizer. However, he could not
determine exactly what he was doing. S2 further stated that he did not receive a
verbal status report on N33701 from the second shift mechanics before they left for
the night.
Ml stated that 11 volunteered to assist Ml and M2 during the second
shift by removing the upper screws attaching the leading edge/deice boot
assemblies to the left and right side of the horizontal stabilizer, while M2 assisted
Ml in removing the lower screws attaching the leading edge/deice boot assembly
to the right side of the horizontal stabilizer. Ml stated that at the end of his shift,
about 2230, he advised Sl of the work that had been accomplished on the
horizontal stabilizers. However, when asked during the Safety Board’s interview,
Ml could not specifically recall whether or not he had informed S 1 that the upper
screws attaching the leading edge/deice boot assembly to the left side of the
horizontal stabilizer had been removed. However, he felt sure that he had. S 1 then
instructed Ml to report the work that he had accomplished to a third shift person
who would be working on N33701. Ml stated that he turned over the job to a third
shift maintenance person (M4), and informed him that most of the screws had been
removed from the leading edge of the horizontal stabilizers (that is, both sides of
the horizontal stabilizer), and those that remained needed to be drilled out. Ml
further stated that he told M4 that only the top screws were removed from the left
leading edge. Ml stated that he then handed M4 the bagged screws that had been
removed during the work. Ml also stated that he did not fill out the backs of the
two work order cards (M-602 cards) to indicate the work accomplished because S2
did not give the cards to him.
S2, when asked why he did not give the M-602 cards to Ml after
assigning the work, stated that given the limited time left on his (evening) shift to
work on N33701, he did not think it was important for the mechanics to have the
M-602 cards. S2 said that normally all routine and watch list maintenance was
performed on the third shift.
M2 stated that he was originally working on another airplane when he
was requested by S2 to help Ml with the removal of the deice boots from N33701.
He stated that he assisted Ml in removing the leading edge/deice boot assembly for
the right side of the horizontal stabilizer and that about 2230 he went back to work
on another airplane. When asked, M2 stated that he did not make any maintenance
entries on the reverse sides of the M-602 cards and that he did not give a verbal
shift turnover to his shift supervisor after helping on N33701.
62
11 stated that he was one of three quality control inspectors assigned
to the second or evening shift. He said that he overheard S2 telling two mechanics
to start removing both horizontal stabilizer leading edges for boot changes. He
said that he had completed his work assignments with 1 hour remaining until his
shift was over. He stated that he informed S2 that he would volunteer to help the
mechanics remove the boots. Since he was the smallest person, he got on top of
the horizontal stabilizers and removed the top screws from the leading edges for
both sides. 11 stated that by the time Ml and M2 had removed the deice boot from
the right side leading edge, he had removed all of the top screws for both sides and
had put the screws into a parts bag, with the exception of about five unserviceable
screws that he discarded. 11 stated further that he left the bag with the screws on
the “man-lift.” At this time the shift changed and he and the two mechanics
gathered their tools and descended to the floor via the man-lift or workstand that
they used to gain access to the horizontal stabilizer. 11 stated that he did not speak
to any third shift mechanics but he did write his part of the inspection department’s
written report, to account for his time. In. that report, he stated that he wrote,
“Assisted mechanics with removal of deice boots.” I1 stated that he left for home
about 2230.
M4, a third shift mechanic, stated that he received a verbal briefing
from Ml that both horizontal stabilizer deice boots needed to be changed on
N33701. He was told that the screws had been removed and that the remaining
screws were stripped. When asked if the attaching screws were removed from the
leading edge of the left side of the horizontal stabilizer, M4 stated that he was not
sure. After receiving the verbal shift turnover from Ml, M4 moved a man-lift to
the left side of N33701. After learning that he was not assigned to work on
N33701, M4 informed a third shift mechanic, M5, that screws were removed and
that the horizontal stabilizer deice boots needed to be replaced. M4 stated that he
saw a bag of screws on the floor of the man-lift. He gave them to M5.
M5 stated that he did not recall receiving a verbal shift turnover from
M4, and was not aware of any previous work performed on the leading edge of the
right side of the horizontal stabilizer. However, subsequent to his interview, M5
submitted a written statement claiming that upon reporting to work, he received his
assignment for the night from S3, one of the third shift supervisors. The
assignment was to help M3 with N33701’s stabilizer deice boot replacement. M5
stated that when he went to N33701, he observed that the deice boot had been
stripped from the leading edge of the right side of the horizontal stabilizer. The
next step was to remove the right side leading edge. After the leading edge had
been removed, a new deice boot was bonded in place. M5 stated that between
0330 and 0430, he installed the leading edge/deice boot assembly on the right side
63
of the horizontal stabilizer, with the help of 12. M5 stated that he used the screws
that were on the man-lift to attach the right side leading edge assembly to the
horizontal stabilizer. He also used new screws he had obtained from stock to
replace those he had previously drilled out. He said that although there were
approximately a dozen screws left over on the man-lift, he did not think this was
significant because of the number of screws he had to replace due to drilling out
and corrosion of the old screws. M5 further stated that after completing the
installation, M3 went into the office to complete the necessary paperwork with I2
and M6, and that he started the job cleanup around the table they had used to bond
the right deice boot to the leading edge. He was also informed that the removal
and installation of the leading edge for the left side of the horizontal stabilizer had
been deferred. He said that since he had received no information that work on the
deice boot on the left side of the horizontal stabilizer had been started, he and M3
had agreed that they would not begin to remove the deice boot on the left side of
the horizontal stabilizer until they completed the work on the right side deice boot.
The two third shift maintenance supervisors, S3 and S4, relieved the
second shift. S3 stated that he arrived at the maintenance hangar about 2200. He
saw that N33701 was in the hangar and that Ml and another mechanic were
peeling off the deice boot from the leading edge assembly for the right side of the
horizontal stabilizer. S3 stated the 11 was lying on the left side of the horizontal
stabilizer, watching them. S3 then asked the second shift supervisor, Sl, if they
had started taking off the left horizontal stabilizer deice boot yet. S3 said that the
Sl looked up at the airplane’s tail and replied, “No.” S3 stated the he knew that his
shift would not have time to replace both boots because he had an airplane coming
in for an “A” check, and that an airplane was needed for the morning launch at
0700. S3 said that since the replacement of these boots was a procedure in
preparation for winter operations, it &as appropriate to postpone changing the left
boot to a later time. He stated that he had told Sl that he was putting the
replacement left side boot back in stock for another night. He then informed Sl
that N33701 would have to come out of the hangar to make room for the airplane
that was coming in for an “A” check. S3 then assigned M3 and M5 to the right
side deice boot replacement. M6 was assigned to the line check. S3 said that after
the deice boot for the right side was replaced and the right leading edge was
reinstalled, he gave M3 the M-602 card to sign off the work on the boot change for
the right side of the horizontal stabilizer. S3 stated that he then looked on the
reverse side of the M-602 card for the replacement for the left side deice boot, and
that because the M-602 card for that assembly did not indicate any work had been
performed, he sent the airplane to the gate for a 0700 departure.
64
APPENDIX D
AIRPLANE INFORMATION
The airplane, U.S. registration N33701, was an Embraer EMB-120,
manufactured in Brazil. The serial number was 120-L77. Continental Express
Airlines acquired the airplane on April 15,1988.
The gross takeoff weight for the airplane, upon departure from LRD
on the accident flight, was calculated by the flightcrew as 22,272 pounds, including
1,8 15 pounds for passengers, 259 pounds for cargo, and 3,100 pounds for takeoff
fuel. The calculated weight for the takeoff from LRD was 3,081 pounds below the
maximum allowable takeoff weight of 25,353 pounds.
The airline’s EMB-120 Aircraft Operations Manual stated, “The
balance of the aircraft is controlled by the load in the aft cargo hold. To keep
aircraft CG [center of gravity] within allowable limits, there are minimum and
maximum loads for the aft cargo hold which vary as the passenger load varies.”
A table provided in the airline’s Alert Bulletin 91-17, dated
September 3, 1991, established a minimum weight of 78 pounds and a maximum
weight of 794 pounds for a passenger load of 11 persons. The documented load of
259 pounds in the aft cargo hold was within CG limits.
65
APPENDIX E
COCKPIT VOICE RECORDER TRANSCRIPT
-
Legend of communication descriptions, abbreviations, acronyms and symbols used
in the attached CVR transcript:
CAM
Cockpit area microphone voice or sound
RDO
Radio transmission from accident aircraft
INT
Cockpit Intercom System
-1
Voice identified as Captain
-2
Voice identified as First Officer
-3
Voice identified as flight attendant
-?
Unidentifiable voice
HSTCNTR Houston Center Controller
*
Unintelligible word
#
Expletive deleted
...
Pause
0
Questionable text
[I
Editorial insertion
AIR-GROUND COEMUNICATION
INTFtA-COCKPIT COMMUNICATION
TIWE 6r
TIME &
SOURCE
CONTENT
0933:36
CAM-2
cruise checklist.
0933:59
CAM-3
something to drink?
0934: 08
CAM- 1
gimme some ice.
0934: 11
CAM-2
could I have some ice please?
0934: 13
CAM-3
just ice?
0934: 14
CAM-2
yeah I've still got the * * . . . * * l ?
0934:19
CAM-3
he's so subtle . . he's just so (soft).
0934:24
CAM
0935: 34
CAM
[sound similar to that of the cabin
door closing]
[sound similar to three knocks on
cabin door and door being opened]
0935:39
CAM-1
(that's better.)
0935:45
CAM-2
thank-you sweetheart.
0935: 46
CAM-3
you're welcome.
SOURCE
CONTENT
AIR-GROUND COBMUNICATION
INTRA-COCKPIT COMMUNICATION
TIME br
SOURCE
CONTENT
0935: 47
CAM-1
thanks a lot.
0935: 48
CAM-3
you're * . . . any time.
TIME f
SOURCE
0936: 03
HSTCNTR
0936: 09
RDO-1
0936:29
CAM-2
CONTENT
jetlink twenty-five seventy four when
able fly heading zero five zero
intercept humble two three four radial.
zero five zero join the humble two
thirty four radial jetlink twenty-five
seventy-four.
do you smell something like paint
thinner or -
0936:34
CAM-1
a little bit yeah.
0936: 37
CAM-2
just a
smell.
0936: 56
HSTCNTR
0937:Ol
RDO-1
jetlink twenty-five seventy-four
contact houston center one three two
point two five.
thirty-two point two five good day.
3
68
AIR-GROUND
INTRA-COCKPIT COMMUNICATION
TIME &
SOURCE
0948:41
CAM
CONTENT
TIME &
SOURCE
COMMUNICATION
CONTENT
[sound similar to that of a frequency
change tone]
0948: 43
RDO-1
ah houston center jetlink twenty-five
seventy-four flight level two four
zero.
0948: 47
HSTCNTR
[3:30 minutes of nonpertinent conversation]
0952:31
CAM-3
0952:36
CAM-2
0952:38
CAM-3
you guys is there any way to urn ..deal with * you know it's so cold I've
got the no I'm
sorry we can't.
I've got the thing back here turned
down -
0952:41
CAM-2
well
0952:42
CAM-3
we're freezing back here.
I
got the -
jetlink twenty-five seventy-four
houston center roger.
INTRA-COCKPIT
TIME &
SOURCE
0952:42
CAM-2
0952: 51
CAM-3
0952: 55
CAM-2
0952: 56
CAM- 1
CONTENT
I got control and I've warmed you up a
little bit and now I've warmed you up
as much as your heart can desire.
well we don't wanna be some roasted
jalapinos back here but.
well it's this this airplane (you see) it's
weird it's like either hot or cold.
0952: 59
CAM-2
that's right.
0953:oo
CAM-3
yeah.
0953:oo
CAM-l
it's like a boeing seven twenty-seven.
0953:02
CAM-2
yeah.
0953:03
CAM-2
0953:08
CAM-3
AIR-GROUND
COMMUNICATION
just pass out the blankets and the
cans of Sterno and tell people I will . . . no problem . . . okay I know
I know I know . . . how much time we got
ten twenty?
TIME &
SOURCE
COMMUNICATION
CONTENT
AIR-GROUND
INTRA-COCKPIT COMMUNICATION
TIME k-c
SOURCE
0953:17
CAM- 1
CONTENT
probably.
0953:26
CAM-1
ten thirty ten thirty-five * * *.
0953:34
CAM
CONTENT
yeah we're doing good ah showing about
twenty-four minutes out . . . so about
what ten thirty at the gate -
0953:24
CAM-2
0953 : 29
CAM-2
TIME &
SOURCE
COMMUNICATION
ah I have turned the temperature up
. . . come back in a couple minutes let
me know what you (feel).
[sound similar to that of the cabin
door closing]
[l:OO minute of nonpertinent conversation]
0954:14
HSTCNTR
0954: 19
CAM
jetlink twenty-five seventy-four cross
five five miles southwest of
intercontinental at maintain niner
thousand.
[sound similar to that of changing the
altitude alerterl
0954:20
RDO-1
okay fifty five miles southwest of
intercontinental at niner thousand
we're out of flight level two four zero
jetlink twenty-five seventy-four.
AIR-GROUND
INTRA-COCKPIT COMMUNICATION
TIME br
SOURCE
CONTENT
TIME &
SOURCE
COMMUNICATION
CONTENT
[l:OO minute of nonpertinent conversation]
0954:34
CAM-2
five five southwest at niner.
0954:44
CAM-1
yeah . . . it's not easy.
0955:31
INT-2
radio check.
0955:35
INT-1
I can hear you loud and clear.
0955:36
INT-2
as you also.
0955:37
INT-1
alright there.
[4:07 minutes of nonpertinent conversation]
0959: 44
HSTCNTR
0959: 46
INT-2
zero five zero.
0959:47
RDO-1
0959:51
HSTCNTR
jetlink twenty-five seventy-four say
your heading.
zero five zero.
jetlink twenty-five seventy-four roger
fly heading zero three zero join the
humble two three four radial gland rest
of the route unchanged.
Y
INTRA-COCKPIT
TIME &
SOURCE
AIR-GROUND
COMMUNICATION
CONTENT
TIME bc
SOURCE
0959: 57
RDO- 1
1000: 02
INT-1
okay I'll (drive) you some ATIS.
1000 : 03
INT-2
alright.
[ATIS information golf received on radio two]
1000:32
INT-1
three zero one zero.
1000:35
INT-2
thirty ten.
1001:02
INT-2
dinger is me.
1001:04
CAM
1001:32
INT-1
1001:45
INT-2
[sound similar to that of three tones
followed by aural indicator: "autopilot"]
well I'm back with you. just came up
it's only ten minutes old. it's a golf
twenty-five thousand thin scattered ten
miles eighty-one degrees winds are
zero two zero at five thirty ten and
they're gonna bring everybody in on two
six or two seven.
okey dokey.
COMMUNICATION
CONTENT
zero three zero join the gland six
arrival twenty-five seventy-four.
AIR-GROUND
INTRA-COCKPIT COMMUNICATION
TIME &
SOURCE
CONTENT
1001:46
INT-1
alrighta.
1001:46
INT-2
thanks.
1001:47
INT- 1
you betcha.
1001:55
INT-2
captured on the right.
1002:lO
INT-2
1002:14
INT-1
1003:07
CAM
1003:08
CAM
1003:09
CAM
1003:ll
CAM
push'n this descent making like the
space shuttle.
huh huh.
[sound similar to objects flying about in
cockpit]
[sound similar to a human grunt]
[sound similar to a fluctuation in
prop rpm (decrease then increase)]
[sound similar to three warning tones
and aural warning: "oil"]
TIME C
SOURCE
COMMUNICATION
CONTENT
AIR-GROUND
INTRA-COCKPIT COMMUNICATION
TIME &
SOURCE
1003:13
CAM
CONTENT
[sound similar to three warning tones,
decrease in prop rpm, aircraft
breaking up, depressurization and aural
warning: "high speed oil" and stall
warning clacker]
1003:19
CAM
1003:23
CAM
1003:26
CAM
1003:30
CAM
1003:32
CAM
1003:36
CAM
[sound similar to three warning tones
aural
and stall warning clacker,
warning: " ew- " and fire warning tones]
[sound similar to three warning tones
and stall warning clacker, aural
warning: "autopil-" and fire warning
tones]
[sounds similar to three warning
tones, aural warning: "eng-" and fire
warning tones]
[sound similar to that of power bus
switching tone]
[sound similar to three warning tones
and stall warning clacker]
[sound similar to aural warning: "aural
unit one channel" and stall warning
clackerl
TIME &
SOURCE
COMMUNICATION
CONTENT
FLIGHT DA
S’l,‘Od
s’ll:oo
79
APPENDIX G
LETTER FROM EMBRAER AIRCRAFT CORPORATION
f
sMmRARAm-OC)APQR#T#N
27b SBW. a4ul9ewt tt Lwdwbk, tL maI 5
Td&lonu QOC) ¶C(3mo
FACSIMILE TRANSMISSION
DATP: February 11, 1992
REF:lB/TL/92-MM
TO FACSIMILE L 202~262-6St6
ATTENTION1 mx COXBOY
COMPANY: mm IIC
DEFT:
FROM:Xumel Xonteiro
COMFANY:f ma- ~p~lucr
DEPT: =C-TL
FROM FACSIMILI: (305)99-9%79
SUBJECT: Roquiroa
Re:
?AGLC:2
~YLBpOatiOXl tt- @It)
Baale Lake Raaident
This fax is intended to clarify some 02 the gue8tions that have been
raised in regard to the i8sue of RI1 (Required Irmpection Item).
1.
The RI1 is an operational requirement FAR 132.427(b)(P).
I understand that this requirement ith;dre.,u,‘,’ of maintenance
l hewn that an
experience gained over the years,
overeight io less likely to ooout during' maintenanoe activities
with increased or additional inspections.
2.
‘The intent of the FAR requirement, whiah is in Part 135 and not
FAR 25, is turned toward maintenance/oontrol to insure the
aircraft is always operated fully eonfigured (per its Type
Certificate) a8 parts of it are not intended to depart from it,
least of all in flight. It ie, therefore, the operator'u duty to
assure that within its organization suitable procedures are
established for the timely accomplishment of designated RIIs.
3.
Was the change of the de-ice boots an RI1 function?
The answer is yes.
On page 1-4 of Continental's GMM (General Maintenance Menual)
under paragraph nEtI entitled "Reguired Inrpection Item List",
there is a list of "Designated Itea@, Item 27 is ~Stabilizern and
the scope of that word in defined in paragraph E. 1. a., on page
5-3 of that came manual, where it is written: II. . . any work
performed on those items.m.
The de-ice boot or the leading edge
80
?UtX9/l’L/92=MM
mparately
assembly,
definition.
4.
or
together,
clearly
Page: 2 o f 2
fit into
#at
Manufacturers design, build,
test fly end have aertifled the
aircraft in a defined oonfiguzation and any deviation from that,
in flight, 18 not avisaged, 8ave for 8ome specific condition8
coneidered by the FAR8 (single engine operation, trim runaway,
etc.).
There i8 no way in which the manufaoture will uoond guess the
PARS and 8ugge8t that one 8y8tem or oomponent requires anything
operator can provide. The
rhort of the best maintenanae the
aircraft must be maintained, servloed and inmpected in aacordanoe
with the manufacturer'8 published and approved operation and
maintenance manual8.
Best regards,
Manuel Monteiro
Technical
Liaison/Safety
M M / s
1
l
81
APPENDIX H
PLOTS OF DYNAMIC FLIGHT SIMULATION RESULTS
Loss of Left Horizontal Stabilizer Leading Edge
w =
9350.
KG
CC
=
25.X
2CG
= - 0 . 3 0
u
I‘
-P +
0
5
TIUE (SEC)
10
I
82
Loss of Entire Horizontal Stabilizer
:f
2CC ---
1
2J
I
‘I
9350.
KG0
C C
-2(.’
.
3
0
Y,
I
I
I
IfI
I
I
83
APPENDIX I
SAFETY BOARD RESPONSE TO FAA
National Transportation Safety Board
W ashington, D.C. 20594
July 31, 1992
Honorable Thomas C. Richards
Administrator
Federal Aviation Administration
Washington, D.C. 20591
Dear Mr. Richards:
On July 21, 1989, the National Transportation Safety Board sent 17 safety
recommendations to the Administrator, Federal Aviation Administration (FAA),
addressing safety issues that were identified during the investigation of the
structural failures of the Aloha Airlines, Inc., Boeing 737 that occurred on
April 28, 1988. Five of these Safety Recommendations, A-89-62 through A-89-66,
were directed to the functions of the FAA’s Flight Standards District Offices
(FSDOs) and the National Aviation Safety Inspections Program (NASIP) as they
pertained to surveillance of the Aloha Airlines maintenance and quality assurance
activities.
On November 21, 1990, as a result of the collision with terrain on Molokai,
Hawaii, involving an Aloha IslandAir deHavilland DHC-6, the Safety Board issued
Safety Recommendation A-90-136 to the FAA. This recommendation addressed
the ability of FSDOs to oversee airline operations and maintenance activities.
The FAA has responded to all these recommendations; however, in some
cases, the Safety Board withheld its evaluation of the FAA’s actions pending the
outcome of another accident investigation involving an air carrier’s maintenance
activities and FAA surveillance. This accident occurred on September 11, 1991,
involving the m-flight breakup of an Embraer Brasilia EMB-120 airplane operated
as Continental Express flight 2574 near Eagle Lake, Texas. Following that
accident, on February 28, 1992, the Safety Board issued Safety Recommendations
A-92-6 and A-92-7 to the FAA regarding the surveillance of air carrier
maintenance departments.
84
The Safety Board has now completed its investigation of the Continental
Express accident. The following addresses the status of the referenced safety
recommendations:
A-89-62
Revise the National Aviation Safety Inspection Program objectives
to require that inspectors evaluate not only the paperwork trail, but
also the actual condition of the fleet airplanes undergoing
maintenance and on the operational ramp.
A-89-63
Require National Aviation Safety Inspection Program teams to
indicate related systemic deficiencies within an operator’s
maintenance activity when less than satisfactory fleet condition is
identified.
In correspondence to the Safety Board dated May 24, 1991, the FAA
indicated that the NASIP procedures have been revised to incorporate the elements
of these recommendations. The Safety Board viewed the FAA’s response as
positive until the accident at Eagle Lake, Texas, showed that some continuing
effort is needed to ensure effective air carrier maintenance and quality assurance.
Because the essential elements of Safety Recommendations A-89-62 and A-89-63
were reiterated in Safety Recommendation A-92-7, the Safety Board classified the
recommendations as “Closed--Acceptable Response/Superseded.”
A-89-64
Evaluate the quality of FAA surveillance provided by the principal
inspectors as part of the National Aviation Safety Inspection
Program.
A-89-65
Integrate the National Aviation Safety Inspection Program team
leader in the closeout of the team findings.
The FAA’s responses to these recommendations dated June 21, 1991, and
August 16, 1991, referred to a meeting of FAA and Safety Board staff personnel on
June 13, 1991. The Safety Board’s staff was advised that the FAA, under current
policy, uses the NASIP findings to evaluate the performance of the FSDOs
responsible for the routine surveillance of the inspected air carriers. Furthermore,
during the June 13, 1991, meeting, FAA staff agreed that they would provide each
team leader with a copy of the initial actions taken to resolve NASIP findings and
that they would consider team leader comments on such actions. Based upon these
staff discussions, Safety Recommendation A-89-64 will be classified as “Closed-Acceptable Action,” and Safety Recommendation A-89-65 will be classified as
“Closed--Acceptable Alternate Action.” However, the Safety Board believes that
the FAA should consider issuing formal documentation to ensure the consistent
and continued implementation of these policies. The Safety Board will continue to
monitor the effectiveness of NASIP actions as they pertain to future accident
investigations.
A-89-66
Enhance the stature and performance of the principal inspectors
through: (1) formal management training and guidance, (2) greater
encouragement and backing by headquarters of efforts by principal
inspectors to secure the implementation by carriers of levels of
safety above the regulatory minimums, (3) improved accountability
for the quality of the surveillance and (4) additional headquarters
assistance in standardizing surveillance activities.
The FAA response to this Safety Recommendation, dated January 2, 1992,
defined a number of FAA initiatives that are intended to improve the performance
and management at the field level and to ensure that FAA field inspectors are fully
supported by FAA senior management. Many of these initiatives are an outgrowth
of Project SAFE (Safety Activities Functional Evaluation), and others involve the
development of improved automation tools, such as the Air Carrier Analysis
System and the Program Tracking and Reporting System. Based on these
initiatives, the Safety Board is classifying Safety Recommendation A-89-66 as
“Closed--Acceptable Action.”
A-90-1 36
Perform a special study of the adequacy of Flight Standards
District Office staffing considering the availability of work hours,
the geographic area of responsibility, and the size and complexity
of the assigned operations.
86
On February 8, 1991, the FAA responded to this recommendation stating
that it had contracted for a study that would revalidate its staffing standards and
would include the availability of work hours, geographic areas of responsibility,
and the size and complexity of operations. This study was originally scheduled to
have been completed in October 1991. Subsequent FAA correspondence on
February 11, 1992, indicated that the completion of the study had been delayed.
Pending the completion of the study and the FAA’s further response, Safety
Recommendation A-90-136 will remain classified as “Open--Acceptable
Response.”
A-92-6
Enhance flight standards surveillance of Continental Express, to
include sufficient direct observation of actual maintenance shop
practices to ensure that such practices conform to the Continental
Express General Maintenance Manual and applicable Federal
Aviation Regulations.
In a letter of May 15, 1992, the FAA indicated its agreement with the intent
of this recommendation. From March 12 through April 3, 1992, the FAA
conducted several inspections of Continental Express maintenance shop practices
in Houston, Denver, Cleveland, and Allentown. Special emphasis was placed on
direct observation to ensure that practices conformed to the airline’s General
Maintenance Manual. No adverse practices were found. However, the Safety
Board expects that the FAA will continue to observe actual shop practices in
Continental Express maintenance facilities to ensure that personnel continue to
comply with the General Maintenance Manual. The Safety Board would like to be
informed of the results of subsequent inspections and will classify Safety
Recommendation A-92-6 as “Open--Acceptable Response” pending further
information.
A-92-7
Enhance flight standards Program Guidelines, including the
National Aviation Safety Inspection Program, to emphasize handson inspection of equipment and procedures, unannounced spot
inspections, and the observation of quality assurance and internal
audit functions, in order to evaluate the effectiveness of air carrier
maintenance programs related to aircraft condition, the adherence
87
to approved and prescribed procedures, and the ability of air
carriers to identify and correct problems from within.
The FAA noted in its May 15, 1992, letter that the NASIP has been revised
to include “hands-on” inspections of employee shift changes and/or interrupted
work and required item sign-offs. Also, on April 8, 1992, the FAA approved an
Airworthiness Inspector’s Handbook Bulletin entitled “Adequacy of
Communication Between Arriving and Departing Maintenance Shifts” to address
this issue. The Safety Board is also aware of other actions to encourage air carriers
to develop internal self-audit programs for better quality assurance. However,
before final disposition of this safety recommendation, the Safety Board would like
further information about the typical frequency of unannounced shop visits by FAA
inspectors to air carrier maintenance facilities and the results of common findings.
Pending further information, the Safety Board is classifying Safety
Recommendation A-92-7 as “Open--Acceptable Response.”
The Safety Board will continue to evaluate any issues involving FAA
surveillance as identified during its accident investigations. However, the Safety
Board acknowledges the positive actions that have been taken by the FAA in
response to these safety recommendations.
Sincerely,
Carl W. Vogt
Chairman