Local safety action
Following the occurrence, the forward galley and forward toilet
drain lines were inspected and no defects were found. The operator
subsequently removed the forward galley and resealed the floor in
the forward vestibule area under the galley and adjacent to the
forward doors before re-installing the galley components. The
operator's maintenance provider designed and manufactured an
approved aluminium moisture shroud that was fitted in the E/E bay
on 11 December 2001. Functional checks on the avionics units were
carried out and the aircraft was returned to service on 13 December
2001.
The operator advised that a safety article would be distributed
to cabin crew with specific feedback in regards to this occurrence.
An operations manual revision would be initiated to include a
discussion about varying emergency descent profiles and aircraft
effects. The operator also advised that this information would be
included in initial and recurrent cabin crew training.
- The aircraft systems faults reported most likely arose as a
result of moisture ingress into a number of electronic components
located in the aircraft E/E Bay. - Moisture ingress was possible due to moisture shields not being
fitted in the E/E Bay following the modification, by the previous
operator, to remove the airstairs from the aircraft. - The moisture shields were deemed by the aircraft manufacturer
to be mandatory equipment. - The cabin floor angle during the emergency descent was reported
as being similar to a normal descent and this may have led the
cabin crew to assess that it was safe to return to their crew seats
at the rear of the cabin during the emergency descent.
The reason for the cabin depressurisation was likely to have
been due to the moisture shrouds not being fitted after the removal
of the airstairs. This permitted the ingress of water into the E/E
bay and the pressurisation controller, resulting in a malfunction
of the operating modes of the unit. The electrical faults found in
the yaw damper coupler electrical filter and the rudder power
control unit may have led to the rudder pedal movement detected by
the flight crew during the descent into Brisbane.
Extensive research has shown that the effect of oxygen
deprivation can be insidious and, as such, cabin crew may not be
able to correctly judge their oxygen intake. Research conducted by
the US Federal Aviation Administration (FAA) Civil Aeromedical
Institute indicates that physical activity such as that performed
by a cabin crewmember will significantly shorten the time of useful
consciousness during an aircraft depressurisation. Based on that
research, the FAA's recommended procedures for cabin crew during
depressurisation are to immediately don the nearest oxygen mask,
sit down, or grasp a fixed object and hold on in order to brace
themselves until given clearance to move about the cabin by the
flight crew.
The operator's emergency procedures for immediate action for
cabin crew following depressurisation reflected that advice.
However, in this occurrence, two cabin crew, while returning to the
rear crew seats, assisted some passengers before taking oxygen
themselves. A further delay to oxygen intake occurred as oxygen
masks, that had failed to deploy automatically above the rear crew
seat, had to be manually released before use.
Some cabin crew appeared to have judged that the angle of the
cabin during the descent was not very steep. The operator's
emergency procedure manual referred only to a `very steep' angle of
descent and other more severe characteristics of depressurisation.
The manual did not discuss the possibility of an emergency descent
that may be less than `very steep' or indications that may be less
severe than those associated with a rapid or explosive
depressurisation. This may have led some cabin crew to believe that
in the absence of other characteristics associated with
depressurisation, they could safely assist passengers while moving
to the rear of the aircraft before using crew oxygen. Remaining
where they were, using oxygen until advised by the flight crew that
a safe level had been reached, may have been a safer practice than
moving through the cabin to reach the rearmost crew seats before
taking oxygen.
The Boeing 737-33A aircraft, registered VH-CZQ, departed
Townsville at 1805 EST on a scheduled passenger service to
Brisbane. Soon after levelling at flight level (FL) 330, the flight
crew observed that the master caution light, the cabin
pressurisation auto-fail light and the pressurisation standby light
had illuminated. The crew completed the non-normal procedure and,
as cabin pressurisation was being maintained, they decided to
continue the flight to Brisbane.
At about 1909, the flight crew experienced physiological
sensations, which indicated that the flight deck was
depressurising. The crew donned their oxygen masks and the copilot
noticed that the cabin rate of climb indicator was displaying a
rate of climb of 4,000 fpm. Shortly afterwards, as the cabin
altitude climbed through 10,000 ft, the crew observed that the
master caution light and passenger oxygen `on' light had
illuminated and heard the cabin altitude warning horn. The flight
crew completed the non-normal procedure for a rapid
depressurisation and emergency descent and advised air traffic
control that the aircraft had left the cruising level due to a
depressurisation. The pilot in command advised the cabin crew on
the public address system (PA) of the descent.
The cabin in-flight service had been completed and the four
flight attendants were at the front of the aircraft when the oxygen
masks dropped from overhead passenger service units (PSU)
throughout the cabin. When the oxygen masks dropped, the cabin
supervisor returned to his seat at the forward left entry door,
donned an oxygen mask and removed the PA handset from its
receptacle. Another flight attendant also returned to his crew seat
at the forward left door and donned an oxygen mask. The two
remaining flight attendants, standing at the front of the aircraft
when the oxygen masks dropped, returned to their crew seats at the
rear left entry door.
Some passengers were assisted by a flight attendant as she
walked to the rear cabin. When the flight attendant reached the
rear crew seat she found that the crew oxygen masks had not
automatically deployed from the overhead stowage. She released the
masks by activating the manual release on the overhead panel, and
was seated just as the second flight attendant reached the crew
seat.
The flight attendants reported that the floor angle during the
emergency descent seemed no different than a normal descent and
they did not notice any movement of cabin or galley items. One
flight attendant reported that she did not feel in danger. Her ears
were `popping', but there were no indications of depressurisation
such as those taught in emergency procedures training. The flight
attendants reported that they `weren't short of breath, felt [they]
had time, there were no objects flying in the cabin or cabin
misting'.
The flight data recorder information showed that at 09:08:55,
while in cruise at FL330, the cabin altitude warning activated
indicating that the cabin altitude had exceeded 10,000 ft.
Thirty-six seconds later at 09:09:31, the emergency descent was
initiated with the thrust levers being closed and the autopilot
level change mode being selected. The pitch attitude then decreased
from 3.2 degrees nose up to 4.4 degrees nose down and the aircraft
began to descend at a rate of 4,600 feet per minute. At 09:11:01,
the speedbrake was deployed as the aircraft descended through FL276
at a rate of 4,500 feet per minute, with a pitch attitude of 3.7
degrees nose down.
The maximum nose-down pitch attitude during the emergency
descent was 4.4 degrees for a 6-second period as the aircraft
descended through FL 285 at 5,000 feet per minute. By the time the
aircraft reached FL199, the pitch attitude had reduced to 1.4
degrees nose down, with a rate of descent of 3,400 feet per minute.
By way of comparison, the pitch attitude during a normal descent of
a B737-300 series aircraft is about 2 degrees to 3 degrees nose
down above FL260, reducing to about 1 degree nose down during the
latter part of the descent.
When the aircraft levelled at 10,000 ft, the pilot in command
advised the passengers by PA that oxygen masks would no longer be
required, and instructed the cabin crew to commence follow-up
actions. The cabin supervisor went to the flight deck to confirm
instructions and advise the pilot in command of the conditions in
the cabin. The flight attendants reported that, although all
passengers were calm, three or four passengers needed reassurance,
but additional oxygen was not required. The flight crew continued
the flight to Brisbane at 10,000 ft and landed at 1953 without
further incident. There were no injuries.
The flight crew later reported that they felt a slight movement
on the rudder pedals on two occasions during the descent into
Brisbane. The crew considered the rudder pedal movements to be
minor and did not take any further action. About the same time, the
crew also noticed that the map display on the Electronic Horizontal
Situation Indicator (EHSI) was incorrect. The aircraft position
depicted on the EHSI differed from the position indicated on the
other navigation instruments.
Operator's Operations Manual
Volume B3 of the operator's Operations Manual contained
information about the safety equipment and procedures. Section 5.1
included information about cabin depressurisation and actions to be
taken by the flight crew and cabin crew following a
depressurisation event. The manual used the term decompression when
referring to depressurisation.
The manual stated that there were two types of decompression,
gradual and rapid (explosive). The information provided in section
5.1 related primarily to the rapid type of decompression. The
manual stated that `the angle in the cabin will become very steep
as the aircraft descends at an approximate rate of 6000 ft per
minute (normal rate of descent is approximately 2000 ft per
minute)'. The manual also contained a warning that required the
immediate use of oxygen by all crewmembers following the deployment
of drop out oxygen masks in the passenger cabin. The manual advised
cabin crew that the effects of a decompression event on the
aircraft were:
- Sudden boiling of liquids;
- Loud noise as air escapes;
- Air becomes thin, cold and dry;
- Fog forms in cabin (should not be confused with smoke);
- Dust and objects blown about;
- Smoke alert devices in toilets may be activated; and
- Flight crew may initiate an emergency descent causing the angle
in the cabin to become very steep.
Aircraft examination
Following the incident, the electrical/electronic (E/E) bay was
inspected and water was found to be dripping from the forward
galley floor into the bay. There was also evidence of moisture
leakage under the forward passenger door and service door. Moisture
stains were found on the racks and ducting within the bay. The
inspection also revealed that the moisture shroud was missing from
above the E1 rack in the forward part of the bay.
A number of avionics units were removed from the aircraft,
including the pressurisation controller, yaw damper coupler,
auto-throttle computer and two stall warning computers. These units
exhibited evidence of water damage, including moisture staining of
component casings and corrosion of connector pins. There was no
record of the maintenance staff finding moisture contamination in
the E/E bay prior to the incident.
The operator stored beverages in polystyrene ice containers that
were stowed in the forward galley. The containers were partially
filled with crushed ice that melted during flight. The flight
attendants on the incident flight did not recall any water spillage
occurring from the ice containers, nor did they recall any
turbulence that may have caused a spillage. The flight attendants
reported that on previous flights on various aircraft there had
been spillage from the ice containers, usually during landing.
Component testing
The manufacturer of the pressurisation controller inspected the
unit and subjected it to a series of tests. A visual inspection of
the outer case of the unit and the circuit boards did not find any
damage or anomalies and a series of functional tests did not reveal
any faults. Although there were indications of moisture staining on
the outer case, there was no evidence of any residual damage to the
unit due to water contamination. The manufacturer was aware of
other instances where water had entered pressurisation controllers
and caused problems in one or more of the operational modes.
An inspection of the yaw damper coupler found that the
electrical filter was shorting out. After the filter was replaced
the unit operated satisfactorily. A dent was also found on the top
cover of the outer housing of the coupler, but it did not affect
the operation of the unit. An inspection of the rudder power
control unit found that the transfer valve was unserviceable due to
an open electrical circuit.
Aircraft information
The aircraft was acquired from another Australian operator
during October 2001 and underwent a pre-delivery inspection prior
to commencing passenger operations on 1 November 2001. The
operator's contracted maintenance provider conducted the inspection
that was intended to ensure compliance with airworthiness
directives and service bulletins.
The previous operator had purchased the aircraft from the United
Kingdom and it was entered onto the Australian civil register on 3
December 2000. Between December 2000 and January 2001 the aircraft
underwent modification and heavy maintenance work at an overseas
engineering facility before entering revenue operations in
Australia. The modifications included the removal of an airstair
from under the forward passenger entry door. During that work the
airstair drip pan and the cloth moisture shroud were removed from
the E/E bay.
The documentation covering the removal of the airstairs
specified that the moisture shroud was to be replaced following the
modification work. However the shroud was not installed because the
kits were temporarily unavailable from the manufacturer. One of the
operator's engineers, authorised by the Civil Aviation Safety
Authority to approve a design modification or repair, assessed that
the absence of the moisture shroud would not affect the safety of
the aircraft. On 29 January 2001, the engineer approved an
amendment to the engineering release that permitted the
installation of the shroud within 12 months of receipt of the
parts.
The aircraft was subsequently operated on Australian domestic
passenger services between February and September 2001 without the
moisture shrouds being fitted. An order was placed with the
manufacturer for the shroud kits and the delivery of the kits was
due in November 2001. The moisture shrouds were not fitted to the
aircraft when it was delivered in October 2001 to the current
operator, who was unaware that the shrouds had not been fitted to
the aircraft.
During the investigation, the aircraft manufacturer provided the
following advice:
`Boeing advises that these shrouds are required in order to
ensure the airworthiness of the airplane. 737 airplanes should not
be used for revenue flight with these moisture shrouds not
installed.'
`The 737 MMEL/DDPG does not provide any relief for these items
to be removed from the airplane. Furthermore, the purpose of these
shrouds is to protect the electronic equipment from moisture
ingress. There have been several reports of 737 airplanes
experiencing uncommanded flight movement due to moisture ingress
into certain electronic components in the E/E Bay.'
On 18 April 2002, another Australian registered Boeing 737-33A,
VH-CZR, sustained a depressurisation incident during a scheduled
passenger service from Auckland to Christchurch, New Zealand. The
incident occurred soon after top of descent when the crew selected
the pressurisation system to the standby mode. The aircraft had a
history of pressurisation controller problems when being operated
in the AUTO mode, until approximately three weeks prior to the
incident when the controller was replaced. The pressurisation
controller fitted to the aircraft at the time of the occurrence
failed during post incident testing, as the selection of standby
mode resulted in the cabin outflow valve moving to the fully open
position.
VH-CZR had a similar service history as CZQ, having been
purchased by the same previous operator and undergoing similar
airstair removal modifications at the overseas engineering facility
prior to entering service in Australia. The aircraft also had not
been fitted with moisture shrouds following the removal of the
airstairs and the moisture shield over the E1 rack was not
installed on the aircraft at the time of the occurrence in New
Zealand.