Canadian Transportation Safety Board focuses on in-flight fires
On March 27, 2003, the
Canadian Transportation Safety Board (TSB) released its long-awaited
report into the September 2nd, 1998 crash of SwissAir Flight 111 (SR
111), a McDonnell-Douglas-11 which crashed near Peggy’s Cove, Nova
Scotia, killing all 229 people on board. The TSB’s multi-million
dollar investigation focused on in-flight fires, which may overcome an
aircraft and its crew before there is time to take advantage of
ground-based ARFF. Ironically, the aircraft was only 66 miles and 20
minutes from Halifax when the tragic events began. This only
underlines the dangers faced many times each day by airliners
venturing over water or inhospitable terrain, hundreds of miles and
hours away from suitably-equipped recovery airports.
This article is
organized into five parts, the first four mirror and summarize the TSB
report; the fifth is an exclusive interpretation of the report for AFJ
readers.
1. Factual
Information
Shortly after the
aircraft reached its initial cruising altitude of 33,000 feet on its
way from New York’s JFK airport to Geneva, the crew detected an
unusual odor in the cockpit. At first, the pilots believed it was
probably an air conditioning anomaly, but within 4 minutes, the crew
had declared a “Pan-Pan-Pan” and requested an immediate return to the
nearest airport.
Air Traffic Control in
Moncton, New Brunswick cleared the aircraft to Halifax, Nova Scotia
and down to 31,000 feet, later clearing it down to 3,000 feet, in
anticipation of a landing on runway 06 at Halifax. The crew was
pre-occupied with checklists for descent and dealing with air
conditioning smoke and smoke/fumes of unknown origin. Unable to make a
stabilized approach directly into Halifax, carrying fuel for a
Trans-Atlantic flight and on the wrong side of Halifax for the 06
approach, the crew requested and was granted time to lose altitude and
to dump fuel.
Thirteen minutes after
first detecting the odor, the autopilot and other electrical systems
(yaw dampers, flight control computers, Left Emergency AC Bus,
Altitude, Airspeed, cockpit displays) began to fail, and a minute
later, the crew declared an emergency. The last communication from
the aircraft was heard only 20 seconds later. Half a minute after
that, the cockpit voice and flight data recorders stopped functioning.
The crew shut the number 2 engine down while still in flight, possibly
because fire-damaged systems gave a false indication that the
tail-mounted engine was on fire. Five minutes later, the doomed
aircraft hit the water in a spiral dive (bank angle estimated between
60 and 110o) at approximately 300 knots, subjecting all 14
crew and 215 passengers on board to an unsurvivable force of “in
the order of at least 350 g.” The aircraft broke up into more
than 2 million pieces, and then sank into water 90 metres deep.
Rescue efforts by local
fishermen, the Canadian Coast Guard and the Canadian Navy soon proved
futile and the operation then shifted into the recovery mode, which
turned out to be a staggering challenge. Five major attempts over 13
months were made to recover underwater debris, including by Canadian
Navy divers, the salvage ship USS Grapple, heavy lift barges,
scallop dragger operations and a suction hopper dredge. Remotely
operated submersibles were used during all phases of the recovery.
Ultimately, TSB investigators were able to recover almost 98% of the
aircraft by weight, much of it in tiny pieces.
The precise location of
the two data recorders, ultimately located by a Canadian Navy
submarine, and recovered by Navy divers on the 4th and 9th
day of the investigation, was complicated by the fact that both were
transmitting on the same frequency. Although they were recovered
largely intact, investigators were frustrated to find that electrical
power to both units had been cut five minutes before the impact,
leaving them without a significant amount of critical data.
Although a major
setback, this provided clues about the crash: both data recorders were
powered from the same source: the Generator 3 (i.e. right engine) bus.
During the emergency checklist dealing with smoke/fumes of unknown
origin, each of the three engine-driven generators buses is
deactivated in sequence to isolate a possible source of smoke/fumes.
This procedure may take several minutes per bus, or 20-30 minutes in
total (35 mins according to the Chief McDonnell Douglas Test Pilot Tom
Moody here). When the FDR failed, the aircraft was still in a clean
configuration - flaps, slats and wheels up. The fact that the aircraft
crashed with the hydraulically actuated flaps set to 15o
but the electrically actuated slats retracted, contrary to their
normal extended position for that flap setting, was yet another
indication that the aircraft was experiencing serious electrical
difficulties.
Heat and Fire Damage
Any fire in the cockpit
ceiling did not penetrate the aircraft skin, nor was there any
discoloration of the aircraft’s white exterior paint. Yet there was
evidence of a fierce fire in the attic area – hidden above and behind
the flight crew seats, all the way back to the cockpit bulkhead and
into the first class and even business class cabin. The inside of the
aircraft’s skin in this area was heavily sooted. The insulation
blankets in the attic, some two layers thick, seemed to protect the
aircraft skin from fire. However, the metalized covering of these
blankets that were pressed against a maze of wiring and ducting in the
attic was damaged or destroyed.
There was also
considerable evidence of heat and fire damage to the air distribution
ducts in the cockpit and cabin areas, which proved helpful in
determining the areas of fire. Vents near the forward galley, located
immediately aft of the cockpit, designed to catch and expel cooking
odors, played a role in the transmission of smoke and heat: two of
these ducts sustained high heat damage. Part of the first class
ceiling area had been exposed to temperatures as high as 593oC/1100oF
for 10 minutes.
Floor carpets as far
forward as the cockpit and seats as far aft as business class also
showed hot material dripping down from the ceiling. The dripping and
sagging cockpit ceiling liner afforded further clues: it begins to
soften and sag at between 246-274oC/475-525oF.
Parts of this area were found to be heavily sooted, discolored dark
brown and black, blistered and bubbled, suggesting temperatures of at
least this magnitude, if not more. A circuit breaker panel on the
right hand side of the cockpit, behind the copilot’s seat, was found
to have been exposed to temperatures similar to those from 430-620oC/800-1150oF
for 10 minutes. The overhead circuit breaker panel above and between
the two pilots’ seats was found to be similarly damaged.
Potential Ignition
Sources
The investigators paid
considerable attention to the area of the aircraft between the cockpit
and a section just into the first class area of the cabin. Of
particular interest was the attic, the area above the ceiling but
below the roof of the aircraft, which was filled with wiring, ducting
and insulation blankets. With unmistakable evidence of a fire in the
attic area of the cockpit and forward cabin, the obvious question
concerned the ignition source.
The investigators
concluded the most likely source was electrical energy. Sections of
four bundled power supply cables for the In-Flight Entertainment
Network (IFEN) were found with melted copper, an indication of an
arcing event. These cable sections, which had been exposed to
temperatures in the range of 500oC/932oF for 10
minutes, were found on the ceiling on the right hand side of the
cockpit bulkhead. Examination of these wires suggested two areas of
arcing, the first, which did not trip the associated circuit breaker,
and a second arc that’s believed to have raised the temperature of the
circuit to a level where the circuit breaker likely did trip. The
sequence and location of these events suggests that the fire started
forward and then initially moved aft with the airflow.
Also of great interest
were the materials used, notably wire insulation and the
thermal/acoustical blankets used in aircraft, in this case, metalized
polyethylene terephthalate (MPET) known by its trade-name as metalized
Mylar®.
Although MPET had
earlier passed the FAA’s vertical Bunsen burner test, the Civil
Aviation Administration of China (CAAC) discovered, following from
three incidents, that once burning, MPET could be completely consumed
by fire. The CAAC brought this to the FAA’s attention in 1996; the
latter agency said that they while they would look into the matter,
the tests used by the CAAC were not required for FAA certification.
Less than six weeks after the SR 111 crash, the FAA announced it would
develop new specifications for aircraft insulation materials. It was
not before 2000 that the FAA mandated the removal of MPET covered
insulation blankets.
One area of concern to
the investigators was the need to fire-harden aircraft systems. It was
noted that the aluminum lines in the flight crew oxygen system could
leak at temperatures as low as 427oC/801oF.
Above this temperature, the oxygen lines had a tendency to rupture,
sometimes in as little as three minutes, introducing pure oxygen into
a fire environment, greatly exacerbating the situation. Similarly,
fiberglass ducting and end-caps showed little tolerance to high heat,
again providing fire with a fresh air supply and a means of
propagation.
The TSB conducted
airflow tests to determine how smoke and fumes would be influenced by
various ventilation fans being turned on and off during the various
emergency checklists. It was found that some switch selections caused
airflow reversal between the cockpit and the cabin.
On Board Fire-Fighting
Capability
In terms of the aircraft’s own fire-protection and fire-fighting
system, it was noted that in keeping with FAA regulations, there were
fire detection and fire suppression systems in the designated cargo
areas in the belly and in the engines and Auxiliary Power Unit (APU)
as well as lavatories. The cockpit and cabin were not required to have
detection or suppression systems; instead there were 8 portable fire
extinguishers, five of which were 2.5 pound Halon 1211 and two five
pound monoammonium phosphate (dry chemical) extinguishers. The
remaining portable extinguisher was a single 2.5 pound Halon 1211 unit
in the cockpit, mounted on the rear wall, out of the immediate reach
of the pilots.
The report noted that,
in accordance with Joint (European) Aviation Requirements, the cabin
crew had received initial and recurrent training on firefighting,
including the importance of identifying the source of the fire, the
location, handling of fire fighting equipment, communicating with the
cockpit and firefighting responsibilities. As was consistent
throughout the industry at the time, there was no training in how to
deal with fires in the cockpit or in the inaccessible areas, such as
the attic.
The investigators also
took great note of the aircraft’s electrical supply system, including
the power fed to the IFEN as they began to focus on electrical arcing
events above the cockpit and over the bulkhead that separates the
cockpit from the first class cabin. This arcing could be caused by
metal-to-metal contact between an exposed and energized wire and a
source of an electrical ground. Arcing events can generate extremely
high temperatures by the arc (up to 5000oC/9,000oF
or more), the vaporization of molten conductor and local gases. If
flammable materials are nearby, this can be the ignition source for a
fire.
The investigators found
a total of 21 wires that showed melted copper - an indication of an
arcing event. One wire was insulated with irradiated ethylene-tetraflouroethylene
(known as XL-ETFE or its trade name, Cross-Linked Tefzel®); the others
were insulated either by polyimide (known as PI or its trade-name,
Kapton®) or ethylene-tetraflouroethylene (known as ETFE or its
trade-name, Tefzel®).
The general purpose wire
in aircraft HB-IWF was polyimide wire, which although light in weight,
a good insulator, resistant to abrasion and producing little smoke
when burned, it was susceptible to arc tracking, a process where the
insulation, when charred, turns into a conductor, and can, over time,
cause a massive arc. The insulation on the IFEN wiring was EFTE, with
the exception of the power supply cables which were PTFE (polytetraflouroethylene,
known by its trade name, Teflon®)
Owing to differing
resistance to abrasion, wires with differing insulation should not be
installed in close proximity to one another, as the tougher insulation
can, under circumstances of heat and vibration, abrasively wear
through the insulation of a wire with softer insulation. PI is the
most resistant to abrasion, EFTE is less resistant to abrasion and
cutting than PI; PTFE is the least resistant to abrasion and cutting.
It is interesting that
the report cites an undated document produced by the MD-11’s
manufacturer, advising “that high vibration and wire-to-wire
abrasion testing has shown that, when properly installed, the mixing
of different approved insulation types has not been a problem.”
The report indicates
that as certified and installed on HB-IWF, the original IFEN system
design did not incorporate an ON/OFF master switch. The ability to
turn the system on or off was achieved by pulling a
circuit breaker. The report addressed the negative implications of
aircraft circuit breakers being used in this manner, including the
wear and tear on the circuit breakers, and the subtle suggestion that
circuit breakers can be regarded as merely a switch of convenience,
rather than a thermal protection device whose tripping should be
interpreted as a serious safety event. In addition, it has been found
that manually tripping breakers can significantly change its rated
trip setting over time.
Of particular interest
to the investigators was the IFEN that was certified and installed in
the aircraft in 1997 in accordance with an FAA Supplemental Type
Certificate (STC). When SwissAir ordered a suite of IFENs for its
fleet, its technical services organization, SR Technics, engaged the
services of the IFEN designer, In Flight Technologies (IFT). However,
IFT had to engage the services of Hollingsead International (HI) to
integrate the IFEN into the aircraft system. The work would all have
to be done in accordance with a FAA Supplemental Type Certificate (STC),
which had been granted by the FAA to Santa Barbara Aerospace (SBA).
The Swiss Federal Office of Civil Aviation (FOCA) was willing to
accept the FAA-approval, as delegated to SBA. Thus, roles and
responsibilities were arguably spread between four or even five
companies and two regulatory authorities, although FOCA did not assume
any direct responsibility for authorizing or overseeing the IFEN
installation.
The TSB report
identifies several irregularities with the installation and
documentation of the IFEN, some of it likely caused by the lack of
familiarity of HI and SBA staff with the MD-11 or its electrical
design philosophy. FAA monitoring of SBA was also deemed to be lax: a
special FAA review done after the crash showed shortcomings in both
SBA’s certification procedures and the FAA’s monitoring of the
project. SBA subsequently went out of business.
There were several
in-flight problems reported with SwissAir’s IFEN prior to the crash –
overheating and short circuits were noted in several cases. The
aircraft’s air-conditioning controller’s range had to be varied to
cope with the heating. Inspection of other SwissAir IFEN installations
showed installation anomalies.
2. Analysis
The 38 pages
comprising this section of the TSB report allow the investigators to
marshal their evidence in many of the areas enumerated above. This is
required if the Board is to support the conclusions, and more to the
point, to gain the Board’s support for recommendations that might
impose significant financial burdens on carriers, regulators, and by
extension, travelers and taxpayers. A properly analyzed finding is
more likely to result in a conclusion, as will be found in the next
section.
This section identified
that neither fatigue, nor incompetence on the part of the flight crew
or air traffic controllers nor criminal activity directed at the
aircraft were involved. However, it did note that all six aircraft
power bus feed cables are routed together near the overhead switch,
creating a risk that all services provided by these cables could be
lost by a single point failure.
Development of Fire
What is particularly
interesting about this section is the degree to which the
investigators, normally disciplined to comment only on facts that can
be ascertained, delve into speculation. As much out of necessity, the
words such as likely, could reasonably be linked to, possible
appear with unprecedented regularity. The most likely source of
ignition for the fire which spread throughout the attic in the cockpit
was “an electrical arcing event involving breached wire insulation
that ignited nearby MPET-covered insulation material.” This event
likely took place above the cockpit bulkhead. Investigators believe
the fire developed and propagated aft, out of the cockpit ceiling
area, and into the ceiling over the galley area. The hot temperatures
likely melted the insulation on Tefzel® wiring, which caused arcing,
and a tripping of various circuit breakers.
IFEN Design, Installation and Certification
The original design of
the IFEN had it wired to the Cabin Bus, which would be the first to be
isolated during any emergency load shedding. However, because the
Cabin Bus could not provide sufficient electrical power to the
original IFEN installation (a full 257 seat configuration) the 115
volt AC Bus was used instead. This change was not signaled to the
pilots, who did not know that the IFEN was powered when they may have
believed it was not.
The investigation looked at the convoluted manner in which the task of
installing the IFEN was managed by SwissAir and regulated by the FAA.
The contracting and subcontracting between SwissAir, SR Technics, IFD,
HI and SBA almost ensured that responsibilities and accountabilities
were diffused. The report stated that along with the hands-off roles
played by the FAA and FOCA, “the overall result was the IFEN STC
project management structure did not ensure that all the required
elements were in place to design, install and certify a system that
would be compatible with the MD-11 Type Certificate.”
3. Conclusions
This is a
summarization of the most relevant findings:
1.
Aircraft standards for material flammability are inadequate.
This allowed fire to spread and intensify which ultimately lead to the
loss of control of the aircraft.
2.
MPET insulation blankets, duct end caps, fasteners, foams,
adhesives are flammable.
3.
Current circuit breaker design does not afford sufficient
protection
4.
The fire most likely originated in the cockpit bulkhead area,
near a power supply cable for the IFEN. This area lacked a fire
detection and suppression system.
5.
Aircrew were expected to use sight and smell to detect
smoke.
6.
The lack of an integrated fire-fighting plan lead to more
effort being given to landing the aircraft than locating and dealing
with the fire.
7.
Certification standards lack consideration of fire as a
failure mode.
8.
The lack of access to certain areas in the cabin (i.e.
attic) made it difficult, if not impossible to fight fires in
those areas.
9.
Checklists for dealing with smoke and fumes were so lengthy
(25-35 minutes) that they allowed other ignition sources.
10.
Emergency checklists did not emphasize the need to land
the aircraft.
11.
Shortfalls in aviation industry installation, maintenance
and inspection of wire were noted.
12.
The fire hazard of contamination (lint, debris, metal swarf
etc) is not fully understood by the aviation community.
13.
High intensity map-lights were a potential source of fire
risk, especially if contamination is present.
14.
Aluminum oxygen lines were susceptible to leaking, pin-holing
by arcs and rupturing into a blowtorch, exacerbating any fire
situation.
15.
Best practices with respect to the use of circuit breakers
are not yet universal. (i.e. don’t use as a switch and any resets
must be justifiable).
16.
Both Cockpit Voice and Flight Data recorders should be
powered by separate sources, and also from a source independent
of the aircraft system.
17.
The FAA’s STC process was fundamentally flawed.
18.
Databases for capturing wiring anomalies were inadequate.
4. Safety Action
The report identifies the safety action that has taken place since the
crash, the safety action required and concerns that TSB has about
safety.
Safety
Action Taken
Some of the safety action taken includes:
MD-11 and aircraft
wiring: The
investigative and regulatory bodies of three countries (Canada, US and
Switzerland), the manufacturer (Boeing) and the airline (then SwissAir)
have taken action to deal with wiring issues in the MD-11 and wiring
in general.
Data Recorders: TSB and NTSB have urged their respective
regulatory bodies to require 2 hours of data recordings (vs. the
current 30 minutes) and to ensure sources of power that are both
independent of each other, and under emergency conditions, independent
of the aircraft.
Insulation Materials:
TSB recommended that regulatory authorities take urgent action to
reduce of eliminate the risk caused by MPET insulation blankets and to
validate all insulation materials against more rigorous standards than
currently exist.
In-Flight Firefighting:
TSB made five recommendations to ensure a more aggressive approach to
fighting in-flight fires within the accessible and inaccessible areas
of the fuselage.
In-Flight Entertainment Systems: TSB recommended a review of
the STC process that approved the system in question. This was
followed up by the FAA, FOCA and SwissAir, the result being the system
in question is no longer approved and is no longer in service.
Circuit Breaker Reset
Philosophy:
TSB noted that a single philosophy on the use of circuit breakers has
yet to emerge.
Standby Instrumentation:
TSB issued two advisories addressing inadequacies in the requirements
for standby instruments, noted the lack of requirements for standby
communication and navigation equipment, and observed on the lack
adequate training in their use under simulated emergency conditions.
Material Flammability
Standards:
TSB recommended revision of standards to preclude any on board
materials that would sustain or propagate fire. TSB also recommended
more rigorous testing of wiring for failure mode as a potential
ignition source or which could exacerbate a fire already in progress.
Safety
Action Required
TSB is still calling for
safety actions in five areas:
Thermal Acoustic Insulation Materials: Regulatory authorities
should develop a more rigorous test regime to prevent the
certification of any materials that could sustain or propagate fire.
Interpretation of Materials Flammability Test Results: The TSB
recommended that regulatory authorities take action to ensure accurate
and consistent interpretation of requirements for flammability.
IFEN STC: TSB recommends that every IFEN system installed under
STC should be reviewed for emergency load shedding.
Circuit Breaker Reset
Philosophy:
TSB recommends that regulatory authorities establish standards for
resetting circuit breakers.
Accident Investigation Issues: TSB recommends regulatory
authorities take measures to improve the intelligibility of Cockpit
voice recordings, implying that pilot boom microphones would provide
more useful data than cockpit area microphones. The Board also
recommended that Quick Access Recorders, used by many airlines to
maintain quality standards, and which often capture many additional
parameters not captured by the FDR, be fed to the FDR, where they are
more likely to survive a catastrophic event. Finally, the board
recommended that regulatory authorities develop harmonized
requirements for cockpit image recording, that could help
investigators actually see what was happening in the cockpit, rather
than attempting to reconstruct it from FDR and CVR data.
Safety
Concerns
TSB identified 10 areas
of continuing safety concern. Those not covered here in the interest
of space have been mentioned elsewhere in this article. Many
expressed frustration at the slow pace of progress on remedial action.
In-Flight Firefighting
Measures:
The TSB is concerned that “there was a lack of awareness in the
industry about the potential seriousness of odor and smoke events.”
Similarly, the Board “remains concerned with the pace of progress
in mandating that all aircraft crews have a comprehensive firefighting
plan that starts with the assumption that any smoke situation must be
considered to be an out of control fire until proven otherwise, and
that an immediate response based on that assumption is required.”
Aircraft Fire-Hardening: The TSB would like to see a more
timely response to the need to rid aircraft of flammable materials and
“disagrees that the eventual reduction or elimination of flammable
materials and anticipated technological advances adequately deal with
the near-term risk. Therefore, the Board is concerned that regulatory
authorities have not taken sufficient action to mitigate the risks
identified in the TSB’s [previous recommendations in this area.]”
Aircraft Wiring:
TSB remains concerned about the inadequacy of flammability testing
requirements for aircraft wire and the limitation of the current
Federal Aviation Regulation, FAR 25.1353(b), concerning the
installation of wire, which in the Board’s opinion, needs to be
resolved. The TSB remains concerned that, the in-service performance
of ETFE wire (Tefzel®) may not be fully known. The ropensity for
certain types of wiring insulation to wet or dry arc-track needs to be
addressed.
Contamination:
The TSB is concerned that the role of contamination (dust, lint, swarf
etc.) in propagating fires is not well enough understood.
Arc-Fault Circuit
Breakers:
Although the TSB felt such devices would provide major improvements
over existing circuit breaker technology, they felt that they would
not trip the circuit prior to the ignition of nearby flammable
material.
Role of FAA: The TSB was concerned that some of the FAA’s
authority to approve modifications installed after manufacture had
been delegated without sufficient controls being put in place. It
found that 10% of In Flight Entertainment systems had been designed,
installed and certified without a means for the flight crew to isolate
the IFE without interfering with essential aircraft systems.
Checklist Modifications:
The TSB was concerned that “given a lack of checklist modification
and approval standardization within the airline industry, airline
operators may unknowingly introduce latent unsafe conditions
particularly to emergency checklists.”
5. Interpretation
The report was the
culmination of a staggering task and represents a mind-numbing
attention to an investigative challenge of mind-numbing proportions.
The investigative team, headed by former RCAF pilot Vic Gerden, broke
new ground in investigating the accident.
The many ironies of this
crash (newish aircraft, reputable operator, competent crew, aircraft
bought down, in part by an IFEN that included gambling) were exceeded
only by the sheer tragedy of the event. With 229 lives snuffed out,
and the loss felt by families, friends and businesses, this was a
terrible event. It is perhaps appropriate the TSB engaged so
liberally in speculation on this investigation – to do otherwise would
have dishonored the dead, and insulted the living. The report makes
sweeping recommendations for the manner in which aircraft are
designed, built, maintained, operated and regulated. And rightly so,
as there are still weaknesses in the aviation safety structure, and as
readers of this journal appreciate more than any, weaknesses in the
safety net that come into action when the primary safety devices
fail.
However, it remains to
be seen, indeed, it already seems to be all too evident that the full
remedial force of this investigation will be treated like so many of
its predecessors – given nodding acknowledgement, but insufficient
action. TSB’s frustration, in the safety concern section, is
palpable. It is doubly ironic that in Canada, where there have been
two commissions of inquiry previously (The SR 111 crash continues a
trend in Canada that there is major crash every 10 years. It began in
1963 with the crash of an Air Canada DC-8 near Montreal, the 1970
crash of another Air Canada DC-8 near Toronto, a series of crashes
that lead to the Dubin Commission of Inquiry in 1980 and the crash of
an Air Ontario F28 at Dryden in 1989. Sadly, the recommendations of
the previous crashes wax and wane, such that we still find ourselves
dealing with issues that were supposed to be addressed, once and for
all, as the result of tragedies past.
Without taking too much
umbrage with an otherwise impressive report, it is noted that
passenger oxygen was never deployed: this would only have lasted for
15 minutes in any event and because it mixed oxygen with cabin air,
would not have saved passengers from the toxic effects of smoke. Even
the flight crew oxygen system, which allowed the pilots to select
between normal diluted flow, 100% oxygen or emergency pressure would
only have lasted 119 minutes - less time than might be required for an
emergency diversion under the worst possible circumstances.
There was speculation
right after the crash by both lay and expert commentators that the
crew might have been able to land the aircraft at Halifax had they
been more aggressive in descending and less concerned with following
the lengthy checklist for dealing with smoke of unknown origin (and
with dumping fuel). Assuming the crew could have controlled the
aircraft in the final moments of its flight, the TSB report states
that “considering all of the factors, the SR 111 landing would have
likely required more runway than the 8,800 feet available on Runway 06
at the
Halifax International Airport.”
This would have created
a ground event that would have involved an aircraft landing with a
substantial amount of fuel and exiting the runway - a major challenge
to any ARFF unit.
AFJ readers may find it
interesting that the report makes only passing reference (two
sentences) to the Halifax airport response team. The report states
that “Aircraft Firefighting Services at
Halifax International
Airport met the availability and equipment requirements of the CARs.”
(Canadian Aviation Regulations),” without citing what category the
airport was operating at (Category 8), or observing on the fact that
the CARs do not meet ICAO SARPs for a number of criteria, including
rescue and response time, even though the Halifax Airport is an
internationally listed airport with significant international
traffic. Readers will note from previous articles in AFJ that
Canadian regulations do not require any more than one Fire-Fighter per
vehicle. The report did give the Halifax ERS unit full credit for its
response time, saying “The Aircraft Firefighting Services were
activated at 0120 and, within one minute, the response vehicles were
in place adjacent to the runway of intended landing.” While the
report engages in considerable speculation about the origins and
effects of the in-flight fire, it does not speculate what might have
happened had the aircraft been able to make it to the runway, which
was admittedly improbable, aside from the fact that it would have
likely gone off the end of the runway, under circumstances of
passengers and fuel that would have severely tested Halifax’s
emergency response services.
TSB investigators found
that industry standards for reporting of fleet-wide electrical
anomalies lacked the level of detail that would have helped in the
investigation. This deficiency in the Service Difficulty Reporting
System has subsequently been addressed, at least in part, by the FAA
and the airline industry. The point here is that lack of evidence is
not proof of a lack of a problem.
The report cited
previous studies of in-flight fires were the time between detection
and the aircraft crashing varied from 5 to 35 minutes. In the case of
SwissAir 111, the time was approximately 21 minutes. Ironically, one
of the methods used to detect fire and smoke is by smell, yet, as the
report pointed out, the highly efficient air filters in aircraft
environmental systems can actually delay the detection of fumes.
Prior to the accident, fumes and smoke were often assumed to be
related to the air conditioning vent systems of galley ovens.
Operational policy was not to initiate a diversion but instead to
follow checklists to identify and deal with the problem - which is
exactly what the crew of SwissAir were doing when they ran out of time
and control.
Talking on television,
one of the TSB investigators opined that they felt lucky the aircraft
hit the water – instantly extinguishing the fire and aside from the
catastrophic impact damage, at least preventing further loss of
evidence from fire. Any fire event, be it pre- or post-crash, has the
potential to destroy evidence. It suggests that once all has been
done to save the people affected, the fire fighter’s response – and
that of regulators and engineers, needs to be focused on saving the
evidence from damage.
Let us hope that the
regulators will not succumb to the same illusion as did the blameless
occupants of this flight did: “Out of sight; out of mind.” This
report, dealing with fire in the sky, and its predecessors, dealing
with fire on the ground, ought never to be set aside by the passage of
time. To do so, is to fail to learn from our mistakes, surely the most
certain sign of stupidity.
The full report can be
seen at:
http://www.tsb.gc.ca/en/reports/air/1998/a98h0003/eReport/sr111_200303.pdf
 Tefzel
Wiring Insulation - Its flaws and an Answer
Mike
Murphy
| 
