International Aviation Safety Association

The Story So Far

The Swissair MD11 accident has opened a lot of eyes about many different aspects of coping with fire in the cockpit. Of course the investigation is as yet incomplete but sufficient is known for us to make the following reasonable assumptions:

a. The pilots were not sufficiently alarmed at the outset for them to try and rush to a landing. In part this is because of a suspension of belief ("This can’t be happening to me"). True fear (and its urgent motivation) is lacking initially because large electrical fires always start off as smaller, seemingly innocuous ones. They get to be bigger catastrophic life-threatening fires because, under the current checklist philosophies (and switchology), the power never comes off the wires. This lack of alarm can also be attributed to a "society of success" attitude that prevails in simulator training. That optimistic doctrine of positivism would have it that, as long as a crew reacts as per "the book" with the correct CRM approach, all will come right in the end. Unfortunately real life is not like that. The current protractedly optimistic trouble-shooting smoke checklist allows an electrical fire to develop (in real life) simply because the power doesn’t come off the wires. In the simulator it’s not realistically portrayed as such a potentially cataclysmic circumstance so pilots are duped into a mindset that has the smoke-in-the-cockpit drill aligned with the straightforward exercises such as an engine fire or turbine failure. In a "kaptonized" aircraft the attitudinally correct approach should be that "all may well be lost if the power doesn’t come off the wires - pronto". The Canadian TSB has confirmed (on 11 Jan 99) that arcing was found on sr111 cockpit Kapton wiring.

b. The MD11 (and probably other airliners) sets the "airconditioning smoke" checklist first (possibly because it’s not going to take power off the busses and disrupt meal service). The fact remains that this is a lengthy checklist and the electrical fire is still being permitted to develop unchecked. The aircon checklist asks four times: "has the smoke begun to reduce" and the pilots must wait to consult the cabin attendants about this. At the end of this checklist the conclusion is that the smoke is not of aircon origin and says that the "Smoke & Fumes of Unknown Origin" checklist should be started. Obviously these two checklists are in the wrong order because aircon smoke (from oil-contaminated bleed air) is just not going to kill (or incapacitate) you – but interim developing electrical system fires will.

c. The MD11 Smoke/Elec/Air Switch has four positions that de-select (and then re-select) a third of the aircon and gens and busses at a time. The checklist calls for a pregnant pause between selections in order to check whether the benign configuration has been yet reached. There is no mention of what to do if a benign configuration is never found. During this checklist control must pass between the pilots as one or the other will lose their flight instruments. The radios will also lose power and the pilots must remember (but probably won’t) to switch their current comms frequency to each successive "live set". The DFDR and CVR are also powered down at certain stages. It may be the case that the battery (and GEN reset ability) is insufficiently protected from a dead short under the MD11’s electrical system. An induced fault in one or both of the epicentric bus-tie sensing relay switches can leave the backup instrumentation vulnerable to the original electrical fault. At some stage during the checklist the #2 tail-engine is liable to flame-out because of power coming off its pumps (and being too high-set to gravity feed). This would tend to (audibly and visually) Xmas-tree the cockpit and distract the pilots (as well as robbing the system of one of two Generators on line). An electrical system short may then be sufficient to trip the remaining generator and leave the battery liable to overload (and/or Batt CB trip).

d. In night or IMC conditions the critical backup attitude indicator is mounted centrally (low and forward) on the centre console – not really an ideal positioning for either pilot’s scan.

e. Emergency flood-lighting is still aircraft battery-powered and vulnerable to a total electrics failure. A fore-head-mounted (atop full-face smoke-mask) focussable light with an integral battery would be preferred to cockpit flood lights that tend to reflect off flight-deck screens and windows.

f. A third man (the old Flight-Engineer) with good systems knowledge and a role in the checklist would have been invaluable. In a two-man crew there is evident overloading (see Annex U). You need a third man to be able to send a technical flightcrew-member down the back on a portable oxy set.(to check and report or fight fire). The MD11 checklist prohibits a crew-member from leaving his seat (and oxy set).see annex L and Annex M plus http://www.geocities.com/Eureka/Concourse/7349/landings3.html

g. Halon or BCF hand-held fire-extinguishers probably wouldn’t make a lot of difference if an elec fire was still powered. But the question must still be asked: "who (anyway) has the time to use it in a two man-crew that is locked to their seats by the umbilical of their oxy systems?". A flight-deck nitrogen inerting system may well be worthwhile considering (see Annex M).

h. The very unwise plumbing of the Swissair Inflight Entertainment System (IFE) into one of the cockpit busses (AC2), vice an ancillary cabin bus, was obviously done because the high current-drawing IFE would otherwise have necessitated a complete revamp of the electrical distribution system (i.e. creation of a Cabin 2 bus). If that had been the case, the Smk&Fumes Cklist would then have killed any IFE-stoked fire at step one (Cabin P/B OFF).

i. The FAA’s obligation (under the FAR) to resolve the dense continuous smoke in the cockpit problem continues to be studiously ignored. The EVAS equipment is one way to go (Annex E and http://www.raytheon.com/rac/rapid/evas.htm ). The Impaired Vision Flight Safety Device (incorporating a full-face smoke-mask) proposal at Annex T is another. Kapton incidents have happened since sr111, luckily with no loss of life….but it is only a matter of time (Annex V and Annex F).

j. Annexes S & M contain sr111 technical commentary and proposed solutions. In particular it is recommended that current Smoke & Fumes checklists be examined closely in light of what is known about Kapton wiring induced electrical system failures and fires. Maintaining power on busses whilst trouble-shooting is simply not a wise move. In fact it is foolishly unsafe to do so. It may not be possible to eliminate Kapton from modern airliners for many years, yet there may be a cheap and practical way to bypass the catastrophic effects of a large-scale flash-over in a (continuously powered) failed Kapton wiring bundle. Annex S and the following Internet site

http://www.geocities.com/Eureka/Concourse/7349/virgin.html

discusses the advantages of a proposed modification that has been termed "the Virgin Bus". It is designedly a TKT wired add-on that would give crews a readily selectable fallback position; a stand-alone "get-you-home" flight essential bus with minimal features that would avoid the undeniable gamble of the present trouble-shooting checklist. Its main feature is that "everybody lives" and this should become a fine selling point for a public that may soon be looking in askance at an airline industry that has suffered its next sr111. The following Internet sites are also dedicated to constructive debate on aircraft wiring and sr111 in particular:

http://members.aol.com/papcecst

http://www.geocities.com/Eureka/Concourse/7349/index.html

http://www.aviationtoday.com/reports/wiring.htm

Conclusion

Neither Airline X’s safety reputation nor its superior maintenance can protect it against the depredations of flawed wiring insulation. A slavish adherence to the long-held doctrine that FAA’s Bulletins and Directives are sacrosanct would not be wise for Airline X. The US FAA and the manufacturer are set on a path that has been decreed by their vested interests. These interests are demonstrably contrary to the safety of the travelling public. As long as no US airline is involved in a fatal Kapton accident the FAA policy of denial will run its course. As you can see from Annex E, foreign airline accidents are of no consequence in determining FAA agendas. Military attitudes (in banning Kapton) are also claimed to be irrelevant. The lessons of the post-Valujet shake-out of the FAA hierarchy have been set aside and FAA loyalties have been hopelessly compromised. Airline X’s best interests must be determined by itself in light of both what Airline X now knows and the public’s forthcoming knowledge of that. Inaction is not an option.