IN THE COCKPIT
When flight-crews detect smoke in the cockpit their main concern has always been to locate the cause and isolate it by switching off the affected piece of equipment or pulling its associated circuit-breaker. In some aircraft this traditional checklist will have flight-crews turning off electrical distribution buses, avionics buses, generators, inverters and individual systems - one by one. This can be very time-consuming. The only way in which they can tell that they have been successful is if the smoke starts to subside. Realistically that is very hard to tell. In the MD-11 the flight-crews rotate a 4 position smoke/elec/air switch through selections that disconnect and reconnect electrical buses, generators, bleed-air sources and air-conditioning packs, looking for a benign combination. The motivation is always that it is assumed that a particular item has failed [malignantly], that it must be identified and power to it cut. Wiring faults are not factored into this procedure.
It has been claimed that the checklist procedure that has been traditionally used for smoke-in-the-cockpit is very inappropriate in a Kapton wired aircraft. Electrical fires have a distinctive odor. The only sensible answer to a Kapton caused fire is to cut all the power as soon as you are aware of what is going on. Failure to do that will mean that an insulation fire can spread and take hold as the arc-wire tracking phenomenon "runs" the wiring looms. In an all-electric jet that is surely not an option, because there would be no flight control for instance. And there is the rub: it is a design consideration that just was not allowed for. The MD-11 checklist procedure meant that the faulty wiring was always being powered. That is a guaranteed formula for a Kapton triggered insulation blanket fire. It is astonishing that Airline X flight-deck crew have no knowledge about Kapton. It is therefore quite likely that they would simply turn the rotary switch or fatally reset a circuit-breaker.
A Possible Solution.
Is there another solution? If you must "kill" the electrical system in order to contain and perhaps extinguish the fire you must have another system with which you replace it. Ideas have been launched about designing and installing a simple Flight Essential Bus [FEB]. It is envisioned as a "get-you-home" minimally basic AC & DC bus that is immediately selectable when a smoke-in-the-cockpit situation arises. The act of turning it on secures the other normal Kapton-wired busses. Its main characteristic is that it is a "virgin" bus - it is normally unpowered. It can be run from any generator, its wire is a stand-alone TKT loom that runs clear of any Kapton harnesses and avoids any commonality. It can provide power [and control power] to only the basic standby flight instruments and engine instrumentation, fuel and hydraulic pumps, lights, communications and avionics that are required for a survival diversion. It is a "get-you-home" configuration that eliminates the sort of inflammatory, crazily optimistic, time-consuming, trouble-shooting checklist that did not work for SR 111. The passengers will not get any hot food or entertainment, but they will get a voucher at their destination. It is emphasized that once you take the power off the faulty Kapton looms the fires are not self-sustaining and will probably fizzle out or be easily extinguished. Leave the power on and they will spread, take hold, generate toxic smoke and damage other vital systems.
An FEB does not require a big modification. It would not be difficult to incorporate it during heavy maintenance, when the aircraft is stripped and the insulation blankets are coming out anyway. It might add a week or two to its scheduled heavy servicing. Each type would need a different design but the modification kits could be off-the-shelf components [switches, relays, junction boxes, etc.]. The design and incorporation costs could come out of litigation between the aircraft manufacturers, FAA, airlines and wire-makers. Each entity must bear some responsibility for the current situation. It only remains for the courts to apportion blame and the insurance companies to force the issue. It is hardly likely that the outcome of the SR 111 investigation will be a recommendation like this. The final results may be inconclusive. It is a difficult investigation because of the loss of Digital Flight Data ecorder data and the aircrafts disintegration, and the fact that the investigators will never achieve a recovery ratio of 100 per cent of the debris. No one should feel surprised if the final outcome will be "undetermined cause", or that it will be generally classified as a "fire of unknown origin suspected to be due to an electrical fault". That way it is easier for all in the industry if it is a "once-off" accident. It will be an actuarial blip rather than being an otherwise defining event, were blame to be laid at the manufacturers door. Of course, Swissair (or KLM) might, in the medium term, quietly sell off all their MD-11s to a third-world operator or cargo-carrier at a bargain price, buy some Airbus A330s and put it all behind them. Everything may just quieten down until the next one.
There are other disasters to be hinted. It is now generally accepted that TWA 800s center tank was ignited by a spark induced in faulty wiring. Few people accept that Valujet 592s cause was the illegally stowed oxygen canisters because of the aircrafts lengthy recent history of electrical problems. The Silkair crash suffered a very familiar loss of DFDR and CVR power, a few minutes before it left altitude and later hit the water at very high speed, but investigators are primarily looking at suicide theories or the B737s history of rudder hard-overs. The Silkair facts fit more neatly into a wiring-caused scenario. It is possible that the Silkair crew were simply trying to resolve an electrical difficulty [that was not urgent or serious enough to warrant an R/T call] when the situation drastically compounded. With these types of crashes you have got to draw conclusions because there are few firm facts to be had from the disintegrated wreckage. Some of these hypotheses may be well wide of the mark however. It will take a few more SR 111 type accidents before the cause is conclusively pinned down. It is not all Kapton but wiring maybe a common denominator.
A checklist that guided the SR 111 flight-crew in the minutes before the crash contained guidelines with disastrous effect. The critical smoke/elec/air switch, described by TSB chief investigator Vic Gerden, could have restored power to damaged wires aboard the MD-11 and thus caused a massive electrical failure, said Julia Johns of Lectromechanical Design Co. "We would not recommend it", said Johns, speaking for the electrical research and design company outside Washington, D.C., known as Lectromec. Johns said reintroducing an electrical current to wires wrapped in the kind of insulation aboard the Swissair jet might be standard practice but it was exactly the wrong thing to do in smoke situations. Aromatic polyimide tape insulation was particularly susceptible, she said, to deterioration from heat and humidity that can lead to arcing, or lightning-like leaks of electricity from wire to wire. " If you continue the source of energy, it will force the sparking and disintegration of the wire", Johns said. ``I would say it would not be prudent to reintroduce electricity. I would say extremely emphatically: Don't do that".
Edward Block was even more emphatic. "They essentially seal their fate by order or instruction", if they followed the plane's checklist, Block said "It's like writing your own death sentence - and that of everybody else on board".
Gerden said instrument displays were affected as pilots tried to trace cockpit smoke. He said the automatic pilot disconnected and other operations shut down "in fairly rapid succession over the next 90 seconds" before two flight recorders, using separate power sources, stopped working. Gerden confirmed the MD-11 had the so-called smoke/elec switch, a rotary selector designed to trace smoke aboard an aircraft. In some planes, pilots simply re-engage tripped circuit-breakers. "It's similar but it does it in a larger scale way", said Gerden. "When you go to one position, it turns off a third of the electrics in the aircraft and one of the air-conditioning packs. You go to the next position and it turns off a different third of the electrics and a different air-conditioning pack. And similarly for the third position. But each third of a system is re-engaged - powered back up - as the dial is rotated", the safety board's Jim Harris added.
Authorities investigating the crash have not yet determined how far into the checklist the Swiss pilots were before the plane plowed into the Atlantic Ocean off Peggy's Cove. The pilots aboard SR 111 were well trained and highly experienced. The captain was an MD-11 instructor himself. Circuit-breakers and how to deal with them are part of the ground training. There are manuals making perfectly clear that circuit-breakers must never be [re]activated "just like that". If, under the circumstances, this flight-crew would have [re]activated the circuit-breaker indeed, they would no doubt have sealed their fate.
The Two Alternatives
A PREFERRED COCKPIT SCENARIO?
We dont have to be visionary when it comes to applying our imagination to what happened in the Swissair accident. We should have no trouble conjuring up two entirely separate theories: of what did transpire - and what might have been [in different circumstances]. In the first instance we see the innocent low-key beginning of the happenstance, the advice to ATC and cabin crew, the follow-up R/T, the descent checklist and then its rapid escalation through donning of masks and goggles and serious checklisting. Then come the profoundly disturbing staccato images - an atmosphere of great urgency [anxiety, concern but no fear], awareness of the deteriorating environment and restricted peripheral vision, unexpected autopilot disconnect, frustration about having to wait until over-water before commencing the dump, denser smoke, upgrading the emergency, deploying the ADG, fuel starvation flame-out of number two tail engine with many resulting warnings and cautions, sudden cockpit darkness, CRT's blank out, internal floodlight reflections off the cockpit windows, misting goggles, instrument power failures and an alarming realization that your partial panel attitude-flying instrument scan and control inputs are very haphazard -even chaotic, eyes streaming with tears, the annoying ceaseless audio of uncancelled system alerts and warblers, intercom and R/T mistakes and confusion, ATC distractions, EICAS announcements, unanswered cabin staff queries, a pervasive and growing feeling of consummate helplessness, breathing and vision difficulties; a Flight Attendant invades the cockpit, evidently distraught - then disappears, total uncertainty as to where the checklist is at. Then, as a vaguely discerned finale ..... seeing the collapse of the other pilot with no-one to help, a feeling of immense solitude, the agony of hot molten plastic dripping from above, difficulty in concentrating, total loss of attitude and altitude awareness and eventually, on the threshold of unconsciousness, the sensation and increased noise levels as the aircraft is accelerating under g and entering its terminal unusual attitude - but peaceably, a final ethereal dissociation from recovery [as a task] because of the mental detachment that stems from the asphyxiating ingestion of toxic smoke. A last poignantly dutiful but pointless pull on the control column as the MD-11 tightens into its final graveyard spiral into the waters off the cove. Declaration of a final Mayday and any thought of external succour is totally irrelevant to such a scenario.
In the second apparition we see a three man crew undergoing the same initial conspiracy of circumstance. In this scene the smoke detector alerts them and the captain declares: "activate the smoke checklist". After the memory recall items and immediate action preliminaries of full-face masks, ATC and cabin staff alerting, the next step is dependent upon captains discretion. However, as with the Swissair instance, visible confirmatory smoke and fumes are evident and this is where the outcomes based approach makes the difference: the Captain orders "select the Flight Essential Bus, you have control. Deploy the EVAS" [i.e. the inflated plastic aid to instrument vision]. They feel a thump as the ADG slips down and locks into place in the slipstream and a charging light illuminates on the CIWS. The first officer is now responsible for ATC comms and flying the aircraft - nothing else, so he begins a gradual descent as per his edicts to ATC. Hes declared a Pan, hes squawking emergency and he automatically has priority, so ATC must divert other traffic and clear levels below him. Its called "take control of your destiny", its not an apologetic approach. It's no longer the polite "request" - it's "require". The aircraft is now on its "virgin" bus, previously isolated from the normal electrical system and presumably having 100% integrity because of that. The captain and the flight engineer commence the challenge / reply portion of the checklist actions. It is evident that despite his killing the normal electrics, the smoke build-up is continuing, that there is a self-sustaining conflagration behind one of the overhead panels. He orders: "Call Mayday, Hold 230 knots, commence fuel dumping to 50 kilo and inert the flight deck". The FE pulls a lanyard on the bulkhead-mounted nitrogen canister and grabs the foam extinguisher from its bracket. The FE, in anticipation, has already [per the checklist] shut off both EDCs so that there is no inflow to the cockpit but he doesnt open the outflow valve- so there is only a gradual depressurization. The flight deck door is already closed so, for their purposes, the cockpit is hermetically sealed and the flame-retardant nitrogen-rich atmosphere dampens the fire. In the absence of oxygen, and with no stoking electrics, no materials fire can take hold. The FE then sprays the adhesively congealing and sealing foam spray onto the overhead panel and the smoke emanations cease. It is now up to the captain. After a pause, he considers the danger is past and orders: "Depressurize and vent smoke, continue with the smoke elimination checklist". Flight is continued, masks off, to their divert field in their significantly reduced [but not crippled] configuration. They have no weather radar, airframe de-ice or engine anti-ice - but at the lower altitudes they are not likely to need it. Flight instruments, hydraulics, avionics and comms are sufficient for recovery. There is no point in further tempting fate and trouble-shooting as the Nitrogen and foam operation was a one-shot affair.
A two-man band working fate or a three-man consortium with a game-plan? Given the alarming scope and urgency of the task, vive le difference. It is possible with todays technology to change the outcome of the in-fuselage fire-in-the-air challenge. To date its been a mish-mash apathetic and disorganized approach to resolving an unidentified issue. Im advocating both a CRM [Cockpit Resource Management] and multi-disciplinary, technical, take-charge / outcomes-based methodology. I readily concede that there are more forms of inflight fire than one confined to the cockpit or electrically initiated. However if the cargo dispatchers and security are doing their jobs we are probably talking about the worst case scenario here. Cabin fires outside the cockpit [or wing events for that matter] make it more important that there is a third man with the technical know-how to go aft, look into it, keep the captain properly informed and take action if necessary.
However theres a lot of inertia out there. Many CEOs just have their fingers crossed that the Swissair report will be couched in terms that enables it to be dismissed [like Valujet, TWA 800, Silkair] as just another "one off.... but casualty figures will probably mount as jets get older and electrics continue to be the silent servant but also tragically, sometimes, the swift unseen suffocating savage.
List of Annexes