As anticipated, my semi-grudging sympathy for the British Airways crew that flew from Los Angeles to London sans a full complement of engines generated its share of arched eyebrows and curious e-mails. And by "curious" I mean hostile.
"Are you on drugs?" asked a reader from Minnesota. "What airman in his right mind would recommend flying 5,000 miles with a failed engine?"
The answer to the first question is a categorical no; the second one is a little more complicated. I'm on the record with something less than a recommendation, actually. "An unorthodox decision," I wrote a week ago, "[and] not one I'd endorse myself." Then, yes, my polemical kicker: "but perhaps not altogether stupid or dangerous, considering the Boeing 747's capabilities on three, or even two, engines."
It's arguable whether or not progressive thinking -- in the sense of things more permissive and freewheeling -- has a useful role in air safety, but now and then a little devil's advocacy is, I think, useful and refreshing.
To wit: a Boeing 747 can fly just fine with an expired motor. Certification rules require the jet -- up to 875,000 pounds of it at maximum takeoff weight -- to lift from the runway, accelerate, climb and avoid off-airport obstructions. All of that after losing an engine at the most critical point of the takeoff roll -- so called "decision speed" or, in the parlance, V1. Penalties apply later on, at upper-altitude cruise. As we saw last week, flight plans are reworked to account for changes in fuel consumption and range.
With two engines gone, things are less predictable, depending on the phase of flight. With powerplants paired beneath each wing, the worst-case predicament for a 747 would be heavy gross weight, low speed (like that during initial climb), and a dual failure on the same side, the asymmetrical thrust (and drag) vectors making a tough situation worse. That's exactly what happened to an El Al 747 freighter in October 1992, shortly after takeoff from Amsterdam. The crew lost control and the plane slammed into an apartment complex, killing 45 people (the exact death toll is disputed, the building having housed large numbers of undocumented immigrants).
During cruise, on the other hand, any double failure would be more easily manageable, provided there's room for a "drift-down" out of the rarefied high-altitude air to the lower, denser conditions needed -- depending on weight -- to maintain level flight. Tricky over the Himalayas but relatively simple over the ocean or most terra firma. The plane would not, by any stretch, be sent plummeting to its doom.
In El Al's case, undetected corrosion of an attachment bolt caused the right inboard engine to shear entirely from the wing, separating in such a way that it took the outboard engine with it. History shows that most "double engine failures" are more accurately the result of structural failure, volcanic ash ingestion, or other extenuating misfortunes.
(Do such asterisks include pilots "having a little fun" while ferrying an otherwise empty regional jet to a maintenance facility for repairs? Potentially. Investigators have yet to determine exactly what went wrong in that incident, and according to the CRJ's operating manual, the engines should not have quit at 41,000 feet. Catch is, the pilots knew something was wrong with the bleed air system, which draws from the engines, and taking the plane to the edge of its performance envelope was, under the circumstances, probably a bad idea.)
Scary, but the likelihood of truly spontaneous multiple engine failure is astronomically low. Off the top of my head I cannot recall a single incident of its ever happening to a jetliner.
On that note, let me clarify something. Last week, in attempting to make a similar point, I wrote of "the extreme improbability of a second failure." This phrasing was poor, and reinforces the gambler's fallacy that one engine's quitting reduces the likelihood of a second. Each motor, on its own, shares the identical likelihood of malfunction. The extreme improbability I meant to imply is that of two failures occurring on a given flight.
On the heels of the British Airways incident, critics of the airline's choice to keep flying have focused on a scenario whereby another of the plane's Rolls-Royces gives up the ghost somewhere en route to London. "Well, what if they'd been out over the Atlantic," asks a reader, "and suddenly another one dies?"
Not to be flip, but you're about as likely to receive a complimentary onboard pedicure from the captain himself as you are to experience two unrelated engine failures on the same trip. And even so, the 747 would simply have kept flying.
Some critics took to playing the nationality card, citing the more rigorous regulatory standard demanded of U.S. carriers per order of the Federal Aviation Regulations (FARs). Though airline policy can, and often does, supersede the government rulebook, let's take a look at exactly what it says. (Bear with me on this, but also know that what follows is one of the more pleasantly comprehensible snippets from the FARs, the majority of which may as well be written in Aramaic.)
FAR 121.565(b). If not more than one engine of an airplane that has three or more engines fails, the pilot in command may proceed to an airport that he selects if, after considering the following, he decides that proceeding to that airport is as safe as landing at the next suitable airport:
1. The nature of the malfunction and the possible mechanical difficulties that may occur if flight is continued.
2. The altitude, weight, and usable fuel at the time of engine stoppage.
3. The weather conditions en route and at possible landing points.
4. The air traffic congestion.
5. The kind of terrain.
6. His familiarity with the airport to be used.
Not having seen an official dossier of flight 268, I nonetheless suspect British Airways could present a convincing argument that, however beholden to its own, differently worded strictures, it effectively considered each of items 1-6 above before proceeding from North America to Europe.
What of the semantics of those two words in sentence 1: "as safe"? "This rule doesn't imply you can head out over an ocean with a dead engine," contends an e-mailer. "That wouldn't be 'as safe' as landing at the nearest suitable airport."
No, it wouldn't. But four engines will always be safer than three. Just as eight would always be safer than seven; 16 safer than 15. But what, then, would be the point of this exemption, which goes so far as to include the open-ended allowance, "to an airport that he selects"? And, if your criterion is based on the infinitesimal likelihood of dual failure, there'd be no two-engine planes crossing the oceans, when in fact most commercial aircraft plying the North Atlantic are twin-powered 767s, A330s, and 777s.
Also, I dislike invoking the scary-sounding connotations of "ocean." As discussed in this space last time, these rules have nothing to do with water and everything to do with distance/time to suitable alternative airports, and plenty of those exist across Canada and across the North Atlantic rim, from Gander to Goose Bay to Keflavik to Glasgow.
Thus, with a little open-mindedness, flight 268's transatlantic hobble might not seem worthy of the skewering it has received from press, pundits and pilots. So why the outrage? Why are so few people willing to endorse what appears to have been a comfortably low-risk wager? The Federal Aviation Administration, authors of that fussily outlined proviso above, is pressing ahead to formally charge British Airways under its "careless and reckless operation" clause.
Some professional flyers chime in:
"Our airline does not permit what British Airways did. And no pilot I know, even if granted the option, would have considered it." -- Doug, Airbus A320 first officer
"I might have made the crossing if the weather was good, but my preference would have been to land for repairs. That's the most conservative position and the general public would have little room to criticize." -- Mark, Boeing 767 first officer
"It was egregious" -- Michael, Airbus A320 first officer
"Had it been just a precautionary shutdown, I'd see no problem. However, tower controllers reported flames and sparks coming from the affected engine on liftoff. If an engine is emitting sparks, all bets are off. What if there was an undetected fire still burning? What if the discharge caused structural damage not visible from the cockpit?" -- Jim, Boeing 767 captain
It's that final quote that gets at the pay dirt. Keeping Jim's words in mind, click here. What's this, B.A. does it again, with the identical ship?
Well, not quite. The two incidents, which uncannily involve the same aircraft, look to be exceptionally similar on the surface: a belly-up turbofan and the subsequent choice to press on for London, still many hours away. (Note also the Associated Press's bogus mention of "headwinds" in recounting the earlier story. To repeat from last week's column, B.A. 268 did not encounter headwinds; it encountered weaker-than-forecast tailwinds.) In truth, circumstances were vastly different. The Singapore-London event was a midflight shutdown brought on by a comparatively benign oil pressure problem. Compare and contrast to a spark-spewing seizure even before the landing gear is retracted. That's peanuts and pretzels, if you will, to a large extent.
(Since we're doing this, how about a glimpse at the culprit 747. G-BNLG, manufactured in 1990, is shown here sporting the "Whale Rider" scheme from B.A.'s former World Images series of tails.)
Flight 268 was, for all intents and purposes, a disabled plane from the moment it left the runway, and the judgment of all involved parties -- company and crew -- was flawed. Not reckless or crazy, but needlessly unsound. You don't take a disabled plane on an 11-hour haul to England. You just don't.
That's the pilot talking, not the statistician, FAA lawyer, oddsmaker or devil's advocate. And in calling out this seat-of-the-pants verdict, he's reminded of that old bit about how to define pornography: You know it when you see it.
- - - - - - - - - - - -
Do you have questions for Salon's aviation expert? Send them to AskThePilot and look for answers in a future column.