Ask the Pilot

How did a suicidal passenger get the door of a pressurized plane open? Plus: Can't any pilot fly most planes? And: The autopilot was on during landing?

By Patrick Smith

Published May 1, 2009 10:15AM (EDT)

Earlier this month over northern Canada, a distressed passenger opened the door of a plane during flight and leaped to his death. In a column some time ago you claimed that an airplane's doors cannot, in fact, be opened by a passenger. If that is true, how was he able to accomplish this?

As a rule, a passenger cannot open the door of a pressurized commercial airliner in flight. This includes regional jets and most commercial turboprops. It can't be done for the simple reason that the outward-acting force of cabin pressurization won't allow it. In addition to allowing you to breathe, pressurization helps keep aircraft doors firmly in place, much the way water helps keep a drain plug secure.

Almost all aircraft exits open inward. Some retract upward into the ceiling; others swing outward; but almost always they open inward first and not even the most musclebound human can overcome the force holding them shut. At a typical cruising altitude, up to eight pounds of pressure are pushing against every square inch of interior fuselage. That's over 1,100 pounds against each square foot of door.

Even at low altitudes, where pressure levels are less, a meager two pounds-per-square-inch differential is more than anyone can displace -- even after six cups of coffee and the frustration that comes with sitting behind a shrieking baby. While I wouldn't recommend it, unless you enjoy being pummeled, beaten and choke-held by frightened passengers who have never read my column, a person could, conceivably, sit there during flight tugging on a door handle to his or her heart's content. It is not going to open (though you might get a warning light flashing in the cockpit, causing me to spill my Coke Zero). You would need a hydraulic jack, and TSA doesn't allow those.

The incident over Canada involved a Beechcraft King Air 200, a small twin-engined turboprop used mostly for corporate flying. The cabin door of the King Air is comparatively small and light. It uses different opening/locking mechanisms than the doors found on commercial airliners and is subject to less pressurization. It would be difficult, though obviously not impossible, to force this type of door at least partway open.

The plane's size would then allow for a comparatively rapid equalization of pressure, at which point kicking or otherwise forcing it fully open would become easier. The plane's remaining, non-suicidal occupants were lucky the open door did not tear away part of the fuselage skin or separate entirely, striking the tail or horizontal stabilizer.

And speaking of King Airs ...

A couple of weeks ago the lone pilot of a King Air died during flight. A passenger, who was also a private pilot with 130 flying hours, took the controls and landed safely. In interviews, he described how he had no idea how to fly the King Air. He needed to be told how to deactivate the autopilot and configure the plane for landing, and at first he could not even operate the radio. Granted, a King Air is more complicated than a small Cessna, but aren't the basics essentially the same? Aren't the lion's share of controls and principles similar?

This is a tricky thing to explain, but going from 130 hours in a Cessna to landing a King Air is a huge leap. I'm not surprised the guy landed the plane safely, but had he crashed the thing, well, that wouldn't have surprised me either.

What sound like minor tasks would be serious challenges to a novice. "Configuring the plane for landing" seems simple enough, but actually it's a complicated ballet coordinating the deployment of flaps and landing gear at various intervals depending on speed and altitude. Deactivating the autopilot? If you've never used or seen an autopilot, how would you know how to turn one off, let alone operate its numerous functions?

Not to mention, the touchdown itself would have been completely different from anything this pilot had experienced -- the speed, the control sensations, the visual perspective. All of it. Sure, all airplanes share vestigial similarities, but operationally they can be very, very different. The devil is in the details.

A similar incident occurred here in Massachusetts several years ago. When the lone pilot of a Cape Air Cessna 402 became incapacitated, a student pilot took over and was able to perform a safe emergency landing. The 402 is a piston-powered twin with seating for 10.

Meanwhile, I stand by my comment about the chances of a non-pilot -- that is, an individual lacking formal flight training -- safely landing a commercial airplane. Despite what you may have seen in contrived, made-for-TV movies, it would be all but impossible. The nonpilot has about as much chance of landing a 737 as somebody without medical training performing successful brain surgery.

On my most recent flight, our takeoff was delayed because of a thrust reverser  warning light. What would have happened if the indicator light had failed to illuminate? Would we have been able to stop on landing?

The surge of power heard just after touchdown is the sound of a plane's engines being powered into reverse. Engine thrust is redirected by the deployment of deflectors. It's not a true, 180-degree redirection, but more of an acutely angled, semi-forward vector like the effect of blowing into your cupped hand. If you're seated with a view of the engines, you can see this quite clearly as the cowling slides open. Once the engine deflectors are positioned, which takes a second or two, engine power is increased (though only so far; full reverse thrust is only a fraction of available forward thrust).

Reverse thrust is helpful, but planes can easily land and decelerate without using it. The brakes and spoilers (drag-inducing panels atop the wing) provide most of the stopping power; reversers are mostly a helping hand. If, during a landing, you hear the pilots applying lots of reverse power, it's probably because they are trying to make a particular runway turnoff point to expedite taxi time, not because they are running out of room. Runway length must always be sufficient, taking weight, weather and surface conditions into account, and this data is calculated without the use of reverse. Whatever help it gives you is a bonus.

Proper and symmetric reverser deployment is something pilots look for on every landing. On touchdown, whichever pilot is not at the controls will watch for the correct indications and say out loud, "Reversers deployed" or "Two (or three, or four) in reverse" --- something to that effect. Should one engine's reverser not deploy for some reason, then yes, there will be some asymmetrical torque to deal with, but nothing the pilots shouldn't quickly recognize and correct for.

I know what you're thinking: What about reversing the engines in flight? There have been one or two aircraft over the years certified for in-flight reversing, but this is definitely the exception. No, the plane will not fly backwards, but it will descend very, very rapidly. And should any sort of asymmetry occur, it's liable to come dropping out of the sky like an anvil.

Modern planes have apparatus that prohibits inadvertent reversal when not on the ground. In 1991 a Boeing 767 operated by Lauda Air, an Austrian charter company, suffered an uncommanded reversal of its left engine shortly after takeoff from Bangkok. The left wing separated, and the airplane crashed into the Thai jungle, killing 223 people. Boeing later redesigned the 767's electro-hydraulic reverser system.

While we were at the gate getting ready to depart, one of the pilots got on the intercom and told us that they were having trouble "getting enough air" to start the engines. What was he talking about? I noticed we were hooked up to a tube coming from what looked like a box on wheels. Was that some kind of air compressor?

Jet engines require high-pressure air in order to start. It rotates the internal compressors to a certain RPM, at which point fuel is introduced. Combustion takes it from there. Usually the air is ducted from the auxiliary power unit (APU), a small turbine engine that provides supplemental power while the main engines are shut down. If the APU isn't working, as seems to have been the case as described above, air can be plumbed in from a ground source -- usually housed on a truck or so-called "air cart." (Not to be confused with the flexible yellow hoses used to deliver supplemental hot/cold air conditioning while parked at the gate.)

Once running, the first engine is able to supply air for starting of the second -- or third, or fourth -- engine. An engine that fails in flight can sometimes be re-started, depending on the nature of the failure. This is possible even without help from an APU or a second engine, as the rotational force provided by wind resistance is usually enough.

Picture a jet engine's anatomy as a back-to-back assembly of geared, rotating discs -- compressors and turbines. Air is pulled in and directed through the spinning compressors. It's squeezed tightly, mixed with vaporized kerosene and ignited. The combusted gases then come roaring out the back.

Before they're expelled, a series of rotating turbines absorbs some of the energy. The turbines power the compressors and the large fan at the front of the nacelle. Older engines derived almost all of their thrust directly from the hot exploding gases. On modern ones, that big forward fan does most of the work, and you can think of a jet as a kind of ducted fan, spun by a core of turbines and compressors.

The most powerful motors made by Rolls-Royce, General Electric and Pratt & Whitney generate in excess of 100,000 pounds of thrust. The thrust is tapped to supply the electrical, hydraulic, pressurization and de-icing systems. Hence the term "powerplant."

In a recent explanation of how a wing provides lift, you described the Bernoulli effect, in which the upper part of a wing is curved and the lower portion is flat, resulting in a pressure differential that keeps the plane aloft. Given this required shape, how is it that airplanes can fly upside down, like you see during air shows and demonstration flights? Wouldn't lift act in the opposite direction, forcing the plane downward toward the ground?

A wing creates lift in two ways. Bernoulli's way is one, but in fact it's the less critical of the two. The second is through the simple principle of deflection, akin to sticking your hand out the window of a fast moving car. Tilt it slightly, increasing the so-called angle of attack, and your arm flies. That's much more crude than the subtle magic of the airfoil, but so is flying upside down.

The negative lift from an upside-down Bernoulli is easily offset by the kiting effect. All the pilot needs to do is hold the right angle, deflecting enough air molecules, and the wing stays aloft. Modern airfoils are meticulously sculpted pieces of high-tech art, but even a rectangular airfoil with no curvature whatsoever can be made to fly in any direction -- if much less efficiently.

I was surprised to discover that February's crash of a Turkish Airlines 737 in Amsterdam is being blamed on the autopilot getting a bad reading from a faulty altimeter. It floored me that the autopilot would be on during landing. Is this standard procedure? I always thought the autopilot was only used once the plane reached cruising altitudes.

It is not uncommon for a jetliner's autopilot to remain on during final approach. Of course, "autopilot" is a misunderstood term. As I've reminded people in the past, it is better to think of it as an autoflight system, with different components able to perform different tasks, up to and including an automatic, hands-off landing. And operating this equipment is never as simple as pressing a button. The crew must manage and monitor the different autoflight functions, making numerous inputs. The autopilot "flies" the plane, yes, but the crew tells it what to do and how to do it.

With respect to the Amsterdam crash, it is suspected that a faulty radar altimeter -- a device that measures exact height over the terrain -- fed incorrect data to the autopilot and autothrottle, which were still engaged at the point of the accident. The autothrottle is a system that adjusts engine power to maintain a preset speed and/or power setting. Although the plane was still at several hundred feet, the autoflight logic believed it was over the runway, and engine power was throttled back for what it thought would be an imminent touchdown. However, there was still adequate time for the crew -- two pilots and a trainee were in the cockpit -- to recognize the errant power reduction and subsequent speed loss, switch off the autothrottle and advance thrust manually. Why they didn't do this is the million-dollar question, and the focus of the investigation.

- - - - -

An invitation from the author to readers in the Boston area:

This Saturday and Sunday, as part of the 11th annual Somerville Open Studios event, my apartment will be open to visitors from 12 noon until 6 p.m. Inside, I will be showcasing the best of my travel photography collection, with more than 50 pictures on display. You may have already seen these pictures via my online archives, but this once-in-a-lifetime opportunity allows you to view them in person -- with a complimentary glass of wine from Trader Joe's.

As an added bonus, visitors get a rare glimpse at the inside of my undersized and crumbling apartment. Even if you're jealous of my travel experiences, you are bound to come away feeling better about your own life while pitying mine. What a treat!

Where: ATP Headquarters, Somerville, near Davis Square
When: May 2 and 3, noon to 6 p.m.
What: Travel photos, cheap wine, schadenfreude

Please e-mail for directions.

Patrick Smith

Patrick Smith is an airline pilot.

MORE FROM Patrick SmithLIKE Patrick Smith

Related Topics ------------------------------------------

Air Travel Ask The Pilot Business