The "miracle on the Hudson": Sullenberger/crew skill and interaction w/ Airbus 320

There are a bunch of good airplane/crew threads going on, so I thought I’d start another one. :wink:

I’ve heard (although I can no longer remember the arguments) that Sullenberger and his crew “only” helped along the innate control systems of the A320 in The Miracle on the Hudson. This sounds asinine beyond words.

Could anyone comment on that comment (or generalizations thereof)?


Who says this? What is meant by ‘helping along the innate control systems’?

I highly doubt there’s a “land plane on river” control system built into airplanes, but I’ll defer to the experts.

I’m no expert, but I can’t imagine a better answer than this Vanity Fair article:

Executive summary: it was an incredibly fortuitous combination of skill, guts and technology that came together.

There was some mention of the “Ditch Switch” in the new Airbuses, but all it did was close various ports (i.e. holes) in the aircraft in case of a water landing, not exactly an autopilot for it. And neither Sullenberger nor his First Officer got around to flipping it before they ditched (had a few other things on their minds), and the plane still stayed afloat.

For what its worth I setup my brother’s MS Flight Sim X (with full CH Yoke, Pedals, and Quadrant) with that exact aircraft type in its exact position, heading & altitude just out of LaGuardia where it hit the birds, plopped myself into the cockpit, killed the throttles, and hit start. And I only made it to the Hudson over the GW Bridge maybe one out of ten tries. My brother, who’s an actual pilot, faired a little better, but it was still not easy. Dead sticking an airliner is like guiding a barge with a canoe paddle…

Perhaps useful to the discussion, here’s a fascinating video that shows a realtime re-enactment:

I was on the edge of my seat.

The pilots who ran out of fuel over the Ontario-Manitoba border managed to land at an old air force base near Gimli. IIRC, it was a 767, and the problem stemmed from converting metric to gallons during fueling. When the engines quit, so did the hydraulics. I recall one analysis where it was mentioned that they practically had to stand on the pedals and haul with all their might to work the controls without power. (Yet, they put down on the runway/dragstrip so well that they split the center guardrail down the middle).

Similarly, there was the case of an aircraft where the centre engine disintegrated over the midwest. The pilots landed using differential engine thrust as they had no working tail controls; unfortunately, on touchdown they hit one wing low, spun and burned but a large number of the passengers survived.

Basically, any time something incredibly serious happens and the pilot manages to make it down, it’s a miracle of experience and training, not a computer program. In the event that I was flying well outside normal parameters, I’m not sure I would want the computer deciding what was best for me.

Look what happened when Air France flight 477 flight computer decided it did not know what was going on over the South Atlantic… it basically shut down and left it up to the pilots to figure things out. (In this case, they did not).

Jaysus. I’ve been in vehicles before that had the power steering quit, and damn near killed myself, I can’t even imagine what that must have been like in a 767.

Not sure if the bit about the hydraulics is accurate, the B767 has a ram air turbine which is an air driven turbine powered that sticks out in the breeze and can power the hydraulics in the case of a double engine failure.

For what it’s worth, I spoke with an airline pilot (an ex-military guy, not sure which service, but a fighter pilot by training) on the week Sullenberger ditched in the Hudson. He said that Sullenberger’s feat was something that many or most airline pilots could have done, but that’s a compliment to the high average skill level of airline pilots, not a slur on the landing itself. His opinion was that Sullenberger made exactly the right decisions in perfect textbook style; not a wildly creative or miraculous landing, but a cool-headed and very technically correct landing. He also said that his first reaction to hearing about the story was “Huh. Good to see that they actually do float.”

If you want answers to specific arguments, you’ll have to provide the arguments. I’ve never heard it suggested before that the plane landed itself in the Hudson.

The most interesting part of this was that he was repeatedly offered multiple runways and had to decide instantly that the only option was a water landing. If he had hesitated, and decided to try and make it back to La Guardia or to the alternate airport across the Hudson he could very easily have lost too much altitude and lost the option for a water landing as well.

There was literally about a 10 second timeframe he had to make that decision, hesitation or dithering would very likely have lost the plane.

I’ve just browsed through some of the accident report. The aeroplane was in alpha protection mode for the final 150 feet of the descent. Alpha protection mode prevents the pilot from stalling by attenuating aft pitch commands at low airspeeds. Basically it appears the captain was trying to raise the nose to reduce the rate of descent and soften the landing but the aeroplane wasn’t letting him because they were close to the stall angle of attack. It’s hard to say whether the aeroplane was saving the day at that point or preventing the captain from using all of the performance available to reduce the impact forces. The protections kicked in when there was still 3.5º of angle of attack to play with. Had they not been available the Captain could have stalled or he could have done a better job than the aeroplane at flying on the edge of the stall.

One point against the captain is that in post accident interviews he stated that the airspeed was a safe margin above the stall, in fact the FDR data showed that it wasn’t, this indicates that he was probably task saturated and had lost some awareness of his airspeed. A point against the aeroplane is that it should have been calling airspeed warnings but these were inhibited by the ground proximity warnings which are designed to over ride the airspeed warnings (hitting a mountain is seen to be worse than getting too slow.)

It is impossible to tell what might have happened had a different aeroplane type been involved. All we know is that the outcome was a very good one, and when you compare it to the Air France accident, any mishandling that may have occurred was minor.

Sure there is - there’s a big button right there marked “DITCHING”. :stuck_out_tongue:

(Yes, that’s just the “ditch switch” referred to by Hail Ants, which didn’t get activated in the Hudson case anyway. And even if it had, the impact ripped several much bigger holes in the fuselage anyway, so it wouldn’t have done much good.)

Good writeup on the “Gimli Glider” on Wiki - and yeah, it was a 767.

I see now that the RAT did deploy but it became less effective at low airspeeds.

I’ve never flown any Airbus product, so I hesitate to comment on this or the Air France mishap lest my comments be given more weight than they deserve. Airbus are truly different airplanes from all others and IMO only folks who fly/have flown them can make truly well-founded comment about the man/machine interface.

I can’t really top what Richard Pearse said in post #12 or appleciders’ friend’s comment in post #10.

Every time I depart LGA now I think about that one & verbalize “Here’s where Sully hit the birds” at about the right altitude / point in space. He truly had only a few seconds to make the irrevocable decision about where to touch down. Essentially he chose to assume the engines were dead *before *they fully quit delivering thrust. Had he chosen to assume otherwise he’d probably have aimed for Tetorboro and ended up in some houses in NJ.

The way I’ve explained it in the past is he was fortunate to be handed a soluble problem. Which he then handled with textbook precision & Hollywood cool.
From reading the NTSB report & his book I conclude the airplane did a pretty good job of making so many noisy alarms that it was all but impossible to think. In other words, to answer the OP, the airplane was hurting, not helping. Subject to my initial caveat about not being an Airbus pilot with first-hand experience on the type.

That excessive distraction in non-standard situations is IMO a real problem with all modern aircraft, and from reading several accident reports seems to be a particularly severe issue with Airbus.

FWIW, his book is a pretty good one. (If the link doesn’t work search for a book titled “Highest duty”)

It explains about the accident in laymen’s terms, and gives an excellent insight into the personality of the Pilot’s pilot. He comes across as rather limited as a personality, but what he has is almost all good-hearted or at least good-intentioned. It’s very much not sensational. All of which fits the pilot phenotype exactly.

They have tried to reenact that situation in simulators but not one flight crew have been able to put the plane down without killing everybody on board. At least not at the time I read or heard about it.

That’s why pilots spend a lot of time in simulators, where they can make such behaviour a reflex, so they don’t have to sit and think about what to do next. A pilot, I once had a beer with, told me that being a flight instructor and simulator supervisor really brought out the nasty in him.

It’s worth noting that pilots do NOT always respond to emergencies with cool-headed technical precision. This week’s release of the transcript of the above-mentioned Air France 447 is a case in point, but there are others.

United Flight 232, July 19, 1989. Here’s a little writeup about it by the pilot himself, Captain Al Haynes.

…such as the commuter plane that crashed in Buffalo in 2009, for almost exactly the same reason. The crew misunderstood weather conditions, the plane slowed to a stall, and rather than diving to gain speed they pulled back on the stick. In both cases the emergency situation was caused by pilot error.

I think the key point is that if 1549 had crashed, nobody would have been saying it was due to crew error.