Eirik: Wasn't there another near-crash a few decades ago that started with the plane doing a roll and then diving straight down? IIRC, it was something like a 727, the pilots saved the plane and landed safly, but then erased the flight data recorder because he felt it wasn't a crash. Speculation was that the flight crew was doing something to test the limits of the plane and did something stupid. Anyone reconzie that?

http://en.wikipedia.org/wiki/TWA_Flight_841_(1979)

The 727 crew was playing with non-approved flap settings at cruise by pulling the circuit breakers at certain flap positions not normally selectable. Rolled twice, but landed safely with substantial airframe damage.

http://en.wikipedia.org/wiki/China_Airlines_Flight_006

747 engine flamed out and the crew didn't follow proper procedures and ended up rolling it and pulling 5G in the recovery. Landed with huge chunks of the elevator missing and other damage.

Click Click D'oh: I kind of assumed they forgot all about that after using it as a convenient excuse to start the crash sequence. Negative Gs and airliners are not friends.

If you're gonna be pedantic, you have to commit.

Anyway, whether or not it's possible to *save* an airliner with a jammed stabilizer by rolling it inverted is entirely open for debate, but there's no question that it's possible to do so and has been tried. Flight's writers pretty clearly based the entire accident sequence off Alaska Airlines 261, the major difference being that the last-ditch attempt to stop the dive by inverting it works in the movie, and for some reason they're able to roll back over again and crash land in a field. Alaska 261, by contrast, made a big, high-speed splash into the Pacific.

I was willing to suspend disbelief in the method of recovery, as the movie as a whole is a far more realistic take on the industry than most people who see it will likely believe.

davidphogan: I'm glad the expert mentioned the absurdity of gliding a commercial jet to a safe landing on the Hudson. If it were in a movie and not real life I'm sure there'd be a bunch of fark experts posting how impossible that would be also.

Airliners have been ditched successfully and moderately successfully before.

Successful:
Pan Am Boeing 337: http://en.wikipedia.org/wiki/Pan_Am_Flight_6
Japan Air Lines Douglas DC-8: http://en.wikipedia.org/wiki/Japan_Airlines_Flight_2

Moderately successful:
Ethiopian Airlines Boeing 767: http://en.wikipedia.org/wiki/Ethiopian_Airlines_Flight_961

As for gliding an airliner, the Canadians have perfected that.
Air Canada Boeing 767 (aka Gilmi Glider): http://en.wikipedia.org/wiki/Air_Canada_Flight_143
Air Transat Airbus A330: http://en.wikipedia.org/wiki/Air_Transat_Flight_236

clear_prop: As for gliding an airliner, the Canadians have perfected that.
Air Canada Boeing 767 (aka Gilmi Glider): http://en.wikipedia.org/wiki/Air_Canada_Flight_143
Air Transat Airbus A330: http://en.wikipedia.org/wiki/Air_Transat_Flight_236

We have a gift (for running out of gas).

Well yeah, if youre crashing and you roll over youll start going up. Basic physics.

Donau: Boeing tests its planes to roll, my great uncle was one of the first commercial pilots to fly the 747, he and a bunch of other PanAM captains went out to were they were testing the 747. They wanted to see if the commercial pilots (many of whom were veteran military aviators) could put the plane through the same paces as the test crew. One of the things they had them do was take the 747 into a roll.

So these planes can roll, but a low altitude roll like that... I dunno. I would think passengers would slip right out of the those shiatty little seatbelts.

Barrel roll vs aileron roll. Barrel roll is 1G, plane doesn't even know it is upside down. Go find the YouTube video of Bob Hoover pouring tea. Aileron roll is negative G, and not recommended for airliners.

Jim DiGriz: clear_prop:Barrel roll vs aileron roll. Barrel roll is 1G, plane doesn't even know it is upside down. Go find the YouTube video of Bob Hoover pouring tea. Aileron roll is negative G, and not recommended for airliners.


An inside (top of the plane points to the center of the barrell) barrel roll would allow for the above. It sounds like the movie is showing an outside (top of th plane points away from the center of the barrell) barrel roll.

Haven't seen the movie, but from the previews it looks like an aileron roll. With the elevator stuck at full nose down, you can't do a barrel roll since it involves pulling up and banking.

Click Click D'oh: No, it's not. Even if you did manage to defy physics and fly the plane upside down using only angle of attack to keep it in the air (and somehow the suck elevators don't make you do outside loops until you stall and crash), there is still no airliner in service with landing gear on the crown of the fuselage so the whole crash and burn ending is still in play. As I have yet to find the switch to momentarily suspend physics in the event of an emergency, that really isn't an issue because the crash and burn will come right when it's expected.We aren't talking about fighter jets or acrobatic aircraft. Airliners don't do the upside down thing very well.

There's no defying of physics occurring there. Every wing is capable of inverted flight - the only question is the degree of efficiency and how much AoA you need to approximate the non-inverted performance of the wing. Whether the aircraft it's attached to can actually roll inverted is also a given - they all can, barring some kind of artificial restrictions on flight envelope like Airbus's fly-by-wire system. So the only real question is whether or not the aircraft the wing is attached to can actually sustain inverted flight. Does it have enough down elevator to maintain sufficient AoA to generate the required lift? Can it's engines operate inverted? Is some other system going to fail as a result of being inverted before any of that matters?

The movie makes the assumption (and nobody here is qualified to say definitively one way or another - it's not tested, listed or easily calculable) that the MD-80 - in a full nose-down trim event, has sufficient elevator authority to generate the required angle of attack to keep the plane in the air while inverted. They also make the assumption that the engines wouldn't survive this (which is a lot easier to guess at than flight control authority - they probably can't last upside down for very long at all), which kind of makes the last question a moot point.

Click Click D'oh: It's been tried, but that doesn't make it possible.

It's physically possible to roll any airplane inverted. Whether it's a sustainable or even recoverable situation once you're there, that depends on the aircraft.

Click Click D'oh: So yeah, it doesn't work in the real world after all.

Never said it could save the aircraft. People who think the crash scene in the movie is totally ridiculous are almost always focusing on the wrong part. There's no underlying physical law that prevents the sequence from happening more or less as advertised from the start of the dive until they're upside down. What doesn't jive is the part of the sequence where they roll the aircraft back over and appear to have solved their nose-down input problem.

Click Click D'oh: Since in order to generate lift using AoA alone, you induce drag that the aircraft could not produce enough thrust to maintain forward speed, would slow down rapidly and the then the whole crashing bit
Click Click D'oh: Elevator authority, although an interesting question, is simply a red herring on your part. There is plenty of authority to say that an MD-80 can not maintain flight on AoA alone independent of (and in fact directly opposed to) wing generated lift.

I get it now. Somebody somewhere along the line taught you that lift generated by the wing is somehow different than lift generated by angle of attack.

That's wrong. There's no such distinction as "wing lift" and "AoA lift", and the things you're confusing them for don't oppose one another when inverted.

Here's why:



This chart shows a symmetrical wing* and the right side is called a coefficient of lift graph. Starting at zero, any increase in angle of attack is met with an increase in lift. This works both ways from zero, since this wing generates a CL (lift constant) of zero when it has zero degrees angle of attack. In the terms you're using, this wing generates absolutely zero lift without angle of attack. At 10° it produces a lift coefficient of 1.2 or so, which gets put into an equation factoring things like wing area, speed and air density into the equation and pumps out a lift value.

Notice, that when the angle of attack moves to -10°, it produces the inverse of it's 10°.. Which means that this wing would produce exactly the same amount of lift whether it's upright or inverted.

Now, this example is for a symmetrical wing, which never produces lift without an angle of attack - aka, no "wing lift" in your terms, I believe.

Here's an example of the AoA/Lift relationship for a non-symmetrical wing (specifically, this is from a business jet - it would be similar enough to a DC-9/MD-80, for which I couldn't easily find a graph):



So the way that you can tell this wing will produce lift at an AoA of zero is that it's CL curve crosses the 0° line with a positive CL - specifically, a CL of about 0.1. This would be more or less realistic for a jet aircraft designed for high speed transportation. That said, in order for this wing to produce anything approaching the required lift to get the aircraft it's attached to off the ground, it will need to include AoA. This wing generates "wing lift", but it amounts to a very small portion of the total lift this wing would be required to produce in actual use. There isn't an aircraft/wing combination cruising around anywhere in the world that doesn't have a positive angle of attack when just flying along. The slower you go, the greater than angle will be. Always.

Note that it is still capable of flying along inverted and generating lift. It requires a slightly higher angle of attack to generate the same CL (what CL is required is determined by the other factors I mentioned above - wing area, speed, air density - but if it needs to generate a CL of 0.5 to fly at 200kts at 10,000 feet, it can do that right side up with an AoA of 3°, or inverted with an AoA of 6°. The wing doesn't care. Reduce the CL required (by having a bigger wing, or flying faster, or being lower where the air is more dense) and you get even smaller AoA values for everything, making inverted flight even easier - from the wing's perspective.

Drag has broadly the same relationship with AoA as lift does, so you're right that the drag would be higher while inverted with a transport category aircraft, but you're wrong when you make the assumption that a plane "can not maintain flight on AoA alone".

tl;dr version: wings don't work the way you think they do.

Click Click D'oh: So wait, you're saying that what's portrayed in the movie isn't completely ridiculous because the first half is plausible... and just ignore that the second part is completely asinine?You work for Mythbusters don't you?

Where'd I say the first bit?