34
$\begingroup$

I have pretty bad flying phobia and often wonder what would happen if all the engines on the craft went out at the same time. I initially had the anxiety-ridden thought that the hundreds of thousands of pounds of metal would nose-dive and explode into the Earth. Then I read somewhere that the plane would actually glide forward as it descended, making a landing where all passengers survive a possibility. What if you were flying over the ocean versus flat land versus mountains? Has complete engine failure ever happened?

$\endgroup$
  • 16
    $\begingroup$ No amount of logic can defeat a phobia. You are safer in a plane than travelling by car or even walking down the street. $\endgroup$ – JamesRyan Oct 6 '14 at 11:04
  • 4
    $\begingroup$ Somewhat worth mentioning that even a HELICOPTER can safely land with engine failure. (If you're curious, it's called autorotation, you can look it up.) $\endgroup$ – Bassinator Oct 9 '14 at 21:39
  • 1
    $\begingroup$ @JamesRyan, that's not true. Learning about the relevant field and the true risks can be a highly effective way of dealing with phobias. $\endgroup$ – Reid Mar 7 at 18:14
60
$\begingroup$

Air Transat Flight 236 experienced a complete power loss over the Atlantic Ocean in 2001. Yes, all passengers and crew survived after the aircraft glided 75 miles to a runway on the Azores islands.

Even in the event of the loss of all engines, an aircraft can keep its critical electrical systems running thanks to the ram air turbine which allows the crew to maintain control of the aircraft and to communicate with ground crews to determine the best course of action for the situation. This was the case with Air Transat Flight 236.

Moreover, consider US Airways Flight 1549 which experienced a complete power loss after takeoff due to a double bird strike. Even in the situation where the pilot had very few options due to being at such a low altitude and being over a populated area, all passengers and crew survived after ditching in the Hudson River. However, in all fairness, ditching attempts do not normally end well.

In all, complete power loss in a modern airliner is extraordinarily rare. While gathering links for this answer, I ran across the "List of airline flights that required gliding" article on Wikipedia. The short length of the list speaks for itself (4 flights in the last decade).

$\endgroup$
  • 2
    $\begingroup$ Good post! BA9 comes to mind, too, a 747 that lost all engines when it flew into a volcanic ash cloud. $\endgroup$ – Peter Kämpf Oct 4 '14 at 7:30
  • 19
    $\begingroup$ See also the "Gimli Glider" incident. The happy outcome was strongly influenced by the captain's experience as a glider pilot. en.wikipedia.org/wiki/Gimli_Glider $\endgroup$ – Jamie Hanrahan Oct 4 '14 at 10:43
  • 1
    $\begingroup$ I think people would recognize it better if you referred to the "Miracle on the Hudson" rather than "US Airways Flight 1549"... $\endgroup$ – Mehrdad Oct 5 '14 at 20:50
  • 19
    $\begingroup$ @Mehrdad only for US Americans. (And I personally prefer a more secular approach to names. It's not priests who fly planes.) $\endgroup$ – ANeves Oct 6 '14 at 11:32
  • 19
    $\begingroup$ @Mehrdad - Also, calling it a miracle IMO takes something away from the awesome airmanship and skill shown by the pilot. There was no miracle, just damn good piloting and calmness under pressure. $\endgroup$ – Jamiec Oct 7 '14 at 12:53
26
$\begingroup$

The fundamental parameters that determine how survivable a plane crash is include the vertical air speed (relative to the ground); where the plane strikes (ideally, it touches down on its landing gear or, in the worst case, on its belly, scraping against the ground); and how long it takes for rescuers to locate the downed plane and provide assistance.

To keep your vertical air speed at a low enough level that there's a chance for people to survive, you must have flight control surfaces working. That means the wings, the vertical stabilizer or tailplane, and ideally, the ailerons and flaps. These systems do one of two things (some of them do both): they either generate lift (the wings do most of this), or they provide attitude control (roll, pitch, and yaw).

Note that, while the engines are not critical to provide attitude control and they are not required to generate lift, the airplane will steadily lose altitude without the engines generating thrust. An airplane is nominally controllable if the attitude controls are working, and the major providers of lift are doing their job (by not, for example, being torn off by an engine explosion). An airplane that is not controllable is almost always doomed to lose all souls aboard. However, an airplane that is controllable but has no engines can, in many cases, land safely, or with a majority of the lives on board being saved.

Survivability mostly depends on the following factors:

  • When the engines fail, why did they fail? If they failed because fuel is exploding in the fuel tanks or starting a fire that could damage the flight control surfaces or hydraulic lines, then things are looking pretty grim. If they just have some minor mechanical problem that causes the engine to semi-gracefully stop spinning, or they ran out of fuel, or the engines "implode", that's less dire. Modern turbofan engines have to be tested and proven that they would vent all their debris out the back of the engine instead of flying around horizontally, so that shards of metal with very high velocity don't cut into the plane's fuselage or wings, which could cause catastrophic damage.
  • When the engines fail, what happens to the debris they make, if any? Similar to the first bullet, but if shards of high-speed metal were to strike the fuselage, wings or tail, it could make the airplane uncontrollable.
  • How high was the plane when the engines failed? The more altitude you have, the better.
  • How fast was the plane's airspeed when the engines failed? The more airspeed you have, the better. Both airspeed and altitude increase the range that the plane can glide before it collides with the ground, which means the pilots have more time to figure out where to land, come up with an approach, and execute the approach.
  • Are the landing gear able to come down on their own by gravity alone? This is always a risk: if you try landing on the fuselage, things will not go very well, especially because the unavailability of the engines means you have one less way to slow down the airplane once you touch the ground (no reverse thrust). If the landing gear deploys successfully, that certainly works in the pilot's favor.

In terms of the locations where you would ideally "want" to lose all your engines (and by "want" I mean in terms of the greatest likelihood of survivability), I'd rank them as follows:

  1. High in the sky, above a populated area. The good news is, populated areas have lots of airports. Airports are the best places to land a plane, because they have emergency crews on the scene and the runway is ideally suited to giving your plane the room it needs so it doesn't collide into anything. The bad news is, if your engines explode and make debris, it falls on people below you. Oh well -- you can't have everything you want in an emergency situation.
  2. Above a small body of water, near a populated area. Hey, it worked well for Sully. Boats were in the area, so they came out to rescue the passengers.
  3. Out in the desert. The B727 test crash on Discovery channel suggested that quite a few people would be surviving that crash. They crashed her on flat ground out in the desert. You'd be hard-pressed to get help any time soon if you were isolated from civilization, but at least the landing gear could do some work to help prevent a catastrophic explosion or fire.
  4. Everywhere else. Landing in hilly or mountainous terrain, very cold places, very wet places far from other people, inside an active volcano, etc. is pretty bad. I don't recommend it. If the plane isn't at least landing on its belly and sliding along the ground (which is itself a pretty bad situation and people are going to die), there are not likely to be many survivors, or any. You definitely don't want, say, the nose or the tail to take the brunt of the initial impact. Problem is, in most of the "everywhere else" cases, it's really hard to land on the belly and use your landing gear to slow down. Fire, smoke, water, and/or impact will be the death of most people on board in these scenarios.
$\endgroup$
  • 2
    $\begingroup$ Very good and thorough answer, though I would also note that the testing given to jet engines to ensure that failures are contained only applies to certain types of failures. In particular, intake fan blade separation is tested. However, sudden failure of a turbine disk, is almost certain to not be contained. QF32 suffered a rather substantial loss of controllability after several fragments cut various control lines. Fortunately, no pieces flew into the cabin and the pilots were still able to divert and land the aircraft safely. $\endgroup$ – reirab Dec 30 '14 at 5:49
  • 1
    $\begingroup$ You might mention that survivability of an off-airport landing, which is typically done with wheels up, is greatly effected by the forward speed of the airplane. Pilots would, of course, control the airspeed to contact the ground at as low an airspeed as possible, in other words slightly above the stall speed. Contact with the ground at 75 knots would be a lot more survivable than at 150 knots. $\endgroup$ – Terry Oct 21 '15 at 1:29
  • $\begingroup$ Very good post. I would add that sliding on the belly by itself doesn't necessarily mean people are going to die. The Asiana crash in SFO was very violent, with landing gear and tail contacting the sea wall in front of the runway and the aircraft almost cartwheeling afterwards, yet the vast majority of passengers survived (sadly, there were 3 fatalities). The Emirates crash in DXB was a ground contact with gear up and subsequent belly-sliding, and everyone inside the aircraft survived (one rescue worker did not). Of course both flights ended up on an even runway with fire services close by. $\endgroup$ – Cpt Reynolds Jul 11 '17 at 18:27
  • $\begingroup$ If you have sufficient altitude (which US Airways 1549 for one obviously didn't), wouldn't an initial problem over a large body of water (reducing the risk of injury or damage to anyone on the ground due to e.g. falling debris), itself near an airport, be even better than an initial problem over a populated area? $\endgroup$ – a CVn Oct 11 '17 at 10:38
4
$\begingroup$

Yep, planes can glide an very long distance, depending on the starting height and atmospheric conditions. All pilots are trained on how to glide as far as possible. In addition it is also EXTREMELY rare for all engines to fail, and modern passenger jets can fly safely even if one engine fails.

$\endgroup$
  • $\begingroup$ True. I suppose a more accurate statement is "Modern passenger jets can fly safely even if one engine fails" $\endgroup$ – Ben Oct 4 '14 at 22:54
  • $\begingroup$ @Ben: All passenger aircraft are required to be able to fly safely with one engine inoperative and were able and required to basically since planes for more than 12 passengers exist (piston engines are less reliable than turbines and the early ones were even less, so in the early days of aviation engine failures were much bigger concern). $\endgroup$ – Jan Hudec Oct 4 '14 at 23:04
3
$\begingroup$

I would like to add a bit of explanation. It is not the engines that keep aircraft in the air, it is the wings that generate lift as they move through the air. Whether the engines run or not does not affect the lift generation.

However as the aircraft moving through air and wings producing lift cause drag that slows the plane down. The engines compensate the drag so the plane can maintain speed. But when the aircraft flies, it also has potential energy due to it's altitude (that had to be provided by the engines during climb) and that can be exchanged for kinetic energy to maintain the speed instead.

The lift is significantly higher than the drag, so the plane only needs to descend quite slowly to keep flying. Glide ratio ranges from around 10 for small general aviation planes to around 60 for good sailplane. A typical value for airlines is around 18 which means from the typical cruise altitude of 10km they can glide about 180km.

Similar principle applies even to helicopters. Their rotor only provides lift when it is spinning, but when the engine fails, air flowing diagonally up and aft through the rotor can keep the rotor spinning and providing sufficient lift to maintain controlled descent. This is called autorotation. Helicopters have smaller gliding ratios, around 5, but they can use the energy of the spinning rotor itself to decelerate in the final phase of landing and so they can do vertical landing even with all engines out.

$\endgroup$
2
$\begingroup$

Sure. As one more example to the excellent ones already posted, British Airways Flight 9 encountered a cloud of volcanic ash from an eruption in the Indonesian archipelago, resulting in the failure of all four of the 747's engines. The plane was able to glide far enough to exit the ash cloud and then get three of the four engines restarted in order to divert to Jakarta, where due to abrasion of the windscreen they were forced to make a full instrument landing at an airport that was not enforcing ILS protection protocols. There were no injuries reported on board.

In the midst of the emergency, the captain, in typical British style, made the following announcement to the passengers:

Ladies and gentlemen, this is your captain speaking. We have a small problem. All four engines have stopped. We are doing our damnedest to get them going again. I trust you are not in too much distress.

$\endgroup$
2
$\begingroup$

An aircraft that has poor glide performance (descends steeply when not under engine power) will be very fuel inefficient and therefore uneconomic and unprofitable to fly commercially. The mass of the plane descending at some uniform rate represents a dissipation of potential energy. If that energy goes into sustaining forward speed, then the plane is in a glide. To keep the plane in level flight at the same forward speed, the engines would have to supply energy in the form of thrust at that same rate as it would be dissipated in a glide. The less work the engines have to do to maintain that speed, the less fuel they use, and the less height the plane would lose in an unpowered glide.

So you can be pretty sure that your economy fare ticket will buy you a seat on a plane that can glide pretty well should the engines ever quit.

Total engine failures are pretty rare events. The most notable events in recent years where all engines failed have all ended with 100% survival (although the aircraft didn't always fare too well). In a couple of cases, the aircraft glided to a runway landing and the landing gear got a bit bent up (hard landings because the pilot has only one try at the approach and may not have working flaps, won't have reverse thrust to aid braking). In another case, the plane landed safely on a grass levee without damage. In one case, the plane landed on water - total write-off, but everyone made it to safety. In one other case, all engines quit due to volcanic ash ingestion, but were later restarted in flight, allowing the crew to make a more-or-less normal landing.

Aircraft nose-dive and crash into the Earth for one of only two reasons:

  1. Someone at the controls points it at the ground
  2. Some vital part of the aircraft structure or control system breaks

Again, these are both extremely rare occurrences. Far more likely someone driving a car will fail to pay proper attention to the task and hit you while you are on your way to the airport.

$\endgroup$

protected by Community Jun 18 '17 at 11:41

Thank you for your interest in this question. Because it has attracted low-quality or spam answers that had to be removed, posting an answer now requires 10 reputation on this site (the association bonus does not count).

Would you like to answer one of these unanswered questions instead?

Not the answer you're looking for? Browse other questions tagged or ask your own question.