-1

I'm doing some research for a potential fan story in the Dresdenverse. Summary: a passenger is flying for the first time in an A-380. The A-380 is fully functional: it's the first flight after a complete maintenance and the pilots are very experienced. However, components slowly stop working on the plane, starting with small stuff: a seat won't recline, a folding tray won't fold open,... However, as time passes more and more components stop working and in the end, pretty much everything on the plane is malfunctioning. They only just make it to the airport, where they make a hard landing. There are minor injuries, but everyone makes it out alive. In the end, the passenger turns out to be a powerful wizard with a technology disruption aura who was being deported to his home country.

The question I have is: in this scenario, what is the absolute bottom requirements for this plane to be able to make it to the ground with no loss of life? I'm talking about ABSOLUTE here: it's okay if the plane after this is a mangled wreck which will never be able to fly again or even be examined for what went wrong. I just want the passengers to survive.

Nzall
  • 892
  • 1
  • 6
  • 15
  • The Minimum Equipment List may be a good start for you - it lists every piece of equipment that is LEGALLY allowed to fail and still allow the aircraft to depart. The list for the A380 is extensive and details how many of each item can fail - the failure of such things would definitely allow the aircraft to fly as you ramp up to the more severe stuff. The obvious culmination would be several engine failures and reverting to Direct Law – Dan Oct 12 '18 at 16:09
  • 7
    You are basically asking about the whole design of reliability engineering of A380. – user3528438 Oct 12 '18 at 16:31

2 Answers2

8

Probably not the answer you are looking for, but I would definitely recommend to read the book "QF32" by Richard de Crespigny about Qantas Flight 32. It is the captain's personal account of what can go wrong in an A380 suffering a major damage (uncontained engine failure, leading to multiple system failures), and how a competent flight crew can save the day with every single person on board walking away from the aircraft afterwards.

Quite an impressive read. It also gives an impression on the levels of redundancy in a modern aircraft like the A380.

The official final investigation report by the Australian Transport Safety Bureau may offer further technical insights to this event.

nabla
  • 1,216
  • 2
  • 13
  • 24
4

Oh, that's a fun one...

Let's start by imagining that all electronics and all moving parts on the airplane fail. It's still structurally intact, but there's no way to control the plane.

If this happens, the #1 problem is the spiral dive. The bank angle will steadily increase further and further. As the bank angle increases, the wings are less and less effective at holding the airplane up, so eventually... they stop holding it up.

In order to prevent the bank angle from increasing, the pilots must have some means of lateral control. This means that they must be able to control at least one engine, or at least one aileron, or the rudder.

This other answer states that if an Airbus's electrical system fails, "the system reverts to mechanical backup, where pitch control is achieved through the horizontal stabilizer and lateral control is accomplished using the rudder pedals."

So if you want as many things as possible to go wrong, while still being a survivable failure, I suggest:

  • All four engines fail.
  • The landing gear fails; it cannot be extended. (If the landing gear could be extended and retracted, the pilots could use it as a flight control.)
  • The entire electrical system fails, meaning that conventional control is no longer possible.
  • The trimmable horizontal stabilizer also fails, meaning that the only flight control available is the rudder.

If the pilots are sufficiently skilled, they will be able to maintain directional control, and avoid both Dutch roll oscillations and phugoid oscillations when they do. If the pilots are able to do that, then their job description becomes very simple: control the direction of the airplane in such a way that when it touches the ground, it's on a runway.

There's one problem remaining with using the rudder only, and that's that there's no way to control the airplane's descent rate. It'll simply descend at whatever rate it "wants to", aerodynamically. This descent rate may be similar to the descent rate of the so-called Gimli Glider (Air Canada Flight 143, which was a Boeing 767). I don't know what that descent rate was, but a couple of web pages describe it as being about 2,000 feet per minute, which is about 20 miles per hour. Keep in mind, that's 20 miles per hour straight down.

(If the pilots are really skilled, they'll induce a phugoid oscillation which is timed in such a way that the touchdown is gentler. But that sounds pretty unlikely.)

If you want to give your pilots even more of a thrill ride, you can have the rudder fail upon touchdown, or even a few seconds (10 or 20 seconds?) before touchdown. The effect will be pretty much the same as a car: the plane will continue in the same direction for a few seconds, but it'll inevitably start to veer to one side. It will likely run off the runway and into the grass.

All in all, the outcome will probably be exactly what you're looking for: lots and lots of components fail, the plane ends up a "mangled wreck", but everyone survives.

Side notes:

  • Even if the landing gear were working perfectly fine, the pilots might decide not to use it. If they extended the gear in flight, but the gear failed to retract, then it would cause drag, resulting in an increased descent rate, which could be disastrous. If they extended the gear just before touchdown, but the brakes failed, then they might overrun the end of the runway; if they ran into an obstacle, that could be even worse than making a gear-up landing.
  • As a little bonus, you could have the cockpit voice recorder and the flight data recorder both fail at the very beginning of the flight. This is likely to leave investigators more confused about what could have caused the accident.
  • I'm a Dresden fan too; let me know if you finish your story!
Tanner Swett
  • 5,771
  • 1
  • 31
  • 58
  • How does one control a phugoid with just the rudder? – Sanchises Oct 12 '18 at 17:34
  • @Sanchises With difficulty! :D I don't know the details, but I can make an educated guess. To damp a phugoid oscillation, you only need to do any one of four things: decrease the airspeed while it's high; increase the airspeed while it's low; lower the nose while it's high; or raise the nose while it's low. The rudder can be used to produce sideslips, which decrease the airspeed, so if I wanted to damp a phugoid with the rudder, my first attempt would be to give rudder input while the airspeed is high. – Tanner Swett Oct 12 '18 at 18:21
  • Rudder input (and accompanying coupled roll) tend to lower the nose. Let a little speed build, then neutralize rudder, and you're above trim speed; you'll almost certainly get a little phugoid motion. – Zeiss Ikon Oct 12 '18 at 18:21
  • 1
    Back in the 1960s and 1970s, rudder-only radio control could be used to fly loops -- put the aircraft into a tight spiral, let speed build enough, and neutralize; the overspeed would raise the nose past vertical, across inverted, and then you'll apply rudder as you passed the bottom to prevent a whipstall. – Zeiss Ikon Oct 12 '18 at 18:23
  • Interesting... Have you included the landing gear in "all moving parts"? And does this need the rudder all the way until the plane has stopped on the ground? if not, what's the absolute earliest I can have it fail during the landing sequence? – Nzall Oct 12 '18 at 18:41
  • @Nzall Good questions. I edited my answer to include that information. – Tanner Swett Oct 12 '18 at 20:38
  • 3
    As far as I can tell, the mechanical law was deleted on A380, likely because A320–A340 never ever reverted to it anyway. Complete hydraulic failure is more likely than complete electric failure and complete hydraulic failure renders the mechanical law inoperative anyway. – Jan Hudec Oct 12 '18 at 20:59