I was watching this video about MH370, and I couldn't help but notice the pilot seemed to have options that could only ever be useful for nefarious purposes.

For instance, the pilot turned off the plane's transponder, he depressurized the cabin, and he powered the plane down so far that the ACARS satellite communications turned off.

Why are these things even possible? Is there ever a good reason for any of them? The only reason I can come up with is for maintenance purposes; but even then there's no reason the pilot should be able to do it rather than ground crew.

  • 4
    $\begingroup$ @Bianfable Partly, but it leaves the question about depressurization open. $\endgroup$
    – Ryan_L
    Commented Apr 6 at 17:49
  • 20
    $\begingroup$ If the pilot is that untrustworthy, pressurisation and transponders are the least of your worries. $\endgroup$ Commented Apr 6 at 21:35
  • 5
    $\begingroup$ Extremely simple case for just about any electrical system on the plane: What if that subsystem develops an electrical fault in flight? Should it take down the entire electrical system, or potentially start a fire, just because one thing broke? If it draws enough current it would trip the breaker (which is what allows the plot to turn it off), but what if it only draws enough to cause heating and excess current draw without enough to blow the breaker? $\endgroup$ Commented Apr 7 at 2:15
  • 5
    $\begingroup$ In contrast to driving your car, it is not possible to turn off the highway and go out to fix a problem when you are in a plane. Sometimes the way to continue to a safe landing is to turn off the disturbing system (it might be on fire). As modern planes does not have a mechanic the pilot has to be able to do everything from the cockpit. $\endgroup$
    – ghellquist
    Commented Apr 7 at 16:35

3 Answers 3


They have control of them because they need to be turned off sometimes. For pressurization, a 737 has two PACKs (Pressurization and Air Conditioning Kits) which can be turned on or off independently. The pilot may need these under many circumstances:

  • To turn off the system on the ground when it shouldn't be running
  • If one PACK is inoperative
  • In certain emergencies like smoke in the cockpit or attempting an airborne restart of the engines

The pressurization system also has an outflow valve that can be used to change the pressurization or quickly depressurize the cabin. Quickly depressurizing the cabin could be required in flight to remove smoke or fumes from the cabin or on the ground in order to evacuate the airplane.


I'm neither a pilot nor an aerospace engineer, but just from my own sense of logic ...

If I was a pilot, I'd want control of as much as possible. With a little imagination I'm sure you could come up with a scenario where it would be a good thing to turn off or take some other action with any system. Any system could be malfunctioning and causing trouble. If an airplane is in trouble, they met well be over the ocean thousands of miles from the nearest land. The pilot wants to have the means to deal with anything that might happen.

Maybe it's true that 99% of the time there would be no reason to mess with X. But what about that other 1%? To say, "300 people were killed because there was nothing the pilot could do to correct the problem, because the engineers figured that, hey, MOST of the time there would be no reason for the pilot to mess with this", yeah, that would be a problem.

Not on the same scale, but ... Just yesterday I had a problem with a default setting in Microsoft Word. Fortunately for me I was able to change the setting, but I really had to search for it. Maybe it's true that 99% of the time, the default setting is fine and there's no reason to mess with it. But my case was the odd 1%.

  • 4
    $\begingroup$ Every switch doubles the amount of states in your system you have to consider, test and verify. It also increases the risk for failure. What if your switch to turn off critical thing X (which should be on all the time anyway) fails in an open state and now you don’t have X? $\endgroup$
    – Michael
    Commented Apr 8 at 7:06
  • 6
    $\begingroup$ @Michael Well, you may think so, but that is generally the path that has been followed in safety-critical systems for decades now. Nobody builds systems where a critical component is not monitored, and as far as possible there are procedures for coping with failures (not necessarily through redundancy). $\endgroup$
    – jpa
    Commented Apr 8 at 8:06
  • 6
    $\begingroup$ @micheal every subsystem already multiplies the possible states by N+1 where N is the number of failure modes. The switch attempts to cut that back to multiplying by two by unifying the behavior of the failure modes. $\endgroup$ Commented Apr 8 at 10:50
  • 4
    $\begingroup$ @michael Yes, if you add a switch to turn off X, you now introduce the possibility that that switch will fail. But many systems on airplanes are extremely complex. They already have hundreds or thousands of possible ways they could fail. If you could guarantee me 100% that the system will never malfunction, but that there is a high probability that a simple on/off switch will fail, sure, adding the switch increases the possible problems. But that's a very unlikely assumption. Adding the switch adds a minuscule amount of possible failure. $\endgroup$
    – Jay
    Commented Apr 8 at 15:13
  • 1
    $\begingroup$ @jpa that works for trivial systems, but complex systems just provide too much information for a human to process easily. So the "every" is not done - even on purpose we skip it. $\endgroup$
    – paul23
    Commented Apr 9 at 15:36

Most of these switches are called circuit breakers.

The general idea is, if something shorts, it gets turned off one way or another. Anything that must be always on is a place to bring all avionics down because of a bus short through the device; and on a plane this big, loss of avionics leads directly to loss of control.

Yes I know most of avionics can be turned off, but there's a few things that actually are critical. If you can't command the hydraulic pumps because of a bus short your future flight time isn't very long.

  • $\begingroup$ Interesting to note that the "Airbus A380 has nearly 100,000 different types of electrical cables, with an estimated total length of about 300 miles… contribute to over 1,100 individual functions." simpleflying.com/airbus-a380-facts-figures-list Although wikipedia says wiring difficulties delayed production by 2 years and almost doubled development costs $\endgroup$
    – Xen2050
    Commented Apr 9 at 6:09

Not the answer you're looking for? Browse other questions tagged .