On 17th April 2018, Southwest 1380, a Boeing 737-700, was cruising about FL320 when it suffered a catastrophic failure to engine 1. It then took 5 minutes to descend to 10,000 feet, and I noted from the ATC recordings that the pilot said it would require a long approach to land at PHL.

Is this because the airplane could not bleed off airspeed fast enough for a short/straight-in approach? Is this the fastest that a large commercial airplane can descend and slow for an emergency landing?

I assume that, to achieve maximum descent and deceleration the remaining engine would be pulled to idle, and that landing gear would be extended at maximum safe speed, which IIRC is 250 knots?

Would it be reasonable to fly cross-controlled with an aircraft in this condition? With no engine thrust requiring rudder input, is that safe at all speeds? How much can it increase drag?

What are the speed limits for extending flaps, and how much do those affect drag?

(I can appreciate that in the case of Southwest 1380 the pilot in command, not knowing the extent of damage, may have been hesitant to do anything that would strain the airframe or controls. So I'm curious as to how long an "unstressed" descent would take, and then how much faster one could descend by using the most aggressive techniques that a pilot would reasonably use in an uncompromised passenger flight in the same model.)

  • $\begingroup$ Related, related $\endgroup$
    – Pondlife
    Apr 21, 2018 at 13:41
  • 2
    $\begingroup$ 32,000 ft down to 10,000 ft is 22,000 feet. 22,000/5min = 4,400 ft/min, that's a pretty good ear popping descent rate. I try and limit my descents to 500 ft/min when flying my 4 seat prop plane. $\endgroup$
    – CrossRoads
    Apr 21, 2018 at 13:47
  • 2
    $\begingroup$ @RalphJ: This is a set of hypothetical questions inspired by a recent accident. It does not require or request any speculation about unknowns in the referenced accident or information that may become known in the future. If you think any question inspired by an accident is off-topic, you should argue that in meta. $\endgroup$
    – feetwet
    Apr 22, 2018 at 13:28
  • $\begingroup$ @fooot: Answers there are informative of part of my question, but this isn't just about getting to safe pressure altitude as fast as possible, but also how quickly a plane can slow for an emergency landing (and what factors might extenuate those procedures if the emergency is a catastrophic engine failure). $\endgroup$
    – feetwet
    Apr 22, 2018 at 17:35
  • $\begingroup$ Note that 5 minutes is plenty of time. Passenger oxygen masks are required to last for 10 minutes. That’s for emergency descent down to 14,000 feet which is a breathable altitude if uncomfortable for some. They usually do 10,000 feet because it is comfortable for most people. $\endgroup$
    – TomMcW
    Apr 22, 2018 at 17:56

1 Answer 1


A 5-minute descent from FL320 to 10,000 feet is pretty quick. As noted in the comments, the descent rate is 4,400 ft/min. For reference, during final approach descend rate is 300~500 ft/min in a small airplane, 500~700 ft/min on an airliner. A typical descent from cruise altitude is performed at 1200~2400 ft/min.

Getting down fast

A "dive bomb" flight-simulator-style descend is not practical if you intend to land. After all, you have to arrest the descend and bring the aircraft to a stable state. You may very well break the plane when trying to pull out of the dive.

Now, what is the fastest way to lose altitude? According to the Boeing 737 QRH, the Emergency Descent checklist (which is conveniently listed on the first few sections) says...

(...) Spreadbrake...FLIGHT DETENT

(!) If structural integrity is in doubt, limit speed as much as possible and avoid high maneuvering loads.

Set target speed to Mmo/Vmo. (...)

In other words, deploy the speedbrakes, then pitch down to max airspeed.

Note that, however, this procedure is only used when it is absolutely necessary to get down as quickly as possible. The Boeing 737 FCOM also contains the following, in bold letters:

WARNING: Use of speedbrakes at speeds in excess of 320 KIAS could result in severe vibration, which, in turn, could cause extreme damage to the horizontal stabilizer.

Landing gear & Flaps

Deploying the landing gear is not good because:

  1. The maximum flying speed with gear extended is 320 knots, while Vmo is 330 knots. Therefore, if you descend with the gear extended, you will lose altitude slower.
  2. The maximum flying speed for extending the landing gear is 270 knots. Therefore, if you wish to extend the gear, you will have to slow down to 270 knots first, then lower the gear handle, wait for the gears to be locked, then pick up speed again.
  3. Flying with the gear extended introduces vibration to the airframe. Deploying speedbrakes already introduces vibration. A damaged engine further adds vibration.
  4. You need the gears to touchdown, so better protect them in the landing gear bay.

Extending the flaps is even worse because the maximum speed for flying with flaps is even lower. The maximum flap extension altitude is also 20,000 ft.

Cross control

Cross-control can probably lose altitude even quicker, but it is not a maneuver you would do on a passenger airliner. When crossing control, the gravity vector is not down, but sideways - meaning passengers will be forced to the side of the seat, a sensation they are not used to. Furthermore, the passengers sitting on the window side will likely perceive the attitude as extreme bank angle and think that the plane is going to crash! The last thing we want in this scenario is a cabin with panic passengers.

  • $\begingroup$ I gather from news reports that some large number of the passengers thought the plane was doomed and they were going to die. The one who did die was partially sucked out the broken window. (Now I'm wondering if, with a full awareness of that condition, slipping with the exposed side of the cabin forward could offset the decompression force through that hole....) In any case, if I were PIC and I had to choose between saving the craft but making the passengers think they're going to die, or keeping them calm but actually killing them in a failed landing, I'd go with the former ;) $\endgroup$
    – feetwet
    Apr 21, 2018 at 16:30
  • 3
    $\begingroup$ Speed and vertical speed are not proportional in this case, so the slower speed with gear in point 1 does not actually mean anything. The vertical speed is given by loss of energy to drag, so if the drag with gear is sufficiently higher, the descent rate could be higher with gear at 320 knots than without it at 330 knots. The second point is a good reason not to extend it anyway though (and it is not part of the emergency descent procedure). $\endgroup$
    – Jan Hudec
    Apr 21, 2018 at 21:02
  • $\begingroup$ This is a very good answer. I’m only a single engine pilot but I can’t imagine willingly going cross-controlled on a twin with one engine out. Especially when you have fuselage damage, I would want to keep the forces as normal as possible. $\endgroup$
    – Ben
    Apr 21, 2018 at 22:44

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