Engine-Out Procedures and Escape Routes

Operational ProceduresCPL · ATPL · Type rating11 min readUpdated Nov 2026
Definition

An engine-out procedure is a route and climb profile, designed by the operator, that a multi-engine airplane flies after an engine failure so that its one-engine-inoperative net flight path stays clear of obstacles after takeoff, or of terrain en route, until it can land or continue safely.

An engine-out procedure is the route a multi-engine airplane flies when an engine fails and the normal route no longer protects it. After takeoff it is an engine-out standard instrument departure (EOSID): a track and climb profile that keeps the one-engine-inoperative climb clear of obstacles. In the cruise over high ground it is an escape route: a pre-planned path toward lower terrain and an aerodrome where the airplane can land.

Both are the operator's work, not the procedure designer's. Published departures assume all engines operating, so the operator builds its own procedures from obstacle and terrain data and the flight manual's one-engine-inoperative performance, and the crew brief the engine-out departure before every takeoff. The rules behind them are CAT.POL.A.210 and CAT.POL.A.215 of EASA's Part-CAT for performance Class A airplanes and, in the United States, 14 CFR 121.189 and 121.191.

Why engine-out procedures exist

A SID or an obstacle departure procedure protects an aircraft that follows its track and climbs at least at the published gradient: 3.3 percent in PANS-OPS unless a higher figure is published, 200 ft per NM for the FAA (see instrument departures). The FAA's AIM says so plainly: ODPs, SIDs and diverse vector areas assume all engines operating, and contingency procedures for an engine failure after liftoff are the operator's responsibility.

With one engine out the gap is wide. A twin need only show a gross second-segment gradient of 2.4 percent, a net 1.6 percent, less than half the default design gradient, and a steep published gradient, such as the 7.5 percent required up to 1,000 ft on the SID from runway 17 at Cannes, widens it further. Flying the SID after a failure gives no assurance of obstacle clearance.

The performance rules fill the gap. For Class A airplanes an engine failure is considered in every phase of flight, and CAT.POL.A.210(c) requires the operator to establish contingency procedures that meet the takeoff obstacle clearance requirements and give a safe route, avoiding obstacles, until the airplane can either meet the en-route requirements of CAT.POL.A.215 or land at the departure aerodrome or a takeoff alternate. The en-route rule asks the same question of the cruise: if an engine fails at the most critical point of the route, can the airplane still clear the terrain?

Exam tip: the SID or ODP gradient is an all-engines figure. Meeting it says nothing about an engine-out climb, and the engine-out net flight path does not replace it when all engines are running.

Engine-out SIDs and special departure procedures

An engine-out SID (EOSID), also called an engine failure procedure or, in US airline usage, a special engine-out procedure, is designed for each runway by the operator or a specialist obstacle-analysis provider. It normally gives:

In the simplest case the engine-out route is the runway track or the SID itself. Where high ground lies ahead, the EOSID may turn toward lower terrain, sometimes earlier than the SID does. EASA guidance, GM1 CAT.POL.A.210, calls the point where an engine failure route leaves the normal departure route the deviation point. A failure before it takes the airplane onto the engine failure route. After it, continuing on the normal departure will normally give adequate clearance, but where that clearance is marginal the operator should check that a flight with an engine failed beyond the deviation point can safely continue along it.

Some procedures rely on visual course guidance from ground references, which EASA counts as visible when they lie more than 45° either side of the track and no more than 20° below the horizontal; the operations manual should then state the minimum weather for using them (AMC3 CAT.POL.A.210). Steeper banks, up to 20° between 200 ft and 400 ft or up to 30° above 400 ft, need approval under CAT.POL.A.240, with flight manual data, visual guidance, and weather and wind limits for each runway.

Obstacle data and the net flight path

The design starts from the net takeoff flight path: the one-engine-inoperative path from 35 ft at the end of the takeoff distance to at least 1,500 ft, reduced at every point by a fixed gradient of 0.8 percent for a twin, 0.9 for a three-engined and 1.0 for a four-engined airplane (see takeoff flight path and obstacle clearance). Obstacle clearance is always shown on the net path, never on the gross path the airplane is expected to fly.

The one-engine-inoperative takeoff flight path that every engine-out procedure is built on: four segments, the gross and net paths and the 35 ft obstacle margin. v1prep schematic.
The one-engine-inoperative takeoff flight path that every engine-out procedure is built on: four segments, the gross and net paths and the 35 ft obstacle margin. v1prep schematic.Illustration © v1prep
EASA, CAT.POL.A.210 FAA, 14 CFR 121.189
Vertical margin 35 ft; 50 ft wherever the airplane is banked more than 15° 35 ft
Lateral margin 90 m + 0.125 D, or half the wingspan + 60 m + 0.125 D below 60 m span 200 ft inside the airport boundaries, 300 ft after passing them
Outer limit of the area 300 m (track change up to 15°) or 600 m (more than 15°) with the required navigational accuracy; otherwise 600 m or 900 m Methods in FAA AC 120-91

D is the distance from the end of the takeoff distance available, or from the end of the takeoff distance if a turn is scheduled before it. In the EASA calculation no track change is allowed until the net path reaches half the wingspan, and at least 50 ft, above the end of the takeoff run available; up to 400 ft the bank is assumed not to exceed 15°, and above 400 ft up to 25° may be scheduled. No more than 50 percent of the reported headwind and no less than 150 percent of the reported tailwind may be used.

A turn costs climb. Unless the flight manual says otherwise, AMC2 CAT.POL.A.210 pairs a 15° turn at V2 with the flight manual's 15° gradient loss, 20° at V2 + 5 kt with twice that loss, and 25° at V2 + 10 kt with three times it. On a wet or contaminated runway the takeoff distance ends at a 15 ft screen, so after a failure at V1 the airplane may initially be up to 20 ft below the net path and clear close-in obstacles by only 15 ft.

The obstacle data come from surveys such as the ICAO Aerodrome Obstacle Chart, Type A, and from obstacle databases. The AIM warns that the takeoff obstacle notes of the FAA's Terminal Procedures Publication cannot by themselves assure compliance with the one-engine-inoperative requirements, and directs operators to precise sources such as GIS databases. At airports prone to strong winter inversions, where warm air aloft reduces thrust above the acceleration height, the analysis may carry an inversion allowance or limit.

The design also fixes the engine-out acceleration altitude. The A320 procedures require it to be at least 400 ft above the airport, high enough for the net flight path to clear obstacles by 35 ft, and low enough that the time limit on takeoff thrust is not exceeded. Under 14 CFR 25.1521 and the engine certification that limit is 5 minutes, extendable to 10 minutes with one engine inoperative where demonstrated; after it, maximum continuous thrust and a lower gradient apply.

Publishing and briefing

The procedure reaches the crew with the operator's runway analysis or performance software, alongside the maximum takeoff mass and speeds for each runway; airlines commonly contract obstacle analysis services to produce it. The clearance is for the SID, so a crew who leave it after a failure tell ATC at once.

The engine-out procedure is part of every takeoff briefing, because the routing may differ from the SID and both pilots must know it before the takeoff roll (see crew coordination, briefings and callouts). A runway or intersection change after the data have been computed means recomputing the performance, cross-checking the new data and briefing the departure and engine-out procedure again. On the A320 the performance cross-check covers the runway, takeoff shift, V1, VR and V2, green dot speed, thrust setting, flaps, trim, packs, anti-ice and the EO ACCEL ALT.

An aircraft control unit with a small dark screen of green and white text above rows of labeled keys and a full alphabet keyboard.
The first officer's MCDU on the pedestal of an Airbus A320. The takeoff data the crew enter and cross-check before departure include the engine-out acceleration altitude, EO ACCEL ALT, alongside the runway, the V-speeds and the thrust setting.Christoph Paulus · Public domain · Wikimedia Commons

En-route escape and drift-down routes

After an engine failure in the cruise a twin usually cannot hold its level. At maximum continuous thrust and the drift-down speed, close to the one-engine-inoperative minimum drag speed, it descends until thrust and drag balance. This drift-down must clear the terrain (see one-engine-inoperative en-route performance).

CAT.POL.A.215 requires the en-route net flight path either to have a positive gradient at least 1,000 ft above all terrain and obstructions within 5 NM (9.3 km) either side of the intended track, or to allow a drift-down from the cruising level to an aerodrome where a landing can be made, clearing the same corridor by at least 2,000 ft. Either way the net path must have a positive gradient 1,500 ft above the aerodrome where the landing is assumed. The engine is assumed to fail at the most critical point, wind is included, and the corridor widens to 10 NM (18.5 km) where the navigational accuracy does not meet RNAV 5. The FAA's 121.191 has the same two methods and margins, with 5 statute miles either side of track, and is applied at dispatch to set the weight over high terrain.

Under AMC1 CAT.POL.A.215 the analyst plots the highest points in the corridor and checks whether one-engine-inoperative level flight 1,000 ft above them is possible. If not, or if the mass penalty is unacceptable, a drift-down is worked out for a failure at the most critical point, clearing the critical obstacles by 2,000 ft, and the minimum cruise altitude is found where the two drift-down paths intersect, with an allowance for decision making. Published MEAs or minimum off-route altitudes (MORA) may be used instead, but at and below 6,000 ft they guarantee only 1,000 ft of clearance.

The result is an escape route: a decision point on the route, a turn toward lower ground and the aerodrome to fly to. Operators publish such routes for mountainous sectors planned on the drift-down method (see operational flight plan).

Exam tip: do not confuse the drift-down decision point with the decision point of the fuel planning reduced contingency fuel (RCF) procedure, in which the flight is planned to its destination via a point from which a second, optional refueling destination can still be reached. It has nothing to do with terrain.

A loss of pressurization over the same terrain is the second case. The crew descend to 10,000 ft or the minimum safe altitude, whichever is higher, and chemical passenger oxygen lasts only about 12 to 22 minutes. Under AMC1 CAT.IDE.A.235 the oxygen quantity assumes a pressurization failure at the most critical point and a descent to an altitude with adequate terrain clearance, so mountain routes get decompression escape routes that reach a safe level before the passenger oxygen runs out. Where the other fuel components do not cover either event, AMC1 CAT.OP.MPA.181 adds fuel to reach a fuel en-route alternate from the most critical point, hold 15 minutes at 1,500 ft above it, approach and land.

Cockpit display with engine gauges at the top and, below, a map of terrain ahead shaded green, yellow and red around the aircraft symbol.
An EGPWS terrain display after departure from Vancouver, the terrain ahead shaded green, yellow and red by its height relative to the aircraft. Engine-out procedures and escape routes are worked out on the ground, so that after a failure the crew follow a known path instead of improvising one past terrain like this.Shawn from Airdrie, Canada · CC BY-SA 2.0 · Wikimedia Commons

Flying the procedure

After an engine failure above V1 the pilot flying keeps control and follows the flight director, whose speed target allows for the failure. On the A320, SRS targets the speed at which the failure was detected, between V2 and V2 + 15 kt; on the 737 the target is V2 for a failure below V2, the speed at the failure between V2 and V2 + 20 kt, and V2 + 20 kt above that.

The airplane then follows the engine-out track: no turn before the published height or fix, no more bank than the procedure assumes (on the 737, 15° until V2 + 15 kt), and a climb at the target speed to the engine-out acceleration altitude. There the crew accelerate, retract the flaps on schedule and set maximum continuous thrust; on the A320, SRS does not disengage by itself at the EO ACCEL ALT with an engine out, so the crew must engage another vertical mode. The procedure clears the obstacles only with the speeds, bank angles and acceleration height it was designed for, and it replaces any noise abatement profile (see noise abatement procedures). The AIM asks pilots to tell ATC as soon as possible of reduced climb capability; even radar vectors assume a climb of at least 200 ft per NM.

In the cruise the drift-down follows the type's procedure: on the 737, ENG OUT mode, autothrottle off, maximum continuous thrust, the ENG OUT speed and maximum altitude on the MCP, then LVL CHG; on the A320 and the E190-E2, green dot speed. Over terrain with an escape route, the crew turn at the decision point as planned.

The engine-out case returns at the destination. At Nice, the minima page of the RNP Z approach to runway 04R tabulates LNAV/VNAV minima by missed approach climb gradient for airplanes able to climb at more than 2.5 percent: for Category C, a decision altitude of 450 ft at 3 percent but 300 ft at 5 percent. A note adds that for public transport operations their use must take into account the failure of one engine during the missed approach (see go-around and missed approach).

Warning: an engine-out procedure belongs to one runway and one set of assumptions. A runway change, an intersection takeoff or a different flap setting can change the procedure or the acceleration altitude, so it is briefed again whenever the data change.

Frequently asked questions

What is an engine-out SID (EOSID)?

An engine-out SID is a departure route that the operator designs for each runway, often with a specialist provider, to be flown when an engine fails during takeoff. Published SIDs assume all engines operating, so they do not protect the engine-out climb. The EOSID gives the track and any turn, the engine-out acceleration altitude and where to go afterward, so that the net takeoff flight path clears all obstacles by the regulated margin.

Can a crew fly the normal SID after an engine failure?

Only where the operator's analysis shows it is safe. A SID is designed for an all-engines climb of at least 3.3 percent in PANS-OPS or 200 ft per NM in the United States, while a twin's net second-segment gradient may be as low as 1.6 percent. The engine-out procedure for the runway says what to fly, which may be the SID track, a different route or an early turn.

What is a drift-down escape route?

An escape route is a pre-planned path for an engine failure or loss of pressurization over high terrain. From a published decision point the crew turn toward lower ground and descend, at the one-engine-inoperative drift-down speed or in an emergency descent, to a level they can hold, clearing the terrain by the required margin on the way to a suitable aerodrome. Operators plan these routes before dispatch.

What is the engine-out acceleration altitude?

It is the height at which, after an engine failure on takeoff, the airplane stops climbing at its engine-out climb speed, V2 or a little above, and accelerates to retract the flaps. On the A320 it must be at least 400 ft above the airport, high enough that the net flight path clears obstacles by 35 ft, and low enough that the time limit on takeoff thrust is not exceeded. It is part of the takeoff data the crew cross-check.

What is the deviation point on an engine-out departure?

EASA guidance uses the term for the point where an engine failure route leaves the normal departure route. Before it, a failed engine sends the airplane onto the engine failure route. After it, the airplane can normally continue on the normal departure, but where obstacle clearance along that route is marginal the operator should check that continuing remains safe after a failure beyond the deviation point.

Test yourself on Engine-Out Procedures and Escape Routes

The v1prep banks cover this topic in Operational Procedures (070), with a worked explanation for every answer. EASA ATPL, PPL, IR and CPL, the FAA written tests and type ratings for the A320, 737, A220, 787, ATR 72 and E-Jet.

Start practicing →
15,000+ questions · EASA & FAA · Free to start

Library articles are written for study and exam preparation. They do not replace your aircraft's approved documentation, your operator's procedures or the regulations themselves.