NTSB investigators examining N772SW's damaged left engine after Flight 1380.
Photo: National Transportation Safety Board, Public domain
Image info
This investigation photograph shows exposed engine components after the aircraft landed in Philadelphia. It documents the physical damage available for inspection; identifying the initiating fracture and the sequence of failures required further examination.
Resized and JPEG recompressed; no content changes.
Geographic context
New York LaGuardia
Dallas Love Field (planned)
Philadelphia (diversion)
Approximate airport and geographical reference positions. This schematic shows the planned journey and relevant locations, not a recorded flight track. Basemap: Natural Earth, public domain.Read supporting evidence
During climb, a low-cycle fatigue crack caused fan blade No. 13 in the left engine to separate. Its impact with the fan case transmitted loads to the surrounding inlet and fan cowl. Parts of the cowl departed and a fragment struck the fuselage near a cabin window. The window separated, causing rapid depressurization and a fatal passenger injury. The crew regained control, descended and diverted to Philadelphia, landing about 17 minutes after the failure.
Read the full incident Follow what happened, from the flight to the investigation findings.
What investigators concluded
The NTSB traced the probable cause to the fatigue crack and the subsequent chain of fan-case impact, cowl separation and window loss. Its analysis identified loads through the radial restraint fitting that the fan-cowl design had not accounted for. The reason the crack escaped the earlier overhaul inspection remained unknown. The NTSB judged the choice of Philadelphia appropriate and found that the crew balanced the emergency workload with the need for a timely landing.
1,789 words / about 9 minutes to read. Based on the investigation records cited below.
A fan blade fails during the climb
Southwest Airlines flight 1380 was flying from New York LaGuardia to Dallas Love Field on 17 April 2018. The Boeing 737-700, registered N772SW, carried 144 passengers and five crew members. At about 11:03, as the aircraft climbed through approximately 32,000 feet, a fan blade separated in the left engine. Parts of the engine inlet and fan cowl then broke away. A component from the inboard fan cowl struck the fuselage beside the window at seat 14A, causing the window to leave the aircraft.
The open window caused rapid depressurization. The crew heard the disturbance and encountered significant vibration, alerts and a sudden roll. The flight data recorded a left bank reaching 41.3 degrees before the aircraft returned towards level flight. The first officer was flying, and the captain initially handled the other duties. The pilots had to establish control, obtain oxygen and begin a descent while the engine failure and cabin-pressure emergency occurred together.
The left engine was brought to idle and then shut down. The crew used their oxygen masks, but the report records initial uncertainty over the switch needed to communicate through the mask microphones. Controllers' repeated calls initially received no clear reply. This temporary communication difficulty formed part of the emergency workload. The crew did establish radio contact and reported an engine emergency while descending. Their early description reflected the indications and circumstances they were experiencing, while the investigation later established the precise structural sequence that had damaged the cabin.
The captain asked the controller for a nearby diversion airport. Harrisburg was suggested, but the crew considered Philadelphia, with the first officer identifying it on a map as a suitable nearby choice. About two and a half minutes after the engine failure, they received clearance direct to Philadelphia International. The captain informed the cabin that the flight would divert there. The decision was made while the aircraft was already descending from cruising altitude with one engine shut down.
The crew's radio report was clarified during the descent: the captain told the controller there was no fire then, but that they were operating on one engine. Controllers arranged lower altitudes and emergency support. The peak recorded descent rate was approximately 5,228 feet per minute. The aircraft initially remained between about 280 and 300 knots during the descent to 17,000 feet. These data show a rapid descent, but the account does not assume that speed alone describes how the crew managed every part of the emergency.
At about 13,600 feet, the captain took over as pilot flying, and the first officer became pilot monitoring. He began the engine fire, severe damage or separation checklist. Below 10,000 feet, the pilots removed their oxygen masks, with a brief further transfer of control while the captain's mask was removed. The captain later explained that she reduced speed to ease the severe vibration. Controllers offered flexibility over speed and the approach. The crew now had to reconcile the need for timely landing with incomplete information about both the aircraft's damage and the condition of people in the cabin.
The flight attendants heard a loud noise and felt the aircraft shaking. They initially went to their jumpseats, secured themselves and used oxygen masks. They then obtained portable oxygen and moved into the cabin to assist passengers. The report describes debris in the cabin and the difficulties created by the noise. One attendant later explained that she answered an interphone call but could not hear the caller. The cockpit's first attempt to obtain a cabin report therefore did not bring a usable answer.
At seat 14A, the passenger had been partly drawn through the window opening after it departed. Passengers and cabin crew worked to bring her back inside and provide assistance. The first officer eventually established contact with an attendant, who described the missing window and injured passenger. The captain learned of the injury at about 11:15. The crew's knowledge of the cabin emergency was therefore acquired during the diversion, rather than being complete from the moment the engine failed.
The captain had requested a long final approach to allow time for checklists. After learning about the passenger's condition, she decided to expedite the approach. Medical personnel were requested to meet the aircraft on the runway. The NTSB's investigation treated the cabin response as part of the same accident sequence as the engine failure and diversion. The emergency was not limited to managing a single-engine landing: the loss of a window had exposed a passenger to fatal injury and created an immediate need for medical attention while the flight crew were still bringing the damaged aircraft down.
The captain chose a five-degree flap setting because she was uncertain about the aircraft's controllability. In her interview, she described substantial drag on the controls. The normal single-engine landing guidance used a greater flap setting, and the airline's checklist did not provide a specific five-degree-flap single-engine landing speed. She selected an approach target by taking the speed for flap 15 and adding 20 knots, reaching 180 knots. The flight data showed the aircraft settling at about 185 knots during the approach.
The captain reported the airport in sight, and controllers cleared the aircraft for a visual approach to runway 27L. The tower reported wind from 280 degrees at 19 knots, gusting to 25. The crew performed the before-landing checklist, and the flight attendants repeatedly instructed passengers to brace. The aircraft touched down at about 11:20:30, approximately 17 minutes after the engine failure, with the flaps still at five degrees and a speed of about 171 knots.
Spoilers extended immediately after touchdown, and reverse thrust was used on the right engine. The aircraft left the runway through a high-speed taxiway and stopped near a fire vehicle. The crew advised passengers to remain seated and follow cabin instructions. Firefighters examined the damage and reported no external smoke or fire. This was a successful diversion and landing of the damaged aircraft, but it could not reverse the injury caused earlier by the window departure. The final report records one passenger fatality and eight minor injuries, rather than describing the landing outcome as an accident without casualties.
After landing, the captain advised emergency personnel that injuries required attention as soon as possible. The pilots completed the shutdown checklist, and the captain briefly entered the cabin. Firefighters arranged buses and paramedics, and the cabin crew disarmed the doors. An announcement asked that the aisle be kept clear for medical responders. The injured passenger remained the immediate medical priority while preparations were made to take the other occupants off the aircraft.
Passenger disembarkation began through airstairs at about 11:52. The report reconstructs these events from the cockpit recording and emergency communications. It also records an attempt to preserve the cockpit recorder in which the captain most likely pulled the flight-data-recorder circuit breakers instead. The cockpit voice recorder consequently continued recording. Investigators noted the potential for a continuously recording device to overwrite useful evidence, although that did not deprive this investigation of the relevant information.
The post-landing sequence is separate from the cause of the window failure, but it explains how the flight ended and how investigators obtained evidence. The aircraft remained on the ground while medical and ground transport arrangements were completed. The passengers did not use an immediate slide evacuation. In the cabin, the attendants had earlier left their jumpseats to provide aid; the NTSB found that their lack of restraint during the emergency landing was not a factor in this accident's outcome, but still identified a training issue. A report can recognise effective assistance while also documenting a precaution that should be improved for another emergency.
The NTSB identified a low-cycle fatigue crack in the dovetail root of fan blade 13. The fracture originated in the area of high stresses from normal engine operation. Examination found the material and microstructure consistent with the specified titanium alloy and no material defect at the origin. The blade had accumulated 32,636 cycles since new. The investigation connected crack initiation to higher-than-expected stresses at the dovetail, rather than treating the failure as an unexplained instantaneous break in an otherwise unchanged blade.
When the blade separated, it struck the fan case at a location particularly important to the fan cowl's structural response. Loads passed through the radial restraint fitting into the cowl, and the associated stresses exceeded the damaged structure's remaining strength. Portions of the cowl left the aircraft. The inboard aft latch keeper hit the fuselage near seat 14A and caused the window to depart. That structural chain, including cowl-component separation, is the NTSB's adopted explanation of the fatal injury.
The blade's crack had not been detected during its earlier overhaul. The report states that the reason was unknown and discusses inspection limitations. Subsequent visual inspections during lubrication had also found no crack, while the likely crack size was below the visual detection threshold. The accident engine did not fall within the applicability of the ultrasonic-inspection bulletins then in use. Those details are not evidence that airline maintenance was identified as the accident cause: the final findings explicitly state that Southwest's maintenance was not a factor. They explain the limits of the inspection arrangements that existed before this failure.
What the final report changed in the safety discussion
The NTSB considered the diversion to Philadelphia appropriate given the aircraft's position, emergency circumstances and the airport's runways and firefighting capability. It also assessed the incomplete checklist work in context. The crew performed most, but not all, of the engine checklist and did not complete three other relevant non-normal checklists. The Board concluded that, in this emergency, their performance appropriately balanced procedural work with maintaining control and achieving a timely descent and landing. That judgement does not turn checklist completion into an optional routine practice.
The report's recommendations extended beyond fan-blade inspection. It called for identification of critical blade-impact locations, fan-cowl redesign, installation on new aircraft and retrofit of existing aircraft. Other recommendations sought closer coordination between engine and airframe manufacturers so that fragment behaviour and loads would be incorporated into nacelle design. The accident showed that retaining engine fragments in some directions did not, by itself, prevent damaging structural consequences elsewhere around the engine.
Cabin emergency training and guidance for an in-flight loss of seating capacity also received recommendations. The final report describes inspection requirements and actions at its publication date; recommendations for redesign and retrofit should not be read as confirmation that every aircraft has since received every change. Flight 1380's account links a fatigue crack to a specific structural failure, then to a cabin breach and a medically urgent diversion. The flight crew landed the aircraft, cabin crew and passengers assisted the injured passenger, and investigators identified design and inspection protections that needed strengthening. The passenger's fatal injury remains part of the outcome despite the safe arrival of the aircraft at Philadelphia.
A simplified reading of the investigation findings. The sequence does not replace the full report or imply that every contributing factor is shown.
Sequence of events
Eastern Daylight Time (EDT), UTC minus four hours. The failure occurs shortly after 11:03:33 in the report chronology. Expand an event to see its evidence.
10:43Departure from LaGuardia
The aircraft took off for Dallas Love Field. The first officer was pilot flying and the captain was pilot monitoring.
Did the blade itself have to penetrate the cabin to cause the fatal injury?
No. The investigation describes a chain through fan-case impact and cowl separation. A departed cowl component struck near a window, which then separated.
Why did the crew land before completing every checklist?
The NTSB found that the crew balanced high workload, aircraft control and emergency descent with the need for a timely landing. Incomplete checklist execution was assessed in that context.
Did the first officer lose control permanently during the roll?
No. The uncommanded left roll reached 41.3 degrees before the first officer brought it back under control. The aircraft subsequently descended and landed.
The report says the crack was not detected at the earlier overhaul for unknown reasons. It does not establish a single proven reason for that missed detection.
The adopted cause describes a structural damage sequence
The probable-cause finding identifies fan-blade fatigue and the structural damage sequence. Initial emergency descriptions should not replace that established mechanism.
Actions and recommendations are described at the dates of the cited records. A recommendation is not proof of implementation.
Recommended: redesign and retrofit of the fan cowl
A-19-17 through A-19-19 called for identification of critical blade impact locations, cowl redesign, installation on new production and retrofit of existing aircraft. These are 2019 recommendations, not a statement that every retrofit is complete.
Recommended: coordinated engine and airframe certification
A-19-20 and A-19-23 sought collaborative analysis of blade impact, fragmentation and resulting loads, with the findings incorporated into nacelle design.
A-19-22 called for Southwest to incorporate lessons in initial and recurrent flight-attendant training, including being secured in a jumpseat during emergency landings.
The NTSB concluded that incomplete non-normal checklist execution enabled a balance between procedural work and control, descent and landing during this emergency.
Matching dossiers for Emergency workload management: No other dossier in this collection.
Read alongside
These comparisons connect specific research topics. Shared circumstances do not establish identical causes. Evidence from both investigations is provided below.
Both incidents show why engine failure protection must consider effects on the airframe. Southwest 1380 involved fan-blade impact loads and cowl separation; Qantas 32 involved an uncontained turbine-disc burst that damaged multiple systems.
Explore the records below for additional context. They open on the publisher's site and are not all included in our report search. Source reuse terms apply.
Supporting factual reports, engineering evidence and operational records for DCA18MA142. The docket also includes submissions by investigation parties; those submissions are not adopted NTSB findings. Read them alongside the final report.
Sources and transparency
Source-based check against the NTSB final report. No independent expert review.
Source limitations
This dossier uses the NTSB final report adopted in November 2019. Initial radio reports of fire are not treated as proof of an engine fire. The earlier overhaul missed detection for an unknown reason, which is retained as an uncertainty. No cockpit transcript or report illustration is reproduced.
Left Engine Failure and Subsequent Depressurization: Southwest Airlines Flight 1380, Boeing 737-7H4, N772SW, Philadelphia, Pennsylvania, April 17, 2018
National Transportation Safety Board NTSB/AAR-19/03, adopted 19 November 2019
US federal investigation report, attributed to the NTSB. Limited excerpts for commentary; third-party material and illustrations excluded. No NTSB endorsement. Source terms / provenance.
Kevin Murphy writes the incident accounts on AirlineIncidents.com, out of a long interest in aviation safety and the lessons each investigation leaves behind.
Prepared by
AirlineIncidents editorial desk
Report citation check
Independent expert review
Not completed
Citation checks compare supporting text with the indexed report pages. They do not constitute independent expert review. How these pages are prepared.
Revision history
: Checked report citations and source hashes. Added credited photographs, geographic context, sequence diagrams, reader answers, evidence limits, report-era safety follow-up, further records, sourced incident comparisons and this editorial record.
: Added a full chronological account with section-level report references and links from the short summary.
