Metal weaknesses like the fatigue found in a jet engine that exploded during a Southwest Airlines Co. flight on Tuesday can lie hidden for years and elude detection with even the most advanced technology.
A crack on a fan blade on the Southwest jet, for example, was on an internal surface that would not have been detectable by a visual inspection, National Transportation Safety Board Chairman Robert Sumwalt said Wednesday. The plane made a successful emergency landing, but a passenger was killed.
As recently as January, the NTSB called on federal aviation regulators to improve how some engine parts are tested during manufacture because of another engine failure that resulted in a massive fireball in Chicago two years ago.
“These are among the most difficult and most critical items that are manufactured by humans,” said John Hansman, an aviation professor at Massachusetts Institute of Technology. “There’s always some risk of imperfection and that’s partly why we do the inspections. But this is really hard.”
The safety board is looking at why one of the fan blades shot off on the Southwest jet’s CFM56-7B engine, which is part of the family of one of the most common power plants in commercial aviation. It broke the engine apart and sent metal parts slamming into the 737.
A portion of the fan blade was found in the engine. No pieces of the engine have been discovered inside the plane, Sumwalt said.
The engine type has a very good safety record but if manufacturing or design flaws were uncovered in the investigation, the agency will take action to address those in use, he said.
“We want to very carefully understand what was the cause of this problem. To be able to extrapolate that to the entire fleet, I’m not willing to do that yet,” Sumwalt told reporters on Wednesday at a briefing in Philadelphia, where the Boeing Co. 737-700 made an emergency landing on a flight from New York to Dallas.
The CFM56-7B is made by CFM International Inc., a joint venture between General Electric Co. and Safran SA.
Investigators are trying to determine whether there are any similarities to a 2016 engine failure on another Southwest 737. While that investigation hasn’t been completed, the NTSB also found evidence of a crack “consistent” with metal fatigue in that titanium-alloy blade, it said in a statement released on the agency’s website after the incident.
The engines in both cases were CFM56-7Bs.
Investigators haven’t been able to determine if the engine on the Southwest jet
had received any special inspection following the 2016 episode. Southwest announced Tuesday it was conducting ultrasonic tests on all engines of the same vintage.
The Federal Aviation Administration last August proposed additional inspections for the CFM56-70 but hasn’t finalized those yet, according to records. Late Wednesday, the agency announced plans to complete that process by requiring inspections on the engines.
If the FAA directive follows the original proposal, engines with more than 15,000 hours of use must be inspected within six months. Engines with fewer hours than that must have the inspections within 18 months. The inspections must be conducted using ultrasonic sensors designed to detect cracks beneath the surface of the fan blades.
Dyes and X-rays
Manufacturers take elaborate steps to avoid metallurgical flaws in engine components that spin at high speeds and must withstand enormous forces and high temperatures.
They use dyes to make cracks on the surface more visible, bombard metal parts with X-rays to peer beneath the surface, or send electric currents or sound waves pulsing through structures that can detect unusual features hidden to the eye.
These techniques can detect most flaws before they cause an explosive failure, as evidenced by the dramatically improved safety record in aviation in recent decades. But they aren’t always foolproof, said Cem Tasan, a professor of metallurgy who also teaches at MIT.
“The main problem is none of these techniques will give you a full picture of what is going on in the material,” Tasan said.
Because of the earlier failure, regulators may want to consider some additional tests on some CFM engines, said James Wildey, a retired NTSB metallurgical expert.
“The time is probably due to increase the inspection interval with these multiple incidents or to develop a more accurate inspection technique,” Wildey said.
In a separate case that prompted NTSB recommendations on engine safety last January, an American Airlines 767 was accelerating for takeoff when one of its jets exploded, creating a fuel leak and a fire that melted the left wing on Oct. 28, 2016. No one died despite a chaotic evacuation on the runway.
A nickel-based alloy disk that shattered in the engine was forged in 1997 and 1998, following an elaborate, multi-step process to ensure it didn’t contain a flaw, according to NTSB. It also received an ultrasonic test designed to ensure there were no hidden weak spots.
Even though such tests have been effective at detecting forging flaws beneath the surface, it didn’t detect the manufacturing error the NTSB described as a “discrete dirty white spot.” After years of use, weaknesses around the spot allowed a growing crack to form that eventually led it to split apart.
While failures on engines are rare, a 2008 report by the FAA found that “a number of rotor disk fracture and cracking events have originated from embedded anomalies,” which were not detected during production inspections.
Engine failures are rare and deaths caused by them are far rarer. The last fatality on a U.S. passenger airline because of an engine failure was almost 22 years ago, according to NTSB records.
Passengers on the Southwest flight had a rough 22-minute ride from the time of the engine failure until the emergency landing, Sumwalt said. After the engine shut down and its smooth exterior tore loose, dramatically increasing wind drag, the plane banked hard to the left by 41 degrees. That’s about twice the maximum bank that a passenger is likely to experience on a typical flight, Sumwalt added.
The passenger who was killed, Jennifer Riordan, worked for Wells Fargo & Co. as a vice president of community relations in Albuquerque, New Mexico, according to a statement released by the bank.
Preliminary reports indicate she was wearing her seatbelt when the engine failed, Sumwalt said Wednesday. The NTSB has a team of experts who will study whether the impact could have been survivable.
Philadelphia’s medical examiner concluded that she died of blunt impact trauma to the head, neck and torso, the Associated Press reported.
–With assistance from Julie Johnsson and Richard Clough
©2018 Bloomberg L.P.