Lion Air Crash: Aircraft Should Have Been Grounded

  • E-Mail this Article
  • View Printable Article
  • Text size:

    • A
    • A
    • A

Indonesian investigators said Wednesday that a faulty angle-of-attack indicator on a Lion Air Boeing 737 MAX 8 wasn’t replaced or repaired prior to the Oct. 29 crash into the Java Sea that killed all 189 people aboard. In a preliminary report on the accident, Indonesia’s National Transportation Safety Committee said the aircraft was unairworthy and should have been grounded.

Although the initial report doesn’t assign definitive cause to the accident, the NTSC was blunt about the airline’s maintenance: "In our opinion, the plane was no longer airworthy and it should not have continued," Nurcahyo Utomo, the NTSC’s aviation lead, told reporters at a press briefing on Wednesday in Indonesia.

The report revealed that the aircraft, a 737 MAX 8 with only 800 hours on the airframe, experienced an angle-of-attack indicator disagree between the left and right side on the previous day’s flight into Jakarta. The crew disabled the aircraft’s electric stabilizer trim and continued the flight, with the first officer flying on what the crew deemed to be more reliable instruments.

Although the crew appears to have squawked the problem, the NTSC reports that maintenance was done on other components, but apparently not the AoA sensor, although press reports differ on this finding. The airplane had a history of unreliable airspeed and AoA indications, but the report doesn’t detail whether these were due to faulty maintenance or defective components.  

While NTSC awaits the arrival of a specialized ship to continue its search for the aircraft’s cockpit voice recorder, data from the recovered flight data recorder revealed that the pilots struggled to maintain consistent pitch, speed and altitude throughout the 11-minute accident flight. The data trace released by the safety agency showed that faulty AoA indications activated the left-side stick shaker almost continuously shortly after takeoff.

The faulty AoA indication also invoked an automatic background trim system Boeing installed on the MAX series called MCAS, for maneuvering characteristics augmentation system, according to the FDR data. Because the MAX aircraft have heavier engines, the center of gravity is biased more forward than on previous models and MCAS is intended to improve pitch feel and provide stall protection. It activates automatically in manual, flaps-up flight at high angles of attack and high load factors, adding nose-down stabilizer trim at a rate of 0.27 degrees per second to a maximum of 2.5 degrees. MCAS will continue to add nose-down trim until the high AoA is resolved.

The system can be temporarily defeated by trimming against it using the yoke-mounted trim switches. But it will resume nose-down trimming after five seconds. The 737 series is equipped with stabilizer trim cutout switches and if activated, the airplane’s electric stabilizer trim is disabled, which also disables MCAS.

The crew that experienced similar problems on the previous day’s flight did use the trim cutout switches, the NTSC said. Because the CVR hasn’t been found, it’s unclear why the accident crew didn’t do the same. "We need to know what was the pilot discussion during the flight. What was the problem that may [be] heard on the CVR. So why the action difference, this is the thing we need to find. At the moment I don't have the answer," said the NTSC's Utomo.

Released flight data shows that flight JT610 flew an inconsistent flight path characterized by pitch, altitude and vertical speed excursions. The captain’s stick shaker activated shortly after takeoff and remained almost continuously active. The captain appears to have used manual nose-up electric trim against MCAS’s nose-down trim at least 30 times, according to the FDR data. Nose-down yoke forces appeared to have peaked at just over 100 pounds. 

Although it’s not implicated as a crash cause, pilot unions in the U.S. have complained that Boeing provided no detailed documentation on MCAS and the system wasn’t covered in differences training for crews transitioning from earlier 737 models. Shortly after the Lion Air crash, the FAA published an emergency AD (PDF) detailing potential fault indications in the AoA, airspeed and control feel systems used on the 737 MAX. Boeing said existing runaway trim procedures—essentially using the stabilizer trim cutouts—could address these problems. The 737 still retains manual trim control wheels, a relative rarity in modern airliners. 

In a statement, Boeing said it’s taking every measure to fully understand all aspects of the accident and continues to maintain that the 737 MAX 8 is a safe aircraft.

Comments (11)

So, from what I gather, if there was no AOA system / indicator there would be no accident. Is that assumption correct? If not, someone please explain.

Posted by: Tom Cooke | November 28, 2018 9:31 AM    Report this comment

Tom:
It APPEARS that the crew failed to disable the electric trim that drives the horizontal stabilizer jack screws. The MCAS system employs the electric stab trim. The AoA sensors themselves - although, in this case, apparently defective - didn't directly cause the accident. The crew on the previous flight did the right thing; the fatal-flight crew likely didn't.

Posted by: YARS (Tom Yarsley) | November 28, 2018 10:20 AM    Report this comment

One crew reported a problem and THEN there was bad maintenance performed followed by a crew that operated at a level that was less-than that of the previous crew who had just flown it successfully. I'm sure the cut-rate airline did not like that report!

Posted by: Mark Fraser | November 28, 2018 10:22 AM    Report this comment

It's clear this aircraft should not have flown until the AOA indication system was repaired, but it did. This failure mode appears to have been considered, so a crash wasn't inevitable even though the plane shouldn't have been flying. Improper MCAS activation should present to the crew as uncommanded stab trim movement, something for which they train regularly. The response for most aircraft, including the 737, is an immediate action checklist, which the crew must be able to recite from memory. Had this crew completed the required immediate actions checklist they could've landed uneventfully.

Unless/until the CVR is recovered we'll never know why this crew didn't respond appropriately, and maybe not even then.

Posted by: Mark Sletten | November 28, 2018 11:32 AM    Report this comment

The aircraft was completely safe to fly, even with defective AOA sensor data, so long as the crew knew, as the first crew did, to activate the stabilizer-trim-cutout switch. It seems the fatal crew did not know about MCAS or the link to horizontal trim, because in the situation they faced, they should have inactivated the autopilot, and when that didn't work, they would have disabled the stabilizer trim, too, and anything else linked to the horizontal stabilizer control surface.

Yes, Boeing should have included the MCAS-trim system in the difference briefs, but the bottom line is that the captain was responsible for knowing the aircraft systems, especially automation and how varying modes and systems interact, because the automation installed to eliminate pilot errors has already gleaned a reputation for causing fatal accidents and is the most complicated aspect of flight management today, where one not only has to know what a system is programmed to do, but how what it does changes in various modes.

It was a new model! The first thing in a pilot's head should be, "What's different here?" followed by a familiarity scan of every control group in the cockpit and a study of the manual.

Posted by: Malcolm Kantzler | November 28, 2018 1:22 PM    Report this comment

Malcolm, your second paragraph is kind of where I was driving at. It appears we have gotten to a point in our zeal to eliminate all accidents that the very systems we are employing are causing accidents. To many systems to know and manage appear to be lending themselves to pilot overload.

Posted by: Tom Cooke | November 28, 2018 2:31 PM    Report this comment

And why isn't there some light on a warning panel or a warning on one of the screens in the design such that they would notice that the MCAS system was engaging? So not only did the MCAS system not get briefed in difference training, the second crew didn't respond correctly to what should have been obvious ... that either an autopilot function or a trim function had gone awry.

Given this new info, I'd say Boeing is partially off the hook although lawyers will gloss over the facts and blame them anyway.

Posted by: Larry Stencel | November 28, 2018 5:20 PM    Report this comment

As I understand runaway trim, the trim function runs in one direction or the other ending up in either full up or down trim. And, in most cases, the trim indicator(s) will follow the the direction the trim goes. Hopefully, you can figure out quickly the direction the runaway trim is going, disable it, and manually over-ride and disconnect the trim runaway to avoid a stall or nose-dive, eventually manually re-trimming for level flight.

However, if you did not know about MCAS on board, and you temporarily get some sort of semblance of level flight, but find the airplane either climbing OR diving every 5 seconds, giving the impression that something is constantly re-trimming or moving the pitch controls potentially in both directions based on attitude, airspeed, and G forces...I can see how an experienced crew would get behind the airplane. It looks from the bar graphs supplied, these up AND down excursions were pretty aggressive. MCAS may have been programmed to respond in .027 degree per second increments but that does not mean it could or did not exceed those on this flight.. Was the MCAS driving the inaccuracies or the inaccuracies driving the MCAS? We don't know. With potentially faulty or split AOA readings, inaccurate airspeed indications, all driving an unknown on-board system, it was a perfect storm for an accident

Posted by: Jim Holdeman | November 29, 2018 10:09 AM    Report this comment

Without maintenance knowing about this system and our lack of knowledge, arm-chair quarterbacking, neither they or us have any idea how MCAS is interacting in the accurate diagnosis of the problem when the airplane was on the ground. Sure, it was supposed to be disabled when on the ground. But was it? How do you troubleshoot a system you don't know exists? How do you "learn" a system you don't know is on the airplane? And if all you have done in the maintenance of the airplane based on previous gripes is now, by the book standards, fixed, why would you not return the airplane to service?

It is obvious, that know one really knows how all of this interacted regarding proper maintenance troubleshooting, pilot responses, and a probable cause. Did the previous crew have a problem right at lift off and hand fly the airplane for the remainder of the flight? Or did this airplane behave badly at the end of the flight? We don't know. But we do know that the last crew had trouble immediately after take-off. And it was a bad situation going rapidly to worse.

The first arm-chair quarterbacking response is Lion Air, its flight crew, and maintenance operation is a foreign,cut-rate airline. Therefore, all information is filtered through that prejudiced. Boeing is going to run with that ball to make sure to feed that inclination, deflecting as much as they can any potential MCAS interference. Boeing has a lot at stake with this MAX 737 and a huge PR problem with domestic airlines regarding this previously unknown MCAS system. So, the easiest financial way out is accuse Lion Air guilty of being a cut-rate airline, demonstrating sub-standard maintenance practices, providing poor pilot training, and lack of concern for its unsuspecting passengers. "Pay no attention to that "MCAS" behind the curtain, for i am the wonderful, powerful OZ."

Think about that the next time we climb aboard Southwest and their new MAX 737. Now, at least we can assume the crew and maintenance FULLY knows about the MCAS behind the curtain. Or do they?

Posted by: Jim Holdeman | November 29, 2018 10:11 AM    Report this comment

Interestingly, the latest report regarding the Lion Air crew's performance did not say they failed to perform the runaway stabilizer trim check list nor does it state the crew failed to turn off the stab trim switches. At this point no one positively knows. Indeed they could have performed these procedures.

This same report explains that the maintenance logs for the accident aircraft recorded problems related to airspeed and altitude on each of the four flights that occurred over the three days prior to Flight 610. The logs indicate that various maintenance procedures were performed, but issues related to airspeed and altitude continued on each successive flight. The logs indicate that, among other procedures, on Oct. 27, two days prior to the incident flight, one of the airplane's Angle of Attack (AOA) sensors was replaced.

And after each round of maintenance, there was a discussion between maintenance and the pilots letting the crew known what maintenance they performed and why. On each succeeding flight the crew felt the maintenance was appropriate resulting in their decision to fly the airplane. However, no matter what maintenance was performed, the attitude and airspeed anomalies persisted.

The ill-fated last crew also consulted with maintenance prior to flight and felt in light of the latest round of maintenance performed, the airplane was safe to fly. In all of the previous maintenance performed, not one person knew of MCAS. Of all the previous four flights, the procedures used to deal with the erroneous airspeed and attitude were performed without the knowledge MCAS was on board.

It appears it started with erroneous airspeed and attitude reporting but aircraft flight handling performance did not change until the flight prior to the accident flight. Like a modern car with ever increasing or multiple fault codes, all you see is a single check engine light. And until you get series of fault codes, the car appears to run just fine. So, you keep driving. but get multiple codes stored in the ECU memory, the right combination or multiple code overload causes the car to quit, in many cases by-passing "limp mode".

It appears there was a sudden but ever escalating issues relating to how MCAS was reacting to the ever increasing, multiple airspeed/attitude anomalies. The previous flight the crew was able to disengage the stab trim and manually trimmed the airplane to an uneventful landing. It also appears that their pitch excursions were not as radical and it did not trigger the stick shaker. But like a car with multiple fault codes, MCAS input became far more aggressive, potentially reacting to all the previous erroneous attitude/airspeed data on top of what additional erroneous data it was currently accruing. Without maintenance knowing MCAS was in this data stream, they did not and could not "wipe " the faulty data from the system prior to the next flight. Thus, things go from bad to worse.

Posted by: Jim Holdeman | November 29, 2018 2:40 PM    Report this comment

In my experience, learning to fly a G1000/KAP 140 equipped C 172, I found that I could put the airplane and myself in a situation where the auto trim would work against my expectation. My instinct is to disengage and hand fly, including the experience of needing to pull the breaker when the disengage button failed to respond.

If I remember correctly AvWeb's early reporting suggested filing this very sad situation under Automation Bias. I have now accumulated 20 hours, including transition in a new SR 22, an aircraft built around automation. It's never crossed my mind to fight a system I can disable. But it's not hard to imagine the puzzle pieces necessary to lead a flight crew in the direction of Automation Bias. It's not hard to imagine a sales team selling features and benefits so fantastically automated as to keep some of the features a secret.

A very sad and costly lesson for all touched by this tragedy.

Posted by: Robert Murray | November 29, 2018 3:51 PM    Report this comment

Add your comments

Log In

You must be logged in to comment

Forgot password?

Register

Enter your information below to begin your FREE registration