How Lion Air's Boeing 737 Max Experienced a Near-Crash The Day Before 2018's Fatal Crash (go.com) 73
ABC News tells the story of Indonesia-based budget airline Lion Air, which had ordered over 200 Boeing 737 MAX 8s at a cost of $22 billion — and what happened on a flight the day before a fatal crash on October 29th, 2018:
[A]fter its first flight in May 2017, the 737 MAX 8 went 17 months without incident. Then, on Oct. 28, 2018, Lion Air Flight 610 from Bali to Jakarta experienced an in-flight emergency as the plane suddenly began to nosedive after take-off. "All of us were screaming like we are in a roller coaster," said Rakhmat Robbi, a passenger on the flight. "To be honest, I [was] think[ing] it's almost like my last flight and this is my last day." The aircraft nosedived four times as the pilots struggled to regain control, according to Indonesia's National Transportation Safety Committee (NTSC). A third pilot who just happened to be in the cockpit was able to help the two pilots resolve the situation and the plane landed safely in Jakarta.
However, according to the NTSC, the crew left incomplete notes about the details of the emergency. "The pilot reported that he had a problem with the speed and altitude indicated on [the] captain's side," said Capt. Nurcahyo Utomo, senior safety investigator of the NTSC. Nurcahyo said the captain failed to mention the plane's trim system had suddenly activated, causing it to repeatedly nose dive. "The pilots were able to control it," said aviation attorney Steven Marks. "They knew they had a problem. But they didn't understand exactly what the nature of the problem was."
Early the next morning, on Oct. 29, 2018, the same plane departed from Jakarta to Pangkal Pinang, Indonesia. Just 13 minutes after takeoff, Lion Air Flight 610 plummeted into the Java Sea. Authorities launched a search and rescue mission immediately, but all 189 people on board died.
The flight data recorder from Lion Air 610 revealed that the plane had gone out of control — it had moved up and down over 24 times before it finally dove into the sea at full speed. "I never knew... any case of the [sic] aircraft that fly down and up and up and down like this," Nurcahyo said. "I knew that the pilot was fighting with the plane." Nurcahyo said the NTSC asked Boeing about the kind of system on the 737 MAX that could have caused it to behave in such a manner. He said investigators were surprised to learn that Boeing had installed a flight control software program that could force the plane into a dive without the pilots' knowledge... MCAS was accidentally triggered on both Lion Air flights because a defective angle of attack (AOA) sensor had transmitted incorrect information about the position of the plane's nose. Although there are two AOA sensors on the 737 MAX, MCAS was only connected to one of them.
"It's a lack of redundancy that appears to me to be unacceptable in airplane design," said aviation journalist Christine Negroni, author of the book "The Crash Detectives..."
Boeing later told the pilots union of American Airlines it hadn't revealed the existence of MCAS in the 737 flight manual "on the grounds that it didn't want to inundate pilots with unnecessary information," according to the article.
ABC also points out that a later investigation by the U.S. Congress "uncovered internal Boeing emails that showed some employees had raised concerns about the 737 MAX while it was still in development, and that they had questioned the safety culture of the company as well."
However, according to the NTSC, the crew left incomplete notes about the details of the emergency. "The pilot reported that he had a problem with the speed and altitude indicated on [the] captain's side," said Capt. Nurcahyo Utomo, senior safety investigator of the NTSC. Nurcahyo said the captain failed to mention the plane's trim system had suddenly activated, causing it to repeatedly nose dive. "The pilots were able to control it," said aviation attorney Steven Marks. "They knew they had a problem. But they didn't understand exactly what the nature of the problem was."
Early the next morning, on Oct. 29, 2018, the same plane departed from Jakarta to Pangkal Pinang, Indonesia. Just 13 minutes after takeoff, Lion Air Flight 610 plummeted into the Java Sea. Authorities launched a search and rescue mission immediately, but all 189 people on board died.
The flight data recorder from Lion Air 610 revealed that the plane had gone out of control — it had moved up and down over 24 times before it finally dove into the sea at full speed. "I never knew... any case of the [sic] aircraft that fly down and up and up and down like this," Nurcahyo said. "I knew that the pilot was fighting with the plane." Nurcahyo said the NTSC asked Boeing about the kind of system on the 737 MAX that could have caused it to behave in such a manner. He said investigators were surprised to learn that Boeing had installed a flight control software program that could force the plane into a dive without the pilots' knowledge... MCAS was accidentally triggered on both Lion Air flights because a defective angle of attack (AOA) sensor had transmitted incorrect information about the position of the plane's nose. Although there are two AOA sensors on the 737 MAX, MCAS was only connected to one of them.
"It's a lack of redundancy that appears to me to be unacceptable in airplane design," said aviation journalist Christine Negroni, author of the book "The Crash Detectives..."
Boeing later told the pilots union of American Airlines it hadn't revealed the existence of MCAS in the 737 flight manual "on the grounds that it didn't want to inundate pilots with unnecessary information," according to the article.
ABC also points out that a later investigation by the U.S. Congress "uncovered internal Boeing emails that showed some employees had raised concerns about the 737 MAX while it was still in development, and that they had questioned the safety culture of the company as well."
Shockingly familiar article... (Score:2, Informative)
https://news.slashdot.org/story/19/03/20/1542246/pilot-who-hitched-a-ride-saved-lion-air-737-day-before-deadly-crash
Really kinda curious if the people who work here read the news they post....
Re: (Score:3)
Serious question: How much does it pay these days to be a Trump astroturfer ? Does he even pay you ? Because he's not in the habit of doing that.
Re: (Score:2)
Do you really think the president decides something like that?
That particular one got involved in all kinds of s***, so yes - in this case he may well have.
The way I remember events:
- The fact that the Lion machine had come close to crashing the previous day was known within 48 hours of the crash.
- The EU was the first to "ground" all aircraft of this type, although the FAA followed suit almost immediately.
Re: (Score:2)
The brass at Boeing didn't want the 737 MAX to be grounded. Boeing's CEO even called Trump to plead for it not to be grounded. But Trump grounded it anyway.
[citation needed]
Re: (Score:2)
Yeah, I was gonna say... didn't we read all about this specific story roughly two years ago?
And not just on Slashdot... it was all over the news, back then.
Re: (Score:2, Interesting)
Re: Shockingly familiar article... (Score:3)
Re: (Score:1)
I wonder if Slashdot isn't just one person now, and their bots.
There have always been dupes but now... ugh
Re: (Score:3)
Captain failed to mention the plane's trim system? (Score:5, Insightful)
Would that be the same MCAS that (a) the pilots were not told about and (b) could not be disabled without also disabling the power-assist to the trim wheels. The smaller wheels being of insufficient force in manual mode to trim the horizontal stabilizer. The wheels being made smaller to accommodate the larger console displays.
‘Boeing later told the pilots union of American Airlines it hadn't revealed the existence of MCAS in the 737 flight manual "on the grounds that it didn't want to inundate pilots with unnecessary information"’
No, the real reason is that Boeing didn't want to spend money on getting the plane recertified and spending money on re-training the pilots.
Yep, the same [Re:Captain failed to mention...] (Score:2)
Would that be the same MCAS that ...
Not only the same MCAA, but the same story.
This article is recycling news from two years ago.
Re: (Score:2)
No, the real reason is that Boeing didn't want to spend money on getting the plane recertified and spending money on re-training the pilots.
That's really the crux of it. Money.
Being a 1960s design that's been extended to fulfill new requirements that it was never designed for has made the whole thing a massive boondoggle by the most recent models. We're talking about a plane from before fly-by-wire systems that's never had one fitted due to the cost of the actual design, the re-certification of the plane and most importantly the re-training of the pilots.
When a new model introduces the kind of instability the 737 Max suffers from with the
Re: (Score:2)
Then management at Boeing decided that this was too expensive and that the system had to be simplified so that it only relied on a single set of sensors, not two, and hence had no redundancy. This is against the basic principles of aerospace engineering,
Quibble: in safety critical systems, you don't use two sensors. They can disagree. One is actually safer (though THIS system should have had three at a minimum, along with a clearly documented "please don't kill us, MCAS" circuit breaker).
Comment removed (Score:5, Insightful)
Re: (Score:2)
Engineers used to face jail time for shit like this. Law and order do not exist anymore.
Re:Heads should roll (Score:4, Insightful)
"Law and order" is only applied to the poor. The rich buy themselves immunity via campaign donations etc. We live in plutocracy.
If this were tied to a specific engineer who intentionally made a lazy shortcut, they probably would be subject to felony prosecution.
But in this case it appears to be corporate group-think, and the trail of the decisions leading to it was probably intentionally not written down so as no individual can be blamed; they'll just all say they remembered the verbal meetings differently. It's similar to what VW did in the smog cheat scandal.
Still, Boeing should be fined heavily and the top-most managers forced out and lose their golden parachutes for at least not being aware of problems.
Re: (Score:3)
Still, Boeing should be fined heavily and the top-most managers forced out and lose their golden parachutes for at least not being aware of problems.
This is the fundamental problem with incorporation. Corporations exist specifically to separate people from responsibility; at the top, the executives, and at the bottom, the shareholders. Neither group should ever be free from responsibility. The executives should be personally responsible for major failures like these under their watch, while the shareholders should always get cashed out LAST when a corporation goes bankrupt. You pay your money, you take your chances.
Re: (Score:1)
Actually, those with "preferred stock" often get priority over common stock-holders when the ship sinks. This is probably because preferred stock holders are wealthier, and thus bribed/lobbied their preference into the rules.
In other words, middle-class stock-holders ARE hosed.
Re: Heads should roll (Score:2)
ALL shareholders should be hosed, and there should not be preferred stock.
Re: (Score:2)
...while the shareholders should always get cashed out LAST when a corporation goes bankrupt.
That's what currently happens, at least in the US.
Re: (Score:1)
Re: (Score:2)
Really? When? Examples?
I know the code of hammurabi had something like that over 4000 years ago.
Re: (Score:3, Interesting)
Full public report from Indonesian authorities is here: http://knkt.dephub.go.id/knkt/... [dephub.go.id]
Skip to page 185, section 2.4.2. Read the whole section, but pay very careful attention to what is being said in this paragraph:
"The engineer in Denpasar provided to the investigation some photos of the SMYD unit during an installation test as evidence of a satisfactory installation test result. The investigation confirmed that the SMYD photos were not of accident aircraft and considered that the
Still not really redundant - needs 3 sensors (Score:3, Interesting)
Re: (Score:3, Interesting)
What they are adding is a "disagree" indicator/warnings to the pilots to alert them that the AOA sensors disagree. At that point, they can take appropriate action. The disagree indication was previously an option that cost a rediculous amount of money for something so simple, so many airlines omitted it to save money.
Anyhow, most aircraft only have two AOA sensors. If the pilots are alerted to the fact that they're unreliable, the pilot can usually figure out which one is bogus pretty quickly (typical failu
They should just get rid of it entirely (Score:2)
How hard can it be for a pilot to learn how to fly the aeroplane without it?!
These semi-intelligent systems have caused many problems. Things like Autothrottle which automatically controls the airspeed on descent, so pilots get lazy and no longer monitor it have caused crashes as well. Either the computer flys the plain, or the human does.
Re:They should just get rid of it entirely (Score:5, Interesting)
Not hard at all.
The problem is, MCAS was installed to make the flight characteristics almost the same as the 737NG, a pre-requisite for gaining support from Boeings largest 737NG customers.
If the aircraft required training as a new type, airlines such as Southwest now have a huge problem - there are only so many simulators available for pilot training, which means that their pilot pool would be split into two for a long time: those pilots who can only fly the 737NG and those pilots who can fly the 737NG and MAX.
Not only is this retraining costly, both in terms of time and money, but it creates scheduling issues - with no retraining required, you can fly a MAX into any airport in the US and any pilot can fly it out again. With retraining required, you don't have that freedom and now need to do some serious juggling to ensure that a MAX doesn't have to sit at an airport after the only pilot available to fly it has time out for the day.
If you are going to require retraining on a new type, then thats just removed a fairly significant barrier to your customer looking elsewhere for their next batch of aircraft - not requiring retraining was a huge positive to getting airlines such as Southwest from not looking at the Airbus A320NEO...
So yes, its possible to remove it and have a pilot learn to fly the aircraft without it, but there are reasons why that wasnt done. As it stands, the training required for conversion to the MAX after the grounding only stands at a couple of hours maximum in a simulator - significantly less than needing to convert to a MCAS-less MAX or an A320NEO, but still significantly more than no training at all...
Re: (Score:1, Funny)
I suspect the issue is that excessive regulation
You've nailed it. Excessive regulation is the real reason why both 737 Maxes crashed.
If only Boeing had had less regulation, less oversight from the FAA, those disasters could surely have been averted. Boeing shouldn't have any pesky regulation at all - it's Airbus that need more regulations, not Boeing!
Re: (Score:2)
I suspect the issue is that excessive regulation means that a quite small change in handling requires extensive type training that is not necessary. So the answer is to put a totally unsafe module into the airlines to avoid that. Rules beating common sense.
The regulators are always going to be blamed when the company in question refuses to take responsible for being honest or sensible. Even the FDA is extremely dependent on pharma companies being highly honest and competent -- unless there are serious injuries in a final trial or when released to the public, it is not implausible to just sweep serious issues under the rug.
The rules here are simple enough. A redesigned aircraft can be fast-tracked in terms of certification and training if the handling of the
Re: (Score:2)
I suspect you will find that the rules are simplistic rather than simple.
"Same" means no training, "Different" means extensive training, with no concept of "Little bit different". Which pushed Boeing to introduce the MCAS.
(I am not saying that people should not be shot for having a system like MCAS that 1. is dependent on a single sensor and 2. cannot be easily overridden by pulling on the stick and 3. was not documented. But I also think the plane might be better off without all these clever aids that de
Re: (Score:2)
Most modern airliners have at least three AOA sensors. Many have four or five.
Re:Still not really redundant - needs 3 sensors (Score:4, Interesting)
Look up the tech documents or just pick a photo and count them. All Airbus airliners have at least three AOA sensors, starting with the very first A300. The A350 and the A220 have four, and so does the Embraer E-jet.
Only Boeing uses two.
Re: (Score:2)
And my understanding is that the two AoA sensors are in mirror locations on the aircraft, is this accurate? If so, both sensors may well be subject to the same sorts of failures caused by siting. The third sensor should be installed in a different location.
Re: Still not really redundant - needs 3 sensors (Score:2)
All Airbus planes are triple redundant in their fly by wire systems, so it's always been three AOA sensors.
Re: (Score:1)
Re: (Score:2)
Having it documented and included in required training for type rated licenses is more than plenty and should have been done from the beginning.
MCAS only activated when throttle was at max, gear was up, flaps were retracted, and when data from two AOA sensors indicate trim needed adjusted automatically.
Now if both sensors disagree with each other MCAS deactivates and the crew receives an alert.
But let's say that both sensors fail in an identical manner and report the same bad results and the trim begins to
Re: (Score:2)
Re: (Score:2)
I thought I read it somewhere but I can't find it now. These articles were more plentiful shortly after the crashes. Some articles I find now mention "take-off throttle" which is near maximum in a fully loaded plane. I think where I read it was in a black box analysis where pilots move the levers to maximum in an attempt to save the aircraft but it was causing MCAS to fight back harder.
Re: (Score:2)
Re: (Score:2)
Re: (Score:1)
"Regular" planes don't have 3 sensors, so why does this make the Max worse than them?
If the two sensors don't agree, the pilots pretty much have to look out the window and/or use proxy factors to guess.
In medicine, this is called a "near miss". (Score:3)
And the thing that separates a good hospital from a bad one is how they respond to a near miss. A high performing hospital treats a near miss as a very serious event that must be addressed promptly. A low performing hospital looks at a near miss this way: "no harm, no foul".
Looks like Lion Air's fault to me (Score:3, Interesting)
Lion Air knew this airframe experienced an unexplained pitch down of the nose and the plane was still allowed to fly the next day? WTF? Did they investigate this problem in any way? Were any corrective measures taken to assure this did not happen? If anything was done then it looks like they didn't do enough.
Even though this plane had MCAS the pilots were trained on how to manage a runaway elevator trim problem. First thing is to turn off the automated trim. AND LEAVE IT OFF! The pilots failed to follow procedure, and in turning MCAS back on only made a bad problem worse.
Boeing certainly screwed up with MCAS. It should not have been allowed to trim past the point to which pulling back the stick would prevent the nose from coming up. I consider the lack of a redundant AOA input as a debatable problem. Having two inputs would make such failures far less likely but would still leave the problem of too much trim for a pilot to override with stick input. How should MCAS react to AOA mismatch? Trigger an alarm and disable itself? If the trim was set too far then it could disable itself in a position that is dangerous.
I believe that there was far too much talk about the lack of a redundant AOA input to MCAS and not near enough that the MCAS was able to set the trim so far from that it left the pilots no room to recover.
I recall reading an article about MCAS where someone proposed implementing a "virtual" AOA sensor based on calculations from other inputs to the flight control system. This could be in addition to having both AOA sensors or as a means to verify a single AOA indication was correct. I don't recall how this virtual AOA indicator works but if implemented, and it works as claimed, then it could detect if both AOA sensors failed. This is important because, again, this MCAS failure could still happen if there is an AOA disagree and MCAS sets the trim incorrectly.
What are the odds ... (Score:2)
Re: (Score:3, Informative)
“Even though this plane had MCAS the pilots were trained on how to manage a runaway elevator trim problem. First thing is to turn off the automated trim. AND LEAVE IT OFF!”
It was MCAS that caused the pitch down, the pilots were unable to d
Re: (Score:3)
It was MCAS that caused the pitch down, the pilots were unable to disable MCAS.
Disabling the automated trim disables MCAS. That was part of the problem pointed out during the analysis of the crashes. The trim motors were on the same switch as MCAS. Without trim motors the trim would have to be adjusted manually using a crank on the flight deck, this is not easy when the flight crew is also trying to pitch up the nose with the stick as there would be considerable force on the elevators at this point. MCAS is easily disabled, there is a switch to do just that. Once MCAS puts the el
Re: (Score:1)
Re: (Score:2)
Why in the world were the trim motors and the MCAS tied together in the first place? Do the pilots not have control of the trim except through MCAS? That alone seems to be an unacceptable hazard. Did Boeing get this idea from Apple? Maybe also tie the autopilot and main hydraulic pumps together in one switch? The cockpit must certainly be uncluttered. Also, in my opinion
is not great,
Re: (Score:1)
Why in the world were the trim motors and the MCAS tied together in the first place?
Likely because they make up a single system in the minds of the pilots and engineers. MCAS was sold as an addition to the old STS to the FAA and the airlines. The use of a single switch to disable the trim motors, STS, and MCAS was to avoid adding more switches, adding more training, and therefore adding more work for the FAA, Boeing, and the airlines. Oh, and another switch adds more work for the pilots. If there's a failure in STS or MCAS then there might not be time to think through which is failing,
Re: (Score:2)
I totally agree that the redundant input, while certainly good and frankly something that should have been there in the first place, is not the ultimate issue. The problem is not so much preventing the system from failing as preventing its failure from crashing the plane. I agree with your take that the MCAS shouldn't have been allowed to do what it did, period.
I do stand by my view, sarcasm and all, that an active system like MCAS should be designed so that it can be shut off separately from power assist
Re: (Score:1)
That is a very enlightening link. Thank you very much. I don't recall ever seeing the function of those switches described in such detail.
Changing the function of the switches was something of a dick move by Boeing. That did remove options for the crew. Without that separation of switch function the only other way to disable MCAS and retain power to the trim motors (at least as I am aware) is to deploy the flaps. The use of flaps in level flight (not in takeoff or landing) was authorized by the FAA to
Re: (Score:2)
Thank you as well! I learned a lot from your comments, I certainly had no idea the systems were tied together. I'm neither a pilot nor aircraft engineer, but I do have an interest in control systems so I've tried to keep up with this whole debacle. (I've also been on the operator side of operator error, so I have strong feelings about the subject) I was curious what the switch looked like and where it was, and that's when I found the article.
I really enjoy these kinds of discussions here, and they are rare
Re: (Score:2)
What's not mentioned is how the FAA allowed Boeing to rubber-stamp safety reports.
I've seen this mentioned in about every /. conversation on the subject so far. Boeing was permitted to self-certify, which means the government wasn't even pretending to do its job here.
Re: (Score:2)
That was the problem - they didn't know. The report of the incident from the pilot didn't include that point:
> "The pilot reported that he had a problem with the speed and altitude indicated on [the] captain's side," said Capt. Nurcahyo Utomo, senior safety investigator of the NTSC. Nurcahyo said the captain failed to mention the plane's trim system had suddenly activated, causing it
Re: (Score:2)
That was the problem - they didn't know. The report of the incident from the pilot didn't include that point:
Then that is pilot error. Perhaps not the same pilot that was on the fatal flight but still pilot error. This then get's back to my point of what might be considered a failure of Lion Air to train pilots properly.
If what you claim is true, and I have no reason to doubt it, then the pilot did report a failure in the air speed and altitude indicators to those responsible for aircraft maintenance. I don't know what is considered acceptable failures in equipment and the aircraft still being considered safe t
Re: (Score:2)
That was the problem - they didn't know. The report of the incident from the pilot didn't include that point:
Then that is pilot error. Perhaps not the same pilot that was on the fatal flight but still pilot error. This then get's back to my point of what might be considered a failure of Lion Air to train pilots properly.
Confused humans sometimes latch onto the wrong thing, and miss the most critical factor that needs to be fixed. Boeing hid the very existence of MCAS from the pilots on purpose. It should not be a shock to anyone that when you hide information from the pilot the pilot may become confused. The pilots are not blameless, obviously, but the aircraft as delivered by Boeing was fundamentally flawed and should never have been certified even if the pilots were all trained perfectly. One simple sensor failure sh
Re: (Score:2)
It should not be a shock to anyone that when you hide information from the pilot the pilot may become confused.
This has been brought up in previous threads, but in some ways, it's reminiscent of the air conditioning configuration changes between 737-200/300 and the -400... various factors (especially the aforementioned air conditioning source change and assumed unreliable vibration monitors) lead an experienced pilot to cut the wrong engine, after assuming that smoke in the cockpit was coming from a particular source. https://en.wikipedia.org/wiki/... [wikipedia.org] My opinion, however, is that Kegworth seemed less like an outrig
Re: (Score:2)
At least with Kegworth, one could reasonably argue there was no change in how the plane "flies" at all, albeit confusion about the altered feedback from an emergency condition turned out to be deadly serious. But nobody would say that any of the individual changes to the aircraft were wrong or inherently more dangerous in their own right. The pilot could have spent more time assessing the situation in that emergency at a very small cost in risk.
IMHO the purpose of MCAS was very clear: Boeing was purposefu
Re: (Score:2)
Confused humans sometimes latch onto the wrong thing, and miss the most critical factor that needs to be fixed. Boeing hid the very existence of MCAS from the pilots on purpose. It should not be a shock to anyone that when you hide information from the pilot the pilot may become confused.
The pilots may have been confused but the protocol for trim motor failure is to turn it off AND LEAVE IT OFF. MCAS, STS (stabilizer trim system), and the trim motors are on one circuit meaning that there is a switch to disable them and pilots were trained on where this switch is located and when to turn it off. MCAS failure would appear to the crew much like a stuck trim switch. Not EXACTLY like a stuck switch but similar, which can lead to confusion. There is a protocol for this and it was unchanged fr
Medal (Score:2)
Shouldn't which of the pilots that managed to workaround the issues get a medal or something?
Re: (Score:2)
Shouldn't which of the pilots that managed to workaround the issues get a medal or something?
More importantly, whichever executive was in charge of MCAS (and there should have to be one) should get some jail time, equal to the sum of the life lost by all passengers killed. That's what the corporate types really fear — responsibility. That's why there are so few great men in business any more — it's now possible to succeed in business even as a coward and a pisspot.
Recertify for 3 person crew (Score:2)
Re: (Score:2)
Agreed. Honestly, the 737-MAX is probably just fine as an aircraft without MCAS assuming the crew is trained properly. Boeing lied in order to cheat on training, so that customer costs would appear to be lower than they truly were. Yes, as you say, the free market is ready to kill the 737, based on the relative merits of comparable aircraft.
All I need to know (Score:3)
And there friends, is why I'll have nothing to do with Boeing any more. It is run by accountants, and lawyers who make really bad decisions.
Inundate the bejabbers out of me if something that can kill and has proven it can kill is going to possibly kill me if I don't know about it.
Boeing should have everything grounded, their entire management staff relieved of duties, and have people that aren't idiots run the place. They are not trustable, and cannot be trustable until that happens.
Re-train *now* (Score:3, Insightful)
MCAS was 'invented' to artificially make the aircraft similar enough to a 'normal' 737 so that the 737 type rating would apply.
If there's anything airlines have at their hands at present, it's plenty of time to re-train their pilots.