Lion Air crash: Jakarta Boeing 737 'had prior instrument error'

[BravoV]

No one wants to sink Boeing, right?

One of the biggest player of the military–industrial complex ? Are you kidding ? 

[Mr. Scram]

One of the biggest player of the military–industrial complex ? Who are you try to kidding at ? 

They're too big to fail and don't want the MAX cancelled. Others may think the aircraft should go, so what do you do? You raise the stakes by building as many as you can. They're basically saying "What are you gonna do about it?" That should give anyone pause what Boeing's intentions and lessons learned really are.

[BravoV]

One of the biggest player of the military–industrial complex ? Who are you try to kidding at ? 

They're too big to fail and don't want the MAX cancelled. Others may think the aircraft should go, so what do you do? You raise the stakes by building as many as you can. They're basically saying "What are you gonna do about it?" That should give anyone pause what Boeing's intentions and lessons learned really are.

The ease and profit of selling civilian product, non military planes like MAX before this fiasco, is much-much more preferable method in making money, much easier and much less stressful, rather than the prolong highly stressful environment, including constant head-ache on maintaining the bribe .. errr ... lobbying the Senate, House of Rep and top brasses at Pentagon and etc for military purchase even they're member of military industrial complex. Also compared to these bunch of big boys, "maintaining" just FAA top officials alone is a walk in the park.

[james_s]

I think the Max will be fine after this, the plane is mostly sound, it was just the idiotic design of that MCAS feature that doomed the two that went down. Other planes in that past have had teething pains, the DC-10 suffered several high profile crashes early on which it never fully recovered from, but it still went on to provide decades of reliable service once the kinks had been ironed out. I do not anticipate seeing any further 737 Max crashes related to MCAS, especially now that everyone is acutely aware of the system and what it does.

[madires]

Have you forgotten that MCAS is the band aid for a make-it-cheap design problem?

[SilverSolder]

Have you forgotten that MCAS is the band aid for a make-it-cheap design problem?

I thought it was just there to make the plane "feel" the same as an earlier model, to avoid recertification and retraining pilots?

[james_s]

I thought it was just there to make the plane "feel" the same as an earlier model, to avoid recertification and retraining pilots?

That's exactly what it's there for. The plane is completely flyable without it, but it has different characteristics than other 737 models so they added MCAS to automatically compensate for these differences so that a pilot trained to fly a 737 could in theory fly the Max without any additional training. Pilots are certified to fly specific aircraft types, different aircraft have different flight characteristics, that doesn't mean there's anything wrong with the plane, it just means the pilots have to be trained on the new model, or you need a system like MCAS to automatically compensate to make the new plane fly like the old plane from the pilot's perspective. There's nothing inherently wrong with this, it's not even a particularly difficult problem to solve, they just screwed up hugely, got cocky and complacent.

This has happened before, the DC-10 I mentioned had three independent hydraulic circuits, the calculated odds of all three systems failing were so low that there was no backup to these redundant hydraulic systems. Unfortunately they overlooked the fact that all three systems pass within close proximity to the #2 engine mounted in the vertical stabilizer and an uncontained engine failure could take out all three systems as happened to United flight 232.

Another classic textbook case of arrogance and complacency is the Therac-25 radiation therapy machine, calculated odds of software failing were so low that the machine lacked any sort of hardware interlocks, however they overlooked the possibility of bugs in the code and a race condition resulted in the horrific suffering, disfigurement and deaths of a handful of people by severe radiation burns before anyone realized what was going on. After fixing the software and adding hardware interlocks the Therac machines went on to provide decades of service without further injuries.

The same will be true of the 737 Max, I suspect the tragic accidents will hang over them forever and limit the popularity of the aircraft to much less than it could have otherwise been, but I have no doubt it will prove to be as reliable and safe as previous 737 variants. I would not hesitate to fly on one.

[madires]

The new fuel efficient jet engines are larger and Boeing decided to move them in front of the wings to avoid a redesign of the landing gear which would harm the simplified certification process. That placement of the jet engines changes not just the flight characteristics, it also increases the probability of stalling. MCAS is meant to mitigate the risk of stalling. Media/press reported these details at the beginning of the investigations.

[james_s]

The new fuel efficient jet engines are larger and Boeing decided to move them in front of the wings to avoid a redesign of the landing gear which would harm the simplified certification process. That placement of the jet engines changes not just the flight characteristics, it also increases the probability of stalling. MCAS is meant to mitigate the risk of stalling. Media/press reported these details at the beginning of the investigations.

No, sorry, that's not correct. ALL jets that have the engines slung under the wings will tend to pitch up when throttle is increased, that is an inherent characteristic of this design. The Max moved the engines and has larger engines so it exhibits this characteristic to a greater degree. Excessive pitch up at low airspeed can induce a stall, but if the pilot is aware of how the aircraft handles then they know how to avoid doing this. The Max does not need MCAS to fly, it needs MCAS to have the same amount of pitch-up tendency under throttle as the earlier 737 models so that the different flight characteristics are transparent to the crew. ALL aircraft have unique characteristics that will get a person into trouble if they don't know what they're doing, that is precisely why pilots of large aircraft have to be certified to fly that specific model of aircraft. Boeing could have decided to skip MCAS entirely, then released exactly the same plane without it and it would have been  perfectly fine aircraft, except crews certified to fly the earlier 737 models would require additional training to fly the new model and customers wished to avoid this.

This is one area where Airbus has an advantage because their planes are already entirely fly by wire so there is no need for a separate system, the human pilot already has no direct control over the control surfaces, the software is tuned to provide the desired handling characteristics.

[Nusa]

This is one area where Airbus has an advantage because their planes are already entirely fly by wire so there is no need for a separate system, the human pilot already has no direct control over the control surfaces, the software is tuned to provide the desired handling characteristics.

With all sorts of fallback modes (automatic or commanded) that have to be learned for each major Airbus type (they have different type ratings, not one for all). It's very different from Boeings approach. Whether it's better is the subject of many discussions. Also, it's not quite right. Airbus does have final fallback mechanical mode on engines, trim tabs, and rudder, intended to give enough control to keep the aircraft level during a complete computer failure and reset. If they don't get at least one computer back fairly quickly, it's very serious.

[madires]

The Max does not need MCAS to fly, it needs MCAS to have the same amount of pitch-up tendency under throttle as the earlier 737 models so that the different flight characteristics are transparent to the crew.

From https://en.wikipedia.org/wiki/Maneuvering_Characteristics_Augmentation_System:

The Maneuvering Characteristics Augmentation System (MCAS) flight control law was implemented on the 737 MAX to mitigate the aircraft's tendency to pitch up because of the aerodynamic effect of its larger, heavier, and more powerful CFM LEAP-1B engines and nacelles. The stated goal of MCAS, according to Boeing, was to provide consistent aircraft handling characteristics at elevated angles of attack in certain unusual flight conditions only and hence make the 737 MAX perform similarly to its immediate predecessor, the 737NG.

[SkyMaster]

...

The same will be true of the 737 Max, I suspect the tragic accidents will hang over them forever and limit the popularity of the aircraft to much less than it could have otherwise been, but I have no doubt it will prove to be as reliable and safe as previous 737 variants. ...

The 737MAX could have been made to be as reliable and safe as previous 737 variants, right from the start; it is really unfortunate that Boeing chosen to fool the FAA, the airlines, the pilots and the flying customers.

[Gyro]

The Max does not need MCAS to fly, it needs MCAS to have the same amount of pitch-up tendency under throttle as the earlier 737 models so that the different flight characteristics are transparent to the crew.

An unfortunate choice of words. The one thing that ill conceived a implemented MCAS, and indeed the whole sordid matter of Boeing's behaviour during the investigation and recertification exercise hasn't been is "transparent".   

[jmelson]

One of the biggest player of the military–industrial complex ? Who are you try to kidding at ? 

They're too big to fail and don't want the MAX cancelled. Others may think the aircraft should go, so what do you do? You raise the stakes by building as many as you can. They're basically saying "What are you gonna do about it?" That should give anyone pause what Boeing's intentions and lessons learned really are.

There's really not that big a problem.  If the previous version of the 737 is OK, then the Max is OK, as long as EVERY pilot knows how MCAS works, what will set it off improperly (damaged AOA sensor) and how to recover (trim nose up, then hit the switches to disable automatic and electric trim).
And, DON'T disable electric trim with the aircraft badly trimmed, use your electric trim to override MCAS until the aircraft is properly trimmed BEFORE disabling the electric trim.

The real problem is that MCAS was NOT known to most pilots, and it behaved differently from a simple runaway trim that all pilots are trained for.
Thus, they didn't recognize this a a form of a trim runaway, and failed to react to stop it.

Jon

[jmelson]

An unfortunate choice of words. The one thing that ill conceived a implemented MCAS, and indeed the whole sordid matter of Boeing's behaviour during the investigation and recertification exercise hasn't been is "transparent".   

Yes, it is pretty hard to understand how NOBODY ever asked "what happens if the AOA sensor fails, and nobody responds to that?"
It is going to kick in MCAS right after takeoff, a high workload time, and doesn't give much of an indication in the cockpit of the 737.
Without strong emphass on training for this condition, the flight crew would be taken totally by surprise.  And, assuming they WERE
trained, would they remember that years later when the condition actually appeared?  This whole scenario where they told almost NOBODY that this
system even EXISTED is totally amazing.

And, of course, Soutwest Airlines is quite complicit in this by fining Boeing $1 million per aircraft if there is any sim time required to learn the Max.
Boeing should have pushed back on that one.

Jon

[SkyMaster]

...  And, assuming they WERE trained, would they remember that years later when the condition actually appeared?

Assuming they were trained, that would have been covered during the recurring training, every six months.

And, of course, Soutwest Airlines is quite complicit in this by fining Boeing $1 million per aircraft if there is any sim time required to learn the Max.
Boeing should have pushed back on that one.

Boeing sales pitch to Southwest Airlines most probably covered the "saving" in training cost. Now, after this fiasco created by Boeing, Southwest Airlines is not "saving" as much, in training.

Boeing lied to everybody, Boeing will probably pay these penalties to Southwest Airlines.

I have a friend who became type rated on the 737MAX, two years ago, and he confirmed to me that all the technicalities surrounding the MCAS were kept secret, from the airlines, by Boeing.

 

[Mr. Scram]

There's really not that big a problem.  If the previous version of the 737 is OK, then the Max is OK, as long as EVERY pilot knows how MCAS works, what will set it off improperly (damaged AOA sensor) and how to recover (trim nose up, then hit the switches to disable automatic and electric trim).
And, DON'T disable electric trim with the aircraft badly trimmed, use your electric trim to override MCAS until the aircraft is properly trimmed BEFORE disabling the electric trim.

The real problem is that MCAS was NOT known to most pilots, and it behaved differently from a simple runaway trim that all pilots are trained for.
Thus, they didn't recognize this a a form of a trim runaway, and failed to react to stop it.

Jon

The whole point of MCAS was to keep the 737 the same as previous models. Requiring any retraining would be a big deal when the competitor does not have such a requirement and that's what Boeing tried to prevent, even at the cost of human lives. That in itself should tell us it's a considerable issue.

[langwadt]

There's really not that big a problem.  If the previous version of the 737 is OK, then the Max is OK, as long as EVERY pilot knows how MCAS works, what will set it off improperly (damaged AOA sensor) and how to recover (trim nose up, then hit the switches to disable automatic and electric trim).
And, DON'T disable electric trim with the aircraft badly trimmed, use your electric trim to override MCAS until the aircraft is properly trimmed BEFORE disabling the electric trim.

The real problem is that MCAS was NOT known to most pilots, and it behaved differently from a simple runaway trim that all pilots are trained for.
Thus, they didn't recognize this a a form of a trim runaway, and failed to react to stop it.

Jon

The whole point of MCAS was to keep the 737 the same as previous models. Requiring any retraining would be a big deal when the competitor does not have such a requirement and that's what Boeing tried to prevent, even at the cost of human lives. That in itself should tell us it's a considerable issue.

also if NG and MAX wheren't considered "the same" an airline that has a mix of NG and MAX can't move pilots around without the added cost and time to keep a dual type rating

[floobydust]

Boeing 737 Max project is criminal conduct in a hundred ways, I don't expect this to go smoothly despite the politics.

We all know what happens when you roll out an ugly baby- a product with safety issues that are covered up to sneak through regulatory. When the problems come up (containment problem) you can't go backwards to make design changes. Hardware is like that. The "software fix" executives push for, very limited and can't cover much for deficient hardware.

Dennis Muillenburg is enjoying his $62M stock, pension , last paycheque and avoiding jail. There's no consequences.
“Boeing executives should be walking away in handcuffs, not with millions of dollars,” Zipporah Kuria, who lost her father, Joseph Waithaka, in Ethiopian Airlines flight 302, said in a statement.

Strange Volkswagen, Audi and Takata have a CEO and few token managers in criminal trial proceedings over their fiascos.

Not a peep about the AoA sensor's poor reliability, if that has been addressed. It was the problem in the first place.

The planes have been in storage an awful long time which adds new dangers.
Recent Dec. 22, 2020 Air Canada 737 Max taken out of storage, had problems "left engine low hydraulic pressure indication" then "indication of fuel imbalance", so they shut off that engine and PAN PAN landed 

[james_s]

There's really not that big a problem.  If the previous version of the 737 is OK, then the Max is OK, as long as EVERY pilot knows how MCAS works, what will set it off improperly (damaged AOA sensor) and how to recover (trim nose up, then hit the switches to disable automatic and electric trim).
And, DON'T disable electric trim with the aircraft badly trimmed, use your electric trim to override MCAS until the aircraft is properly trimmed BEFORE disabling the electric trim.

The real problem is that MCAS was NOT known to most pilots, and it behaved differently from a simple runaway trim that all pilots are trained for.
Thus, they didn't recognize this a a form of a trim runaway, and failed to react to stop it.

Jon

The whole thing was the perfect shitstorm of poor design decisions, poor software QA, politics, arrogance, complacency and carelessness that all came together and blew up. It almost any one of these things had been done differently the whole catastrophe could have been avoided. It would have been trivial to limit the authority of MCAS so that it could not wind the trim down beyond the range that could be overridden by the pilot's control input, that was an absolutely massive oversight that even a novice QA engineer should have caught. The ball was dropped at so many levels that a disaster was virtually guaranteed. If a system is intended to be fully automated to the degree that the crew is not even aware that it exists then it damn well better be rock solid with redundant everything and extensive testing. Even if no additional training is required to operate the aircraft, the crew should still be aware of the existence of the system that makes this possible.

[BravoV]

Part of the problem solved ...  a good sign for Boeing ...

-> U.S. fliers less familiar with Boeing 737 MAX crashes two years on, but wary when reminded

[james_s]

Part of the problem solved ...  a good sign for Boeing ...

-> U.S. fliers less familiar with Boeing 737 MAX crashes two years on, but wary when reminded

That's not too surprising. The same happened with the DC-10, and that involved multiple incidents with several different iterations of "fixing" the cargo door and other problems. Eventually people were flying on them and the related MD-11 for many years. The MD-80 too, I flew on one about a year after the jackscrew failed causing that Alaska Airlines flight to crash, a guy who's office was a few doors down from mine and his wife both died on that. It crossed my mind when I sat down in the plane but it didn't stop me from flying on it. I've always been fascinated by airplanes and pay attention to exactly what kind of aircraft I'm on and occasionally I've chosen flights specifically to ride on a particular type of plane, but I suspect most people are probably only vaguely aware if they know at all.

[ve7xen]

Not a peep about the AoA sensor's poor reliability, if that has been addressed. It was the problem in the first place.

That's almost beside the point. AoA sensors are going to fail. Whether 1/1,000 hours or 1/1,000,000 hours, there are (or Boeing would hope there will be) enough of these birds in the air that failure is inevitable and must be handled gracefully. Obviously a high failure rate would be a concern, but it's hardly the biggest problem here, and warrants a far less aggressive regulatory response than not reacting to AoA failure appropriately does.

According to the report on the Lion Air flight, they are speculating that the AoA failure was not due to hardware failure per se, but a nonstandard (however, approved) calibration procedure in use at the re-manufacturer, and speculate that the operator had the test equipment in 'REL' mode, introducing a bias into the AoA signal. I'm not sure the Ethiopian investigation has speculated on what caused that failure.

I've been following this whole debacle closely since the first accidents, and it just keeps getting more and more on the Boeing side. It's really pretty egregious the level of engineering and particularly management failure that happened here. From negotiating a 'no training' deal with airlines before they even understood what training would/might be required, to actively discouraging customers from giving their pilots training, to covering up the existence of the system, to the software QA issues, to the 'oversight' of this deadly failure mode in their failure analysis, ignored QA problems in manufacturing, etc. etc. etc. it's just... malicious failure all the way down.

IMO the Final Order does enough to prevent future incidences like this one, I think the MAX will be safe from MCAS at least, once they get the return from storage issues ironed out. But all of this really doesn't make me want to get on a Boeing aircraft, both because who knows what other danger is lurking behind this management, and who wants their money going to such a company?

[floobydust]

The AoA sensor is another shitstorm, hardware failure plus its undocumented relative calibration problem:
"...the output from Resolver 2 was correct when the internal heaters were operating but stopped working again when the temperature dropped.
The investigation discovered that a loose loop of the very fine magnet wire from the primary rotor coil was trapped in the epoxy which was meant to hold the end cap insulator in place. The trapped magnet wire thus adhered to both the end cap insulator and the rotor shaft insulator which had very different coefficients of thermal expansion (CTE). As the wire expanded and contracted to the two different environments, the wire became fatigued and showed multiple ridges and cracks before breaking. The wire failure created an intermittent open circuit, dependent on temperature. The sensor worked fine at temperatures above 60°C (140°F) but failed at temperatures below that."
https://fearoflanding.com/accidents/accident-reports/lion-air-610-the-faulty-aoa-sensor/

[ve7xen]

The AoA sensor is another shitstorm, hardware failure plus its undocumented relative calibration problem:
"...the output from Resolver 2 was correct when the internal heaters were operating but stopped working again when the temperature dropped.
The investigation discovered that a loose loop of the very fine magnet wire from the primary rotor coil was trapped in the epoxy which was meant to hold the end cap insulator in place. The trapped magnet wire thus adhered to both the end cap insulator and the rotor shaft insulator which had very different coefficients of thermal expansion (CTE). As the wire expanded and contracted to the two different environments, the wire became fatigued and showed multiple ridges and cracks before breaking. The wire failure created an intermittent open circuit, dependent on temperature. The sensor worked fine at temperatures above 60°C (140°F) but failed at temperatures below that."
https://fearoflanding.com/accidents/accident-reports/lion-air-610-the-faulty-aoa-sensor/

I'm not sure what you mean by undocumented relative calibration problem. This was a problem, but it was a documented and allowed procedure that was used. The calibration issue was due to the remanufacturer using a nonstandard piece of test equipment and associated nonstandard procedure, with the approval of the FAA, which included this mode that was not part of the recommended test equipment. This procedure was documented and approved, it just failed to correctly deal with the setting of this REL switch. As this wasn't a recommended procedure, I don't think you can blame Boeing or the AOA vane subcontractor for that.

Yeah, the initial failure that triggered replacement with the miscalibrated sensor is certainly interesting, but it definitely doesn't strike me as a shitstorm. Possibly a QA issue, possibly just an unexpected consequence of the manufacturing process, should be easy to fix going forward and isn't obviously a result of malice. It's not made or designed by Boeing, and I haven't seen anything implicating the sort of management problems at Boeing at their subcontractors (yes, Collins failed to catch the AOA DISAGREE alert issue, as did Boeing, but they were then ordered not to fix the problem because ??). These kind of manufacturing issues always come up with new aircraft in one system or another, get resolved, and because the designs are usually proper, don't result in crashes. Aircraft are incredibly complicated, and to expect no issues at all of this type isn't reasonable, which is why the engineering needs to account for occasional failures.

Indeed, what actually happened in this case - the problem was detected by the flight computers, the plane flew several legs safely (probably should have been grounded earlier, yet...), eventually the problem was isolated, and the part replaced - is what one would expect to happen. In fact, if I understand the fault correctly, I don't think it would be likely for this original failure to trigger MCAS at all, an open circuit is an outright failure that would be detected, and the sensor was otherwise providing correct readings. Ultimately (bar MCAS), the AOA vane is a non-critical and 'fault tolerant' system on the 737 - I can't easily find if the 737NG and 737MAX use the same AOA sensor part; I would guess that they do or one of very similar design, and NG has been flying in the 1000s without AOA vanes contributing to an accident for 20 years.

Even if you're right and this is a complete disaster of an AOA sensor design or manufacture for some reason, failure still shouldn't have had any effect on safety of flight, even complete loss of AOA data isn't an issue for any 737 (which is why it's only doubly redundant), so I think this is just an interesting footnote to the story. If the design of the system on the Boeing side wasn't such a clusterfuck, the inconsistent readings would be detected, trigger a disagree, and the faulty data wouldn't be used for anything.