Page 5 of 10 FirstFirst ... 34567 ... LastLast
Results 121 to 150 of 287

Thread: Lion Air Flight 610 Crash

  1. #121
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,439
    you would think the only reason to fit two aoa sensors would be to turn the system off if they disagree or if one dies. If reports are correct the system believed the faulty sensor instead of shutting itself down and relying on the pilots. Even having seen enough software errors in my life, it is hard to believe that mode of failure slipped through. The protocols for dealing with one , two or three sensors are old school basics. It's very sobering for a regular flier.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  2. #122
    Join Date
    Oct 2005
    Posts
    26,546
    Yes I agree, it's really hard to imagine how a design conversation could go like,
    "Should we use info from both AoA or just one?"
    "Well, there's an old saying, a man with one watch knows what time it is, a man with two is never sure."
    "But what if it's actually important to know the correct time?"
    "Don't worry, we'll give the pilots a way to turn off the power to the stabilizer if they don't trust the system."
    "What if they are just taking off, have no leeway to lose altitude, and are gaining so much speed that they cannot adjust the stabilizers manually?"
    "What are the chances of that?"
    "Um, well, shouldn't we at least warn them about this possibility?"
    "They might object and not want to fly at all. Ignorance is bliss."
    "Shakespeare actually said, where ignorance is bliss, 'tis folly to be wise. He wasn't saying a plane crash is bliss."
    "Shakespeare doesn't sign our paychecks."
    Last edited by Ken G; 2019-Apr-17 at 08:30 PM.

  3. #123
    Join Date
    Mar 2004
    Posts
    18,310
    Quote Originally Posted by Extravoice View Post
    From what I’ve read, it seems like the Ethiopian pilots followed the “runaway stabilizer” procedures up to the point that they found that they couldn’t turn the trim wheels by hand.
    At that point, they went off-script, but that may have been the right call.
    I look forward to the final report.
    I watched this video from a pilot mentioned earlier in thread:

    https://m.youtube.com/watch?v=9jNbayma9dM

    If I understand correctly, he believes they were so focused on getting pitch under control that they didn’t watch airspeed, which made it impossible to manually adjust pitch, and eventually led to other problems. They turned the system back on, so they could use the motorized adjustment, but that was not the recommended procedure, since MCAS would kick in again and make things worse. Essentially, they had a time window to get things generally under control and didn’t make it in time, due to a problem they shouldn’t have had.

    Edit: I should add that he discusses how a crew in that situation potentially might be able to get the plane under control. Basically, through delegation of tasks, with one focused on flying (including watching airspeed) while other is working the problem. He was careful not to blame the crew, but discussed how human factors come up in emergency situations, and how they are sometimes dealt with.
    Last edited by Van Rijn; 2019-Apr-17 at 11:24 PM.

    "The problem with quotes on the Internet is that it is hard to verify their authenticity." ó Abraham Lincoln

    I say there is an invisible elf in my backyard. How do you prove that I am wrong?

    The Leif Ericson Cruiser

  4. #124
    Join Date
    Mar 2004
    Posts
    18,310
    Quote Originally Posted by profloater View Post
    you would think the only reason to fit two aoa sensors would be to turn the system off if they disagree or if one dies. If reports are correct the system believed the faulty sensor instead of shutting itself down and relying on the pilots. Even having seen enough software errors in my life, it is hard to believe that mode of failure slipped through. The protocols for dealing with one , two or three sensors are old school basics. It's very sobering for a regular flier.
    My impression is that they didnít account for the effects of the repeated ratcheting behavior, and that turned a relatively minor issue into a catastrophic one. From what I have read/seen, MCAS was supposed to make fairly minor adjustments so this plane would fly like earlier models, so new certification and training wouldnít be needed. Introducing a second sensor would add more potential failure modes and complicate the software design for something that, if it failed, (I believe was thought) would only present minor issues for the crew.

    Ultimately, they didnít consider the ramifications of the combination of effects (including changes made during design) and how that could affect a flight crew not prepared to deal with that issue. This was software design failure, but they probably could have resolved it with more time for testing and code fixes.

    "The problem with quotes on the Internet is that it is hard to verify their authenticity." ó Abraham Lincoln

    I say there is an invisible elf in my backyard. How do you prove that I am wrong?

    The Leif Ericson Cruiser

  5. #125
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,439
    i agree, I have been involved in "what if" sessions and it is not a random walk, you go though a list of failures and think about consequences. then you decide if the consequences merit redesign or back up spending. In a plane the consequences are predictably bad during take off in particular, for any event like bird strike, loss of engine, and so on. (Sully comes to mind) The reset of zero (if I heard correctly) in a "switch off, switch on" thought experiment should have been thought through to redesign in my opinion because it removed any chance of manual correction. The consequence of rotation from the repositioned engine thrust should have been a major case for "what if" sessions.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  6. #126
    Join Date
    Oct 2005
    Posts
    26,546
    Quote Originally Posted by Van Rijn View Post
    My impression is that they didn’t account for the effects of the repeated ratcheting behavior, and that turned a relatively minor issue into a catastrophic one. From what I have read/seen, MCAS was supposed to make fairly minor adjustments so this plane would fly like earlier models, so new certification and training wouldn’t be needed.
    That's actually not true. The MCAS was designed to keep the plane from stalling, but that certainly does not in any way imply "minor adjustments" to the stabilizer in any situation where the MCAS is reading data from the AoA that tells it the stabilizer is way out of proper trim (which is exactly what happened), because it will always completely believe that data despite all else, that's how it's programmed. And that's the problem right there, it gets all its information from a single gizmo that can be bent by a bird.

    In the Ethiopian accident, the MCAS apparently thought the stabilizer was way up, and in imminent danger of stalling the plane, even though a host of other information that MCAS does not use at all (like the other AoA, and the airspeed, and the actions of the pilots, and who knows what else) indicated that the opposite was true. But the point is, if the MCAS is radically wrong in its impression of the situation, then the adjustments it makes are not minor, they are quite radical. They're not exactly sudden, but they alter the stabilizer angle downward by some 2 degrees (a huge amount) over some minutes, then it waits some more minutes, and does it again if it is still getting wonky data.

    That is, very simply, always going to crash an airplane in takeoff. This is the crucial thing to understand, the MCAS, in concert with a bad AoA sensor, is programmed to crash a plane, it's just that simple. The designers of the system would have known this, it's really pretty obvious in the hindsight of this accident. Anyone who can see that this will be the outcome of a bad AoA sensor would know that the pilots need to be told how to turn the MCAS off, and they were told how to do that, and they did it. They just did it too late, because the airspeed was up. But I'm not sure it's fair to say they didn't keep an eye on the airspeed, I think it's fairer to say they didn't want to throttle down because they were at low altitude. It's almost like no one in their MCAS scenarios had considered an MCAS failure at low altitude, that really changes everything and is the central cause of this crash.

    It would sure let Boeing off the hook to blame the pilots here, but I don't think it fits the facts-- though it's possible that other pilots might think they would have fared better in that same situation. The sense I got from the pilot in the video linked to by schlaugh is that what the pilots did was not grossly negligent, they were simply in a very tough situation and were understandably slow to realize that the MCAS is programmed to crash a plane under those circumstances, until they were literally being crashed by it. They must have felt pretty desperate to turn the MCAS back on after it had put them through such a panic before that, but they clearly did not expect the reset to zero, or they would have known it was sheer suicide to turn it back on.
    Introducing a second sensor would add more potential failure modes and complicate the software design for something that, if it failed, (I believe was thought) would only present minor issues for the crew.
    There's no way they could have thought that, the MCAS will always crash a plane in that situation unless it is turned off and kept off.
    Ultimately, they didn’t consider the ramifications of the combination of effects (including changes made during design) and how that could affect a flight crew not prepared to deal with that issue. This was software design failure, but they probably could have resolved it with more time for testing and code fixes.
    Well we can certainly agree there needed to be code fixes. The obvious one is to take input from both AoA sensors, and either use whichever one is consistent with the airspeed indicator (which they didn't think to use at all), or at the very least let the pilots switch to whichever one they trusted. The evidence is quite clear that this is an appalling design failure, because it's not a subtle principle that individual gizmos can break, and they can break at low altitude where the pilots need to be able to act on the full range of available information.
    Last edited by Ken G; 2019-Apr-18 at 01:42 AM.

  7. #127
    Join Date
    Mar 2004
    Posts
    18,310
    Quote Originally Posted by Ken G View Post
    That's actually not true. The MCAS was designed to keep the plane from stalling
    I primarily based my comments based on another video by the previously mentioned 737 pilot. Here's a link to a full video about his comments on a software fix for MCAS (about 13 minutes long in total):

    https://www.youtube.com/watch?v=zGM0V7zEKEQ

    I found it very interesting, and I'd recommend watching the entire video (along with his others on the 737 Max), but here's a link to the point in the video where he discusses the primary purpose of MCAS:

    https://www.youtube.com/watch?v=zGM0...youtu.be&t=149

    I'll try to transcribe (with some paraphrasing) a few points, but fair warning, it's easy to make mistakes, so please watch the video:

    Quote Originally Posted by Paraphrasing part of the above video
    MCAS was not meant as a stall prevention system in itself. It was primarily meant so that the 737 Max has a single type rating, the same rating as other 737s. There are some circumstances when the 737 Max wouldn't have the same "feel" as earlier models, and MCAS was meant to replicate that "feel." The new plane is an inherently stable design, like the previous models, but the nose can get a bit light. With the MCAS system turned off, seasoned 737 pilots could barely tell the difference from previous 737s. The significance of the single type rating was to save money, so pilots wouldn't need to be specifically trained for the flight changes there would be without MCAS. However, the original MCAS design proved to be much too powerful. In simulator testing, the MCAS system could get the plane in an unrecoverable situation within 40 seconds.
    So if I understand it, MCAS was supposed to have fairly minor effects, not meant to fix an unstable design that could easily stall, but there were design mistakes that made it do things that were not intended.

    Quote Originally Posted by Ken G View Post
    In the Ethiopian accident, the MCAS apparently thought the stabilizer was way up, and in imminent danger of stalling the plane, even though a host of other information that MCAS does not use at all (like the other AoA, and the airspeed, and the actions of the pilots, and who knows what else) indicated that the opposite was true. But the point is, if the MCAS is radically wrong in its impression of the situation, then the adjustments it makes are not minor, they are quite radical. They're not exactly sudden, but they alter the stabilizer angle downward by some 2 degrees (a huge amount) over some minutes, then it waits some more minutes, and does it again if it is still getting wonky data.
    Yes, that's the repeated ratcheting behavior I was referring to. It would keep ratcheting the stabilizer angle. My impression is that is what could turn a fairly minor issue for the crew into a catastrophic one, and I think that is the key issue that wasn't accounted for in software design. My impression (again based on what I've read in articles and seen on videos) is that without that design mistake, MCAS, even with reading incorrect data from a sensor, wouldn't cause extreme problems that could put the plane quickly in an unrecoverable situation. I strongly suspect other decisions came from that: If it wouldn't have such dire effects, it might not have been considered that significant issue if there was a sensor failure that would affect MCAS, and other failure modes with dual sensors and additional software complexity might have looked more risky.

    Quote Originally Posted by Ken G View Post
    That is, very simply, always going to crash an airplane in takeoff. This is the crucial thing to understand, the MCAS, in concert with a bad AoA sensor, is programmed to crash a plane, it's just that simple. The designers of the system would have known this, it's really pretty obvious in the hindsight of this accident.
    I disagree. I think hindsight bias is involved here. If the designers had understood MCAS would crash a plane, I strongly doubt they would have allowed it to go forward without modification.

    I expect this was more of an issue of a system rushed out the door without sufficient testing or consideration of "What If" scenarios (as mentioned by profloater). That's a nasty thing about testing engineering in real life - the combinations can kill you.

    Quote Originally Posted by Ken G View Post
    It would sure let Boeing off the hook to blame the pilots here, but I don't think it fits the facts-- though it's possible that other pilots might think they would have fared better in that same situation. The sense I got from the pilot in the video linked to by schlaugh is that what the pilots did was not grossly negligent, they were simply in a very tough situation and were understandably slow to realize that the MCAS is programmed to crash a plane under those circumstances, until they were literally being crashed by it.
    It may be possible a crew could have recovered the plane, but I'm quite sure that wouldn't let Boeing off the hook since the problem shouldn't have occurred in the first place. At most, I suspect it might be used as another example in lessons for crew in dealing with extreme conditions.

    "The problem with quotes on the Internet is that it is hard to verify their authenticity." ó Abraham Lincoln

    I say there is an invisible elf in my backyard. How do you prove that I am wrong?

    The Leif Ericson Cruiser

  8. #128
    Join Date
    Oct 2005
    Posts
    26,546
    Quote Originally Posted by Van Rijn View Post
    So if I understand it, MCAS was supposed to have fairly minor effects, not meant to fix an unstable design that could easily stall, but there were design mistakes that made it do things that were not intended.
    Yes, I can believe it was supposed to be something different from what it is, it wasn't supposed to crash airplanes. It sounded like you were saying the MCAS only makes minor adjustments and the pilots escalated it into a problem, but I think I misread your intention there.
    I strongly suspect other decisions came from that: If it wouldn't have such dire effects, it might not have been considered that significant issue if there was a sensor failure that would affect MCAS, and other failure modes with dual sensors and additional software complexity might have looked more risky.
    I agree the problem was underestimated. The question is whether that was because the designers were not very good at anticipating the failure modes of their designs, which would be a strange failing for people in that business, or they were under pressure to understate the potential dangers. I don't know which was the case, and the poor communication from the designers to the pilots could have come from either source.

    I disagree. I think hindsight bias is involved here. If the designers had understood MCAS would crash a plane, I strongly doubt they would have allowed it to go forward without modification.
    The MCAS is designed to overrule pilots and adjust the stabilizer quite aggressively. It basically says the pilots cannot be trusted to keep the nose down because they are used to a different plane and it costs too much to train them on the new plane, so it needs to intercede on their behalf. But it gets input from a single sensor, and if that input is completely wrong, it will crash a plane if there is no leeway to lose altitude. I don't see how the designers could possibly not have known that, there's no need for any hindsight. You simply ask, what happens if the AoA sensor breaks on takeoff, and the MCAS keeps trying to bring the nose down while the pilots are trying to gain altitude? Seems pretty darn inevitable to me, they will have to turn the MCAS off or die. So that's why they told the pilots they might have to turn it off, but the design is such that you can't do that without going to manual control.

    On that latter issue, did you see the part in the schlaugh video where the pilot describes a common training exercise where the instructor distracts the pilot and puts the nose down and the plane picks up speed? Then the instructor turns off the power to the stabilizer, and tells the pilot to salvage the situation. The pilot finds they cannot manually raise the stabilizer to recover proper trim, so they have to think to lower the nose even farther to take pressure off the stabilizer and make it easier to raise manually. Of course you will lose altitude during this maneuvre, so it's not something the instructor would do during takeoff. Since it's a standard training exercise, it can't be too bizarre of a design issue-- manual stabilizer control during takeoff (after high speeds are acquired) must be a well known danger issue.

    I expect this was more of an issue of a system rushed out the door without sufficient testing or consideration of "What If" scenarios (as mentioned by profloater). That's a nasty thing about testing engineering in real life - the combinations can kill you.
    I certainly agree that the appalling design flaw could have arisen from improper testing or considering "what if" scenarios. My point is that not having sufficient testing and "what if" thinking is the very definition of an appalling design flaw.


    It may be possible a crew could have recovered the plane, but I'm quite sure that wouldn't let Boeing off the hook since the problem shouldn't have occurred in the first place. At most, I suspect it might be used as another example in lessons for crew in dealing with extreme conditions.
    The question now is what are they going to have to do to get the MAX flying again. Let's hope they either get better designers this time, or relieve the pressure on the old ones to come up with some kind of slapdash solution. Perhaps Boeing has learned a painful lesson that if you pay too much attention to short-term profits, it can bite you in the long run. It's too late for those passengers to benefit from the lesson.
    Last edited by Ken G; 2019-Apr-18 at 02:42 AM.

  9. #129
    Join Date
    Jun 2005
    Posts
    13,536
    Quote Originally Posted by Ken G View Post
    The question now is what are they going to have to do to get the MAX flying again. Let's hope they either get better designers this time, or relieve the pressure on the old ones to come up with some kind of slapdash solution. Perhaps Boeing has learned a painful lesson that if you pay too much attention to short-term profits, it can bite you in the long run. It's too late for those passengers to benefit from the lesson.
    Only semi-non-seriously. Maybe they should redesign it, with a smaller engine... They'll lose some fuel efficiency apparently, but that's a pretty good trade-off for not crashing.
    As above, so below

  10. #130
    Glom's Avatar
    Glom is offline Insert awesome title here
    Join Date
    Sep 2002
    Posts
    11,275
    The thing is that as soon as you get into the business of giving the flight crew the capability to make adjustments to the MCAS, you are introducing differences from a flight crew perspective, which means loss of commonality. Boeing had to bury MCAS under the hood to avoid this.

  11. #131
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,439
    I am grateful for Van Rijn's comments, it makes a lot of sense to try to maintain the feel of the previous 737 and with the intention of a small bias to the pitch trim, and the incentive of selling with no pilot retraining. The ratchet effect is a major design error in that case as is the provision for action if the two AoA sensors fail to agree with each other. It is not hindsight to recognise that in any system, even systems much simpler than an aircraft, changes always bring unintended consequences and in my experience this is especially true of systems where hardware and software are involved together. I will admit that sometimes the customer is the test house because there is no lab substitute for field testing.

    During my time in the industry we suffered the terrible super stall unintended consequence of high tail aircraft and I recall reading the report of the test pilots who really did try everything during a long, doomed fall from high altitude in a superstalled plane. But those design flaws like the metal fatigue failures that bought down Comets before my time, were new problems that could not be anticipated whereas these two crashes seem at this stage to come from haste and not learning well established lessons in change management.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  12. #132
    Join Date
    Feb 2005
    Posts
    11,354
    Quote Originally Posted by Trebuchet View Post
    Yes, as I understand it that is the reason for MCAS.
    I wonder if the longer fuselage on the 737 Max 9 might allow that to go back into service a bit sooner--counterbalancing the larger engine placement

    Quote Originally Posted by Van Rijn View Post
    Basically, through delegation of tasks, with one focused on flying (including watching airspeed) while other is working the problem. He was careful not to blame the crew, but discussed how human factors come up in emergency situations, and how they are sometimes dealt with.
    An idea:
    The pilot gets the glass cockpit--but the co-pilot has all analog gauges from the 1970s--and we have flight engineers back with a combination of digital and analog systems. Expensive--but worth it.
    Last edited by publiusr; 2019-Apr-23 at 07:12 PM.

  13. #133
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,439
    I feel we are seeing a too slow reorganisation in the wake of these disasters and this is allowing doubt in the safety of 737 family and the doubts will spread in the market. If passengers vote with their feet then the airlines will do that too. It seems to me there should have been further lessons in PR crisis management that have not been applied here. It's not yet too late for some "transparency" but with a share holder meeting imminent, I do not see it happening.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  14. #134
    Join Date
    Oct 2005
    Posts
    26,546
    Now this: (https://www.cnn.com/2019/04/26/polit...rts/index.html) "The day after Ethiopia's minister of transportation released a preliminary crash report on Ethiopian Airlines flight 302, four Boeing employees called an Federal Aviation Administration whistleblower hotline that allows employees and the public to report aviation safety issues. A source familiar with the matter says the hotline submissions involve current and former Boeing employees describing issues related to the angle of attack sensor -- a vane that measures the plane's angle in the air -- and the anti-stall system called MCAS, which is unique to Boeing's newest plane."

    To me this certainly paints the picture of disgruntled designers who felt they were not being heeded by management and concluded they needed to air their concerns externally. It appears there has been a huge communication problem at Boeing, where pilots are kept in the dark about how the systems in their own plane work and are encouraged to fly them with minimal retraining, while design engineers with safety concerns end up concluding that whistleblowing to the FAA is the only way to get their point across to the management. Given all this, I would say a shareholder's meeting is the perfect time for these issues to be addressed-- someone at Boeing needs to give a clear signal that a significant course correction is underway. I would have to say there is significant evidence that Boeing, in its quest to increase profits, has left safety well behind, and that is not going to go down well in the long run, and with shareholders with long-term earnings in mind.

    The only question about the whistleblowers is, why would they wait for it to happen twice? I suspect they probably wouldn't-- they only waited for it to happen twice to air their concerns externally.
    Last edited by Ken G; 2019-Apr-27 at 02:41 PM.

  15. #135
    Join Date
    May 2007
    Location
    Earth
    Posts
    10,240
    What MCAS was supposed to do was to keep the slope of pitch angle (measured from the trimmed condition) vs stick force constant (or at least the same as it was on earlier 737 variants), so a new type endorsement wasn't required. Had Boeing installed a stick shaker* (to warn pilots of impending stall) or a stick pusher (to keep pilots from stalling the aircraft), they would have increased the training burden on the airlines and possibly required pilots to get a separate endorsement for the new model, which would have required a change in marketing, as it could no longer be sold as an aircraft that would require no new training for pilots. What happened with the 737max was that the larger, more forward engine nacelles pushed the center of lift farther forward at high angles of attack, even before the tips stall, and reduced the slope of stick force vs pitch angle; MCAS was there to counter that. Since it was a control system crutch and Boeing argued** that it wasn't something to correct a safety of flight issue, they didn't need it to be as well-vetted as it would have been for a stick pusher or even a stick shaker.

    It's actually pretty common for swept-wing aircraft to have control systems, such as stick pushers or stick shakers, because the stall behavior of swept-wing aircraft, especially swept-wing aircraft with high-aspect ratio (long, thin) wings can be difficult, as swept wings tend to stall from the tips; there were a number of crashes of 727s, DC-9s, and Tridents due to the combination of swept wing and T-tail*** that led to crashes from unrecoverable high-angle of attack condition called "deep stall." See, for example, https://www.pprune.org/archive/index.php/t-334159.html, https://www.defensedaily.com/a-tale-...uncategorized/, and https://aviation.stackexchange.com/q...ecover-from-it.

    There is a lot of literature about the high-angle behavior of swept wings, e.g., Harper & Maki A REVIEW OF THE STALL CHARACTERISTICS OF SWEPT WINGS, https://ntrs.nasa.gov/archive/nasa/c...9640014908.pdf, Chambers & Gfrafton, Aerodynamic Characteristics of Airplanes at High Angles of Attack https://ntrs.nasa.gov/archive/nasa/c...9780005068.pdf, and https://ntrs.nasa.gov/archive/nasa/c...9790079923.pdf, and https://ntrs.nasa.gov/archive/nasa/c...0030063954.pdf. The engineers at Boeing were quite aware of all of it, and also quite aware of the results of their CFD modeling.





    -------------

    * See this article: https://www.flyingmag.com/how-it-wor...-shaker-pusher Note that the article does have some rather lazy editorializing; neither stick pushers not stick shakers are "lazy AoA indicators"; many transport aircraft, especially swept-wing aircraft with T-tails will go into deep stalls, which are not recoverable.

    ** Rather obviously, they argued successfully and wrongly.

    *** I do know that 737s do not have T-tails; deep-stall behavior is one of the reasons they do not. On the other hand, I suspect that just about every swept-wing aircraft with rear-mounted engines has either a stick shaker or a stick pusher.
    Information about American English usage here and here. Floating point issues? Please read this before posting.

    How do things fly? This explains it all.

    Actually they can't: "Heavier-than-air flying machines are impossible." - Lord Kelvin, president, Royal Society, 1895.



  16. #136
    Join Date
    Oct 2005
    Posts
    26,546
    And note that arguing that all this is well understood only makes the situation worse, because it takes off the table the oft-seen "hindsight is 20-20" counterargument. What seems pretty clear is that if the one AoA sensor used by MCAS fails during takeoff, and the pilots don't immediately realize the source of the problem (among all the other things that could be going wrong), then they are in a very serious predicament that can be expected to risk catastrophic crash.

    Worse, after this situation did indeed lead to one crash (Lion Air), there were efforts to downplay the severity of the problem, and then when it led to a second crash (Ethiopia) there was still an effort to keep the planes in the air. That situation is made only more appalling by arguments that explain how well understood are the deep stall characteristics of a swept-wing aircrafta, because it points to the need for fail-safes in the stall mitigation software. At least with stick pushers and stick shakers, the plane is informing the pilots rather than fighting them. Is there any way to interpret this situation that does not look like Boeing trying to cut corners on safety to increase profit?

    Even the FAA seems complicit in the effort. Their words ring pretty false when they insist that safety is their main concern while certifying the MCAS use of a single AoA sensor, and won't even admit to having made a pretty obvious mistake. Here is the quote of the FAA: "The single angle of attack sensor was considered in relation to a variety of other factors, specifically well-known pilot procedures that would mitigate the effects of a failure. MCAS design, certification tests, and cockpit procedures were evaluated using a standard industry approach to failure analysis." In light of the evidence, I'd say that statement is approaching pretty close to wrong-is-right style doublespeak. They won't even say they were wrong to certify it, but there's no new information than what they already had, yet those planes are all on the ground right now. How can the FAA logic hold water under those circumstances?
    Last edited by Ken G; 2019-Apr-27 at 11:00 PM.

  17. #137
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,548
    A YouTube clip of a Nova PBS presentation is yet another example of popular media getting the physics wrong. They asserted that the larger engines mounted farther forward caused an increased pitch-up tendency by moving the center of gravity forward. It's physics 101 to see that moving the center of gravity forward should make it pitch down a bit if all else is equal. As I understand it, the aerodynamic properties of the larger nacelles farther forward are the reason for the pitch-up tendency at high angles of attack, and having the effective thrust center a bit farther forward contributes to it. The melodramatic style of the Nova piece further alienated me.

  18. #138
    Join Date
    May 2007
    Location
    Earth
    Posts
    10,240
    The FAA relies, quite heavily, on the honesty and good faith of the manufacturers. This usually works; engineers and management normally have enough imagination to realize what is at stake. Sometimes this fails, like here with Boeing, and the failure to properly deal with MCAS, flawed maintenance procedures with American Airlines Flight 191, and omitted bird ingestion tests with the CF6.
    Information about American English usage here and here. Floating point issues? Please read this before posting.

    How do things fly? This explains it all.

    Actually they can't: "Heavier-than-air flying machines are impossible." - Lord Kelvin, president, Royal Society, 1895.



  19. #139
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,439
    I have looked through the many crashes listed for the 737 (in WP) over nearly fifty years and none it seems were due to stall at take off even though in many ways this is a difficult phase of any flight with max engine power and steep angle of attack. This modification seems therefore a home goal against Boeing. Super stall or deep stall is not the issue here and all pilots know the issues surrounding take off. If, as it now seems clear, the modification was just to maintain the stick feel, it was not thought through as it should have been, for failure of one sensor , no pilot warning (an option) and a ratchet effect of the software reset of trim angle. Tragic as the two crashes are, it is the aftermath of cover up that will damage Boeing the most because among many historical disasters, these two should have been avoided by plain thinking about the changes and the second, Ethiopian crash especially so.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  20. #140
    Glom's Avatar
    Glom is offline Insert awesome title here
    Join Date
    Sep 2002
    Posts
    11,275
    Quote Originally Posted by swampyankee View Post
    The FAA relies, quite heavily, on the honesty and good faith of the manufacturers. This usually works; engineers and management normally have enough imagination to realize what is at stake. Sometimes this fails, like here with Boeing, and the failure to properly deal with MCAS, flawed maintenance procedures with American Airlines Flight 191, and omitted bird ingestion tests with the CF6.
    But probably the most important reason for independent evaluation is a fresh set of Androsses to pick up what was missed by mistake rather than deliberate lack of safety focus.

  21. #141
    Join Date
    Oct 2005
    Posts
    26,546
    Actually, it would seem to me the most important reason for independent evaluation is very much the deliberate lack of safety focus. This is true of all kinds of industries, it is not always obvious to a corporation leadership that safety needs to be a prime focus when it reduces profit, especially when accidents can be blamed on someone else. That's why pilots often have to push back against being blamed for crashes and near-crashes, even in situations where the pilots' actions were heroic (such as the celebrated case of Sully Sullenberger, or in the case of the Gottrora crash which also had no deaths-- had those been complete plane losses, one imagines the pilots would have been easy targets). Interestingly, the Gottrora crash was also a situation where plane software was fighting the pilots in ways the pilots did not understand, preventing the pilots from avoiding a crash. Apparently this is a commonly accepted design option-- prevent the pilot from doing something wrong, even when the pilot is acting on complete information that the software is not using, and don't necessarily make sure the pilots are even aware that they will be fought against and how to disable that wonderful feature. I get that pilot error is an important problem and is mitigated in many ways, but it seems to me a key principle when overruling pilots is to be sure that the system doing the overruling is using the same crucial information that the pilots have access to. When the pilots know much more than the software, it seems folly to act on limited information even as the pilot is desperately trying to act on complete information. Nightmarish, actually-- straight out of something HAL might do.
    Last edited by Ken G; 2019-Apr-29 at 07:46 PM.

  22. #142
    Join Date
    Feb 2005
    Posts
    11,354
    Quote Originally Posted by Trebuchet View Post
    In fact, the main reason for the existence of MCAS is the that the bigger engines are further ahead of the wings and would increase the tendency to pitch up with added thrust.
    Here is a quote from folks in the space industry that you would think would also apply to aircraft designers:

    "...rocket aerodynamicists are sticklers for telling their chief engineers that they are allowed to modify anything they want inside the rocket, but they canít change a thing with the rocketís outer mold line, speed, or weight distribution or else all of their carefully calculated models instantly become worthless. These handcuffs make it hard to introduce something like a new SRB, but in exchange, the aerodynamicists promise to keep the rockets from falling into the ocean. That last part tends to keep the peace."

    --from
    http://www.thespacereview.com/article/3658/1


    One last thought. I can't help but wonder about something....

    Had this not been yet another in a long series of blasted twin-jets---had it been a tri-jet, would the greater weight of the center, tail engine been enough to push the CG back, lessening the need for the MCAS to begin with?

  23. #143
    Join Date
    Jun 2004
    Location
    The Great NorthWet
    Posts
    14,289
    Boeing, on the other hand, allowed the interior decorators to dictate exterior lines of the 787. And I'm not making that up.
    Cum catapultae proscriptae erunt tum soli proscript catapultas habebunt.

  24. #144
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,439
    Quote Originally Posted by publiusr View Post
    ...
    ...
    Had this not been yet another in a long series of blasted twin-jets---had it been a tri-jet, would the greater weight of the center, tail engine been enough to push the CG back, lessening the need for the MCAS to begin with?
    It’s more subtle, the engines being forward would tip the nose down and a rear engine would tip it up. But the engine airflow presumably affects the wing lift as well as the thrust line being under the c of g so that high thrust lifts the nose. If this effect is more at high angle of attack it must be aerodynamic such as by blowing away the downdraught off the wings. Speculation, but it is not the extra forward weight.
    sicut vis videre esto
    When we realize that patterns don't exist in the universe, they are a template that we hold to the universe to make sense of it, it all makes a lot more sense.
    Originally Posted by Ken G

  25. #145
    Join Date
    Feb 2005
    Posts
    11,354
    Quote Originally Posted by Trebuchet View Post
    Boeing, on the other hand, allowed the interior decorators to dictate exterior lines of the 787. And I'm not making that up.
    Ack!

    A question. Had Sully been flying a tri-jet, does he still have to land in the Hudson?
    Last edited by publiusr; 2019-Apr-30 at 09:27 PM.

  26. #146
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,548
    Quote Originally Posted by publiusr View Post
    Ack!

    A question. Had Sully been flying a tri-jet, does he still have to land in the Hudson?
    If the ingested geese took out all three engines at the same altitude, I would say yes. If he still had one engine going, he might have been able to get to Teterboro.

  27. #147
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,548
    Quote Originally Posted by publiusr View Post
    Here is a quote from folks in the space industry that you would think would also apply to aircraft designers:

    "...rocket aerodynamicists are sticklers for telling their chief engineers that they are allowed to modify anything they want inside the rocket, but they can’t change a thing with the rocket’s outer mold line, speed, or weight distribution or else all of their carefully calculated models instantly become worthless. These handcuffs make it hard to introduce something like a new SRB, but in exchange, the aerodynamicists promise to keep the rockets from falling into the ocean. That last part tends to keep the peace."

    --from
    http://www.thespacereview.com/article/3658/1


    One last thought. I can't help but wonder about something....

    Had this not been yet another in a long series of blasted twin-jets---had it been a tri-jet, would the greater weight of the center, tail engine been enough to push the CG back, lessening the need for the MCAS to begin with?
    My educated guess is that changing it to a tri-jet would have made it a different enough airplane that the need for new type certification for the crew would have been ruled out.

  28. #148
    Join Date
    Jun 2004
    Location
    The Great NorthWet
    Posts
    14,289
    Informative, if annoying because of the editing, video by Blancolirio.

    I fear Boeing's CEO may be getting bad advice from attorneys and bean-counters.
    Cum catapultae proscriptae erunt tum soli proscript catapultas habebunt.

  29. #149
    Join Date
    Feb 2010
    Posts
    728
    Iím late to this thread, so apologies if this has already been covered. Has Boeing explained why MCAS, as designed, was OK with input from one sensor only?

  30. #150
    Join Date
    May 2007
    Location
    Earth
    Posts
    10,240
    Quote Originally Posted by publiusr View Post
    Here is a quote from folks in the space industry that you would think would also apply to aircraft designers:

    "...rocket aerodynamicists are sticklers for telling their chief engineers that they are allowed to modify anything they want inside the rocket, but they can’t change a thing with the rocket’s outer mold line, speed, or weight distribution or else all of their carefully calculated models instantly become worthless. These handcuffs make it hard to introduce something like a new SRB, but in exchange, the aerodynamicists promise to keep the rockets from falling into the ocean. That last part tends to keep the peace."

    --from
    http://www.thespacereview.com/article/3658/1


    One last thought. I can't help but wonder about something....

    Had this not been yet another in a long series of blasted twin-jets---had it been a tri-jet, would the greater weight of the center, tail engine been enough to push the CG back, lessening the need for the MCAS to begin with?
    Probably not; the c/g would be in the same place, as the forward fuselage would be relatively longer.
    Information about American English usage here and here. Floating point issues? Please read this before posting.

    How do things fly? This explains it all.

    Actually they can't: "Heavier-than-air flying machines are impossible." - Lord Kelvin, president, Royal Society, 1895.



Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •