Page 10 of 11 FirstFirst ... 891011 LastLast
Results 271 to 300 of 313

Thread: Lion Air Flight 610 Crash

  1. #271
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,498
    As mentioned earlier, if two sensors are used in any system you cannot employ voting so you disable the system if they disagree or both give nonsensical outputs, and issue a warning. this is really basic systems design. Even without a warning the MCAS could be programmed to stay off if the sensors disagree but a warning seems a very good idea to tell the pilots to take it easy in the climb..it is really hard to explain as a designer, why you would let a control system operate when the AoA sensors disagree. It’s not a tiny detail you could overlook. And those sensors were known in the beginning to be accident prone.
    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. #272
    Join Date
    Oct 2005
    Posts
    26,574
    Quote Originally Posted by Jean Tate View Post
    For example, there was considerable "scope creep" in MCAS; initially a failure due to a sensor malfunction was not considered because "sensor input" was not part of the original specs (or something like that). When the scope was expanded, the initial design team was not informed.
    Yuck, I think it is safe to say this is clearly no way to build an airplane. Surely people who build airplanes ought to know this!
    I'm now a long way from software development, testing, and project management, but the sorts of things Boeing did surely violated a great many basic cannons of "best practice". A separate failure is the FAA's (and in hindsight other regulatory/certification agencies): given the known havoc "bad" software systems can have, why were they so relaxed about accepting what Boeing told them about MCAS?
    Seems like a bit too cosy of a relationship, to say the least. Those who know the way these relationships are being set up these days cannot be terribly surprised, but let us hope this incident pulls the wool off the necessary eyes.

  3. #273
    Join Date
    Oct 2005
    Posts
    26,574
    Quote Originally Posted by profloater View Post
    it is really hard to explain as a designer, why you would let a control system operate when the AoA sensors disagree.
    It's no problem, you just let the control system guess which one is right. After all, you still have a 50-50 chance of survival, so you can't blame the system if it's your luck that's bad.

  4. #274
    Join Date
    May 2007
    Location
    Earth
    Posts
    10,246
    Quote Originally Posted by Ken G View Post
    Yuck, I think it is safe to say this is clearly no way to build an airplane. Surely people who build airplanes ought to know this!
    Seems like a bit too cosy of a relationship, to say the least. Those who know the way these relationships are being set up these days cannot be terribly surprised, but let us hope this incident pulls the wool off the necessary eyes.
    As to the first, yep: they should. So should their managers and executives. For the second, one hopes, but one is not optimistic. The penalties for safety failures of this kind aren’t paid by the people bearing the responsibility. At most, some engineering line group leader or staff engineer will get fired.
    Last edited by swampyankee; 2019-Jun-20 at 01:45 PM.
    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.



  5. #275
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,498
    Quote Originally Posted by Ken G View Post
    It's no problem, you just let the control system guess which one is right. After all, you still have a 50-50 chance of survival, so you can't blame the system if it's your luck that's bad.
    I wonder if the FAA have that sense of irony? As a friend said to me when something went wrong like that "If I didn't have bad luck, I wouldn't have any luck at all." I wonder if the new software includes guessing, or requires more hardware (like a "I'm guessing" warning light)
    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. #276
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,498
    Quote Originally Posted by swampyankee View Post
    As to the first, yep: they should. So should their managers and executives. For the second, one hopes, but one is not optimistic. The penalties for safety failures of this aren’t paid by the people bearing the responsibility. At most, some engineering line group leader or staff engineer will get fired.
    My bet is there are prior memos from engineering about exactly this issue.
    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

  7. #277
    Join Date
    May 2007
    Location
    Earth
    Posts
    10,246
    Quote Originally Posted by profloater View Post
    My bet is there are prior memos from engineering about exactly this issue.
    ...and prior memos from management saying "it's not a problem" or "STFU."
    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.



  8. #278
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,498
    Incidentally, why do we talk about moving the stabiliser, surely we are referring to the elevators and elevator tabs?
    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

  9. #279
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,744
    Quote Originally Posted by profloater View Post
    Incidentally, why do we talk about moving the stabiliser, surely we are referring to the elevators and elevator tabs?
    Two different animals, two different but similar functions.

    In the 737 the stabilizer is trimmed during flight, either by the trim adjustment switches by the pilots, or manually with the trim wheels, or by the autopilot - or by MCAS.

    The elevators move when the pilot (or autopilot) pulls or pushes the yoke to control the aircraft pitch, ideally in gentle increments.

    The elevators are much smaller than the stabilizer so in a tug of war for pitch, the stabilizer wins. IIRC the crash investigators found the elevators in a nose up position and the stabilizer in a nose down position.

    Try this for more details:
    https://www.grc.nasa.gov/www/k-12/airplane/elv.html

    And from that site:

    The stabilizer is a fixed wing section whose job is to provide stability for the aircraft, to keep it flying straight. The horizontal stabilizer prevents up-and-down, or pitching, motion of the aircraft nose. The elevator is the small moving section at the rear of the stabilizer that is attached to the fixed sections by hinges. Because the elevator moves, it varies the amount of force generated by the tail surface and is used to generate and control the pitching motion of the aircraft. There is an elevator attached to each side of the fuselage. The elevators work in pairs; when the right elevator goes up, the left elevator also goes up. This slide shows what happens when the pilot deflects the elevator.

  10. #280
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,498
    Quote Originally Posted by schlaugh View Post
    Two different animals, two different but similar functions.

    In the 737 the stabilizer is trimmed during flight, either by the trim adjustment switches by the pilots, or manually with the trim wheels, or by the autopilot - or by MCAS.

    The elevators move when the pilot (or autopilot) pulls or pushes the yoke to control the aircraft pitch, ideally in gentle increments.

    The elevators are much smaller than the stabilizer so in a tug of war for pitch, the stabilizer wins. IIRC the crash investigators found the elevators in a nose up position and the stabilizer in a nose down position.

    Try this for more details:
    https://www.grc.nasa.gov/www/k-12/airplane/elv.html

    And from that site:
    Ok thanks for that, all clear.
    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

  11. #281
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,600
    I saw what looked like a 737 Max on final approach to Dulles today. It had the distinctive winglets. I am guessing it was a test flight.

  12. #282
    Join Date
    Oct 2005
    Posts
    26,574
    Was it bobbing up and down like a roller coaster?

  13. #283
    Join Date
    Jun 2004
    Location
    The Great NorthWet
    Posts
    14,342
    I think Aviation Partners has similar winglets for retrofit. Hang, on, let me check...
    ...Yes, they do. Scroll down a tad.
    Cum catapultae proscriptae erunt tum soli proscript catapultas habebunt.

  14. #284
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,600
    Quote Originally Posted by Trebuchet View Post
    I think Aviation Partners has similar winglets for retrofit. Hang, on, let me check...
    ...Yes, they do. Scroll down a tad.
    That means I probably saw an NG model retrofitted with new winglets.

  15. #285
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,600
    Here is a new Seattle Times article. It goes into a lot more technical detail than we have seen before from the news media.
    https://www.seattletimes.com/seattle...st-safeguards/
    The account of the technical details of the handling issues and the unsuccessful attempts to resolve them aerodynamically reinforced my inclination to see an analogy to the modernization of the USS Midway in the late 1960s. In that case it appears that the Navy went overboard (pardon the pun) and gave her a flight deck as large as those on the new Forrestal-class carriers, with only about 3/4 of the standard displacement. This overloaded the ship and made her run too low in the water, with less freeboard than was desirable. Retrofitting the hull with blisters restored the freeboard but gave the ship roll characteristics that precluded flight operations in heavy seas which the supercarriers could handle. With that experience they chose to settle for a lesser project on the sister ship Coral Sea.

  16. #286
    Join Date
    Jun 2005
    Posts
    13,562
    Quote Originally Posted by Hornblower View Post
    In that case it appears that the Navy went overboard (pardon the pun) and gave her a flight deck as large as those on the new Forrestal-class carriers, with only about 3/4 of the standard displacement.
    Sorry, this is sort of a derail, and maybe should be moved to a new thread, but I was just wondering if it is still common to use "she" for ships. I always use "it" myself, but am a bit curious about how prevalent that is, as well as in versus outside the US.
    As above, so below

  17. #287
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,600
    Quote Originally Posted by Hornblower View Post
    Here is a new Seattle Times article. It goes into a lot more technical detail than we have seen before from the news media.
    https://www.seattletimes.com/seattle...st-safeguards/
    The account of the technical details of the handling issues and the unsuccessful attempts to resolve them aerodynamically reinforced my inclination to see an analogy to the modernization of the USS Midway in the late 1960s. In that case it appears that the Navy went overboard (pardon the pun) and gave her a flight deck as large as those on the new Forrestal-class carriers, with only about 3/4 of the standard displacement. This overloaded the ship and made her run too low in the water, with less freeboard than was desirable. Retrofitting the hull with blisters restored the freeboard but gave the ship roll characteristics that precluded flight operations in heavy seas which the supercarriers could handle. With that experience they chose to settle for a lesser project on the sister ship Coral Sea.
    Addendum: I can envision in principle an active roll damper system such has been used successfully to reduce roll in passenger ships. I don't know whether or not it is practical for aircraft carriers. MCAS works fine for augmenting the trim function of an airplane provided it is not misprogrammed as this one was.

  18. #288
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,600
    From May 24:
    Quote Originally Posted by Hornblower View Post
    Last night I picked the brain of a recently retired systems engineer who has worked on software projects of similar magnitude and complexity. I asked him about having the computer doing the cross check I described above. He said yes indeed, but reminded me that it is complex and must not be rushed into production. He said it is entirely possible that there are other bugs lurking there that have not caused trouble so far, but could turn around and bite you if you do not do exhaustive testing after making the revision. If that takes six months or more, so be it. He agreed that it was unfair to blame the pilots for the missed opportunities under the circumstances.
    From June 12:
    Quote Originally Posted by Hornblower
    I just checked some online news, and it appears that I can paraphrase the various officials as saying something like, "This will take as long as it takes to get it right." Now they do not appear to have possibly false hopes of resolving this in the next few weeks.
    From CNN today:
    https://www.cnn.com/2019/06/26/polit...law/index.html
    During simulator testing a faulty microprocessor reportedly caused a similar nosedive despite the software revision. This could be one of those previously hidden bugs my engineer friend cautioned about. I am no expert but I would advocate replacing the flaky devices instead of trying to program around them.

  19. #289
    Join Date
    Mar 2004
    Posts
    18,329
    Quote Originally Posted by Hornblower View Post
    From CNN today:
    https://www.cnn.com/2019/06/26/polit...law/index.html
    During simulator testing a faulty microprocessor reportedly caused a similar nosedive despite the software revision. This could be one of those previously hidden bugs my engineer friend cautioned about. I am no expert but I would advocate replacing the flaky devices instead of trying to program around them.
    It's really hard to say without a better description of the problem. They might be referring to some subtle interaction between software and hardware that can be resolved with a code change. Or it may be about how the flight computer falls back to a redundant system. Maybe the full system is too slow to detect hardware failure, and perhaps that can be resolved in software? I'd also say that previously hidden bugs aren't the only issue, but that new failure modes might have appeared with the software update. The good news is that they are testing it more carefully this time, and both Boeing and the FAA have a huge incentive to get it right this time (I expect Boeing can survive the previous crashes, but I wouldn't give them great chances if they had another crash due to the same issue).

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

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

    The Leif Ericson Cruiser

  20. #290
    Join Date
    Mar 2004
    Posts
    18,329
    Quote Originally Posted by Hornblower View Post
    Here is a new Seattle Times article. It goes into a lot more technical detail than we have seen before from the news media.
    https://www.seattletimes.com/seattle...st-safeguards/
    That is an excellent article, with a lot of detail. A lot of things we suspected here. The initial concept would have been understood to be low risk because of scope, but there definitely was scope change and scope creep, there was management pressure that limited testing, especially in regards to changes, they didn't consider how repeated activation would increase the problem, and they didn't consider how stick shake and other effects would confuse pilots and expected pilots would quickly realize there was a stabilizer runaway if there was a problem.

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

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

    The Leif Ericson Cruiser

  21. #291
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,744
    New update from Juan Brown. Apparently the microprocessor "failure" isn't really a failure and those vaunted aviation experts at CNN have got it wrong. Act shocked. The problem is that the microprocessor in question controls the rate of change when using the electric trim switches and the test pilots don't like that rate when electrically trimming the stabilizer; they want it to be faster. So it's not that the microprocessor is forcing the nose down; the plane is already nose down.

    https://youtu.be/isy9yAU6ajQ

  22. #292
    Join Date
    Feb 2005
    Posts
    11,403

  23. #293
    Join Date
    Mar 2007
    Location
    Falls Church, VA (near Washington, DC)
    Posts
    8,600
    Quote Originally Posted by publiusr View Post
    No, no, a thousand times no. It leads off saying the microprocessor could send the plane into a nosedive, as did the CNN article itself. I will go with Mr. Brown's comments. With the exception of the Seattle Times, I am pretty peeved with the mainstream media on this topic.

  24. #294
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,744
    Quote Originally Posted by publiusr View Post
    Geeze, that's even worse than CNN because the IE writers not only did not understand what the pilots were finding, but now they've decoupled it from the MCAS testing that uncovered the undesirable rate of change. :headbang:

  25. #295
    Join Date
    Feb 2005
    Posts
    11,403
    From something with the word "engineering" in the title no less...

  26. #296
    Join Date
    Mar 2004
    Posts
    18,329
    Quote Originally Posted by schlaugh View Post
    New update from Juan Brown. Apparently the microprocessor "failure" isn't really a failure and those vaunted aviation experts at CNN have got it wrong. Act shocked.
    Yes, I shouldn't have even tried to guess what the real issue was. It was obvious that the explanation was at least lacking important detail.

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

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

    The Leif Ericson Cruiser

  27. #297
    Join Date
    Oct 2005
    Posts
    26,574
    And Sullenberger thinks they should require simulator training for the MAX-- avoiding that was the whole point of the MCAS fiasco. Sully's is a novel idea-- just train the pilots to fly the plane they are actually flying, rather than tricking them into feeling like they are flying a plane they are more used to. Imagine Sully himself trying to fly a plane that is trying to trick him into thinking it's a different plane-- it's really condescending to the pilots, and takes away their ability to make on-the-spot decisions to save emergency situations.

    I would have thought "know thy plane" would have been a kind of mantra for pilots, why would Boeing try to circumvent that? It would be the ultimate irony if they end up having to train MAX pilots anyway-- something they could have done from the start, without loss of life or stock options.

  28. #298
    Join Date
    Apr 2011
    Location
    Norfolk UK and some of me is in Northern France
    Posts
    8,498
    I must be naiive to think that was the reason we have an FAA to insist on testing and to have independent experts especially for any new design. Such an expert might look at those flappy things and ask what they do. “Oh they measure the AoA? What if they get hit by a bird? Did you check that in a simulator?” Which is worse? That never happened or it did happen and was quietly buried?
    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

  29. #299
    Join Date
    Aug 2005
    Location
    NEOTP Atlanta, GA
    Posts
    2,744
    I'm starting to think that Juan Brown may not have it quite right. While the FAA test pilots may indeed want a faster response on the rate of trim change, Boeing stated that the new issue can cause uncommanded movement (as CNN reported). From Boeing's press release:

    CHICAGO, June 26, 2019 – The safety of our airplanes is Boeing’s highest priority. During the FAA’s review of the 737 MAX software update and recent simulator sessions, the Federal Aviation Administration (FAA) identified an additional requirement that it has asked the company to address through the software changes that the company has been developing for the past eight months. The FAA review and process for returning the 737 MAX to passenger service are designed to result in a thorough and comprehensive assessment. Boeing agrees with the FAA's decision and request, and is working on the required software. Addressing this condition will reduce pilot workload by accounting for a potential source of uncommanded stabilizer motion. Boeing will not offer the 737 MAX for certification by the FAA until we have satisfied all requirements for certification of the MAX and its safe return to service.
    That doesn't sound like a rate of change issue, that sounds like a new problem which forces a trim adjustment. Or maybe both?

  30. #300
    Glom's Avatar
    Glom is offline Insert awesome title here
    Join Date
    Sep 2002
    Posts
    11,277
    Quote Originally Posted by Ken G View Post
    And Sullenberger thinks they should require simulator training for the MAX-- avoiding that was the whole point of the MCAS fiasco. Sully's is a novel idea-- just train the pilots to fly the plane they are actually flying, rather than tricking them into feeling like they are flying a plane they are more used to. Imagine Sully himself trying to fly a plane that is trying to trick him into thinking it's a different plane-- it's really condescending to the pilots, and takes away their ability to make on-the-spot decisions to save emergency situations.

    I would have thought "know thy plane" would have been a kind of mantra for pilots, why would Boeing try to circumvent that? It would be the ultimate irony if they end up having to train MAX pilots anyway-- something they could have done from the start, without loss of life or stock options.
    Because Boeing had a panic attack after the launch of the A320NEO and were desperate to promise anything to American Airlines to not lose them to Europe. Since antigravity wasn't going to be credible, they had to settle for promising an aircraft that wouldn't have retraining costs, silly though that may have been.

Posting Permissions

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