Slashdot is powered by your submissions, so send in your scoop

 



Forgot your password?
typodupeerror
×
Transportation Technology

Chinese Carriers, Ethiopian Airlines Halt Use of Boeing 737 MAX 8 Aircraft After Crash (reuters.com) 182

China's aviation regulator today grounded nearly 100 Boeing Co 737 MAX 8 aircraft operated by its airlines, more than a quarter of the global fleet of the jets, after a deadly crash of one of the planes in Ethiopia. From a report: However, a U.S. official said it was unclear what information the Chinese regulator was acting on because the investigation of Sunday's crash, the second involving the latest version of the narrowbody jet, was in the early stages. Speaking on condition of anonymity as the topic is sensitive, the U.S. official said there were no plans to follow suit, as the jet had a stellar safety record in the United States and there was a lack of information on what caused the Ethiopian crash.
This discussion has been archived. No new comments can be posted.

Chinese Carriers, Ethiopian Airlines Halt Use of Boeing 737 MAX 8 Aircraft After Crash

Comments Filter:
  • by Anonymous Coward on Monday March 11, 2019 @09:17AM (#58252720)

    I'm going out in a limb here but maybe they were acting on the information that another one had crashed.

  • by Anonymous Coward

    Normally, I'm pretty happy to fly on any well maintained airplane, but a second crash within 5 months, where the early indications are that the plane crashed itself despite the best efforts of the pilots to prevent that, would make me cautious of flying on a 737 MAX 8.

    How does that quote go: "To lose one parent may be regarded as a misfortune; to lose both looks like carelessness."

    Unless Boeing get this sorted out very fast then folks will reinterpret "MAX 8" to be the maximum amount of time in minutes that

  • by holophrastic ( 221104 ) on Monday March 11, 2019 @09:23AM (#58252758)

    when last I checked, a "lack of information" is a great reason to avoid something dangerous. Actually, it might be the one and only and best reason to avoid anything dangerous -- from bicycles to bungee jumping. Get informed first. And if you thought you were informed, and suddenly you discover that you aren't informed because you can't explain something that happened, well then, avoid again until you become informed again.

    In other words, let someone else run the tests. That's exactly what test-pilots are for.

    • by caino59 ( 313096 )

      No kidding. Saying that it's had a stellar record in the US is somewhat less meaningful when it has only been in commercial use for less than 2 years.

    • by Nidi62 ( 1525137 )

      when last I checked, a "lack of information" is a great reason to avoid something dangerous. Actually, it might be the one and only and best reason to avoid anything dangerous -- from bicycles to bungee jumping. Get informed first. And if you thought you were informed, and suddenly you discover that you aren't informed because you can't explain something that happened, well then, avoid again until you become informed again.

      In other words, let someone else run the tests. That's exactly what test-pilots are for.

      That's my thought as well. 2 high profile crashes of a new airframe within a few months, one of which was directly caused by an undocumented and therefore untrained for "safety feature" and the other with no immediately identifiable external (ie not integral to the aircraft itself) causes (bad weather, explosion, etc), the immediate reaction would be to suspend flights with aircraft of that type. There are at most about 50 737 MAX 8s in operation with US carriers (I could only find total MAXs delivered,

    • by ceoyoyo ( 59147 ) on Monday March 11, 2019 @09:57AM (#58253050)

      There aren't really that many Max 8s flying, and they haven't been doing so for very long, so a couple of crashes seems like a bad sign, particularly since they were very similar (immediately after takeoff). The Max also has a lot of new technology and new aerodynamic design, so it's not just a minor upgrade over previous 737s, which do have a long safety record.

      The 737 Max is unstable at higher angles of attack so Boeing added a bit of software to correct if the angle of attack starts to get dangerous. From the Lion Air crash it sounds like there might be problems with the sensors, causing that system to improperly engage and actually put the plane into a dangerous dive.

      Ethiopian Airlines got burned by the 787 battery fire issues too.

      • by Anonymous Coward

        Automobiles get in about 1 fatal accident per 10,000 vehicles per year.

        The 737 MAX is at about 1 per 100 per year, since the average age of the fleet at about 20/month being produced is about half a year.

        • Not even going to do the math, but you're comparing a time unit (per year) to a non-time unit. (total vehicles that exist)

          What you need to do is normalize it by converting the number of vehicles to hours driven. Then your units are both based on time, and you can make a reasonable comparison.

          The way it is now, you're including grandma's 1 trip to the store per week as if she was driving full time like a taxi, or a commercial airplane.

  • by RockDoctor ( 15477 ) on Monday March 11, 2019 @09:35AM (#58252862) Journal
    UK reporting (this is about start-of day time in USA?) is that Boeing shares are 12.9% down, which is enough to get any company's attention.

    The ground search (well, digging into the ground) has located both FDR (Flight Data Recorder) and CVR (Cockpit Voice Recorder) for the crashed aircraft, which should help greatly in determining what the problem was/ is.

  • by JoeyRox ( 2711699 ) on Monday March 11, 2019 @09:36AM (#58252874)
  • by Anonymous Coward

    Today's nyt quotes an eyewitness as saying the plane was trailing smoke and making strange noises before it went down. It was carrying lots of UN people. It might have nothing to do with the MCAS system and everything to do with the political system.

  • by Jzanu ( 668651 ) on Monday March 11, 2019 @10:03AM (#58253094)
    The Chinese response is the most appropriate, especially since they are not blinded by the fear of damaging profit margins for Boeing. The fact that at least 2 incidents with "unknown" cause involve the same aircraft means its continued use while the cause is unknown is a risk. China recognizes it as an avoidable risk, which it is in reality.
  • Bad Sensor (Score:5, Interesting)

    by Thelasko ( 1196535 ) on Monday March 11, 2019 @10:10AM (#58253150) Journal
    If I recall, the previous crash has been linked to a bad angle of attack sensor. [flyingmag.com] This sensor is only used by a new stall protection feature [seattletimes.com] in the 737 Max. When it fails, the stall protection algorithm thinks the plane is stuck in a nose up orientation, and tries to force the nose down... into the ground.

    There are several things that should happen:

    1. Interim corrective action. Disable stall protection on all 737 Max aircraft.
    2. Quality control investigation into the angle of attack sensor reliability.
    3. Implement diagnostic algorithms into the control strategy to detect failed angle of attack sensors automatically. A failed sensor should disable the stall protection feature automatically, and alert the pilot.
    • Not only those, they need to change how this feature works. The system is simply too dangerous as-is.

      I'm shocked that this idiotic "feature" wasn't disabled on all 737 Max 8s after the first crash. This is a system that will override the pilot's inputs to use the elevator to pitch the aircraft downward at any time, it's normally on. There's no indication to let the pilot know that this system is doing this. If one sensor glitches out it will cause a crash rather quickly unless the pilot knows how to recogni

      • Why can't the computer make a determination that a sensor is faulty? That is to say, if the computer thinks the nose is pointed down because one sensor says this, but altitude is decreasing rapidly and airspeed is high and the pilot is pulling back on the stick frantically, can't the computer recognize that the nose might actually not be pointed down and ignore that sensor? Allowing a single senors to take complete control over the plane from the pilot seems like really bad programming.
        • It is bad programming and unfortunately, having automatic aircraft control systems use a single sensor when many are available seems to be the norm. For example in the Air France 447 crash, the stall warning system only used the notoriously troublesome pitot tube sensors and didn't look to GPS inputs for a second opinion. Neither did the pilots...

          But I still don't think it's a good idea to let a computer silently take control of an airplane from the pilot, even without a sensor as a single point of failure.

          • Because the GPS inputs only show ground speed and are completely useless for determining airspeed. Given a sufficiently strong headwind an airplane can have only little ground speed and seemingly park in the sky, but with an airspeed high enough to keep flying.
            How exactly are you suggesting to measure airspeed other than with a differential pressure system?

            • The GPS could be used to sanity-check changes in airspeed - if there's a massive change in the airspeed reported from the pitot tubes and the plane is flying somewhere close to level, there should be a large change in ground speed as well. If there's none, then that's a red flag that the pitot tubes are acting up. GPS can also be used for reliable altitude readings, which the pitot tubes could also report incorrectly if they malfunction.

              • Like I said, airspeed and groundspeed are completely unrelated. A gust of wind can massively change airspeed, but barely change groundspeed. You cannot sanity check apples by comparing them to oranges.

              • by Shinobi ( 19308 )

                Two things:

                First, as Dunkelfalke said, doing delta comparisons between ground and air speed as an indicator is not reliable at all, due to winds etc.

                Second, GPS isn't very reliable during storm conditions, like what AF447 flew through, so would thus not be trustworthy.

    • by Shotgun ( 30919 )

      Regardless of the number of sensors on the plane, the number one / primary instrument should be what is seen out the front window. I would not fly a plane that had an automatic system that I could not override, and I consider it a crime to even sell tickets for such a beast.

      • by x0 ( 32926 )

        I would not fly a plane that had an automatic system that I could not override, and I consider it a crime to even sell tickets for such a beast.

        So, no Airbus for you, right? Unless the aircraft is in direct law, it's *always* under some type of automated control.

        http://en.wikipedia.org/wiki/Flight_control_modes/ [wikipedia.org]

        m

        • Not just no airbus, the "primary instrument is what is seen from the front window" would mean no airplane that can be flown under the IFR. That means piddly general aviation stuff only, but these normally aren't allowed to carry paying passengers. Real pilots are trained to trust the instruments first, not what they see or what they think they see.

      • by Nidi62 ( 1525137 )

        Regardless of the number of sensors on the plane, the number one / primary instrument should be what is seen out the front window.

        Remember a year or 2 back when Airbus was talking about using digital windscreens in the cockpit?

    • When it fails, the stall protection algorithm thinks the plane is stuck in a nose up orientation, and tries to force the nose down... into the ground.

      Incidentally what is noted from this incident so far is that the rate of ascent during the latest accident was irregular. I don't mean rate of descent either before someone tries to be funny about all crashes having an irregular ascent.

    • They also should have experience on how to handle the situation. Last time it was the radar altimeter that caused unexpected crashes: https://en.wikipedia.org/wiki/... [wikipedia.org]

    • And this lies at the heart of why, until flight-control systems give up the 'thick stacks of bloated code model' and go back to using simpler designs that are easier to understand (massively-parallel analog circuits will eventually see their day, IMHO), I shall continue to Fly Chevy.

    • From what I read, because they used the 737 frame but moved the engines, stalling is easier, so they NEED something like the MCAS. Just turning it off is not an option. I am a software engineer, so when I read that in order to cover for an aeronautics engineering flaw of the aircraft they turned to software, I shuddered...

      • The problem is Boeing trying to shoehorn modern engines on a 50 year old airframe specifically built to have the shortest possible landing gear so the aircraft could be used in unsophisticated airports.

    • by mjwx ( 966435 )

      If I recall, the previous crash has been linked to a bad angle of attack sensor. [flyingmag.com] This sensor is only used by a new stall protection feature [seattletimes.com] in the 737 Max. When it fails, the stall protection algorithm thinks the plane is stuck in a nose up orientation, and tries to force the nose down... into the ground.

      There are several things that should happen:

      1. Interim corrective action. Disable stall protection on all 737 Max aircraft.

      My first thought as well, but apparently the engine nacelles have been moved further along the wing and higher so that the risk of stalls have increased. That's why the MCAS (anti-stall) system was introduced in the 737 MAX. If this can't be corrected to the satisfaction of the CAA and others, the type may have to be withdrawn, at least until it's re-engineered.

  • by Anonymous Coward on Monday March 11, 2019 @11:12AM (#58253694)

    www.nytimes.com/2019/02/03/world/asia/lion-air-plane-crash-pilots.html

    to summarise: we added a few lines of code that the pilots don't know about to make the plane do something the pilots have no idea the plane might decide to try and do: ie, depending on input from one little sensor, the computer might try and shove the nose into the ground.

    And the "you couldn't make this bit up" bit in the article:: on previous planes without this new software, if you felt the nose was being shoved into the ground for some unknown reason, you could (wait for it, wait for it) "pull back on the stick", and that would do what pulling back on the stick has done in aeroplanes like forever, i.e. bring the nose up (in this case, by disabling any mad sensors/sensor readings).

    (If I were the one conscious person on a plane, having to fly it, that is the single thing I would know to try to do.)

    But not any more - with this new feature, *that method of escape has been removed*.

    - We're going to crash!

    - Pull back on the stick!

    - Computer says no!

    • And good. So far a computer has caused 1 maybe 2 crashes. Pilots on the other hand yanking sticks back in error / terror have caused hundreds of crashes in history.

      Several industries including airlines, cars, and the process industry have made major strides in safety precisely by realising that for easily describable scenarios a computer should have the power to say no.

      The question isn't why wasn't the pilot given control, the question is only why did the computer fail.

  • by Applehu Akbar ( 2968043 ) on Monday March 11, 2019 @12:48PM (#58254498)

    Over 150 people just died, which is 3 Chernobyls. This means that aviation is a dead-end technology that cannot possibly be made safe at reasonable levels of cost. Germany takes the lead, mothballing all civilian aircraft now in use. From now on, Germans will use their rail network to carry domestic traffic. For international travel, Germany will build a new fleet of ships, wind powered and made of sustainable tree derived materials.

    The US will take a more measured approach. No new planes will be ordered, but airlines will continue to operate with existing craft until they age out, whereupon they will be replaced by buses. The UK will do the same, but will order one more aircraft from China, specially designed with 12 engines and parachutes for each passenger, to cost GBP 10 billion and be delivered in 2025.

    • Although I agree the feer of nuclear is often irrational, at least you have the choice of not flying. Even if you chose not to use nuclear electricity (say, you get only your own off grid solar panels), you will still be affected by a nuclear plant meltdown.

      Now, of course an airplane can fall on you while you are on the ground, but this is highly unlikely.

    • An important factor to consider is the number of people who fly successfully every day for every one that dies. You can't call something a failure if it works statistically close to 100% of the time. The important thing is that each crash be investigated and if there was a gap to the regulations that contributed to this we will close those gaps and keep it as close to 100% as possible in the future.

      Soon self driving will need the same kind of system or it will never work either.
  • Latest news from Reuters suggests that plane suffered some kind of a problem that caused it to emit smoke while in the air:

    https://www.reuters.com/articl... [reuters.com]

    Could be anything from engine trouble to a bad case of Mohammedianism.

  • How come some relatively simple sensors malfunction so often in aircraft? Especially AoA or pitot sensors?
    And why don't they run a backup system like GPS sensor, or cheap run-of the mill barometric and accelerometer sensors, and at least run sanity checks against these? This way, the aircraft could at least know that MAYBE something is s bit 'off' with the main sensors, and maybe light a 'soft warning" lamp?

The question of whether computers can think is just like the question of whether submarines can swim. -- Edsger W. Dijkstra

Working...