back to article Boeing big cheese repeats pledge of 737 Max software updates following fatal crashes

Boeing chief exec Dennis Muilenberg has repeated earlier promises that a software update for the troubled Boeing 737 Max airliners is coming "soon". In an open letter published last night Muilenberg acknowledged the "shared grief for all those in mourning" after the separate crashes of two 737 Max 8s within a few months - …

Page:

      1. Anonymous Coward
        Anonymous Coward

        Re: Want to try to reprogram it so it feels and drives like an F1?

        "For a successful technology, reality must take precedence over public relations, for Nature cannot be fooled."

        Who said it, where, when, and why? No prizes, it's just to remind readers that it's all been said and done before, and *some* people have learned lessons.

        RIP.

      2. dfsmith

        Re: Want to try to reprogram it so it feels and drives like an F1?

        Bicycles are an unstable system. As long as there are active inputs, it can be controlled quite consistently.

        1. Lars Silver badge
          Coat

          Re: Want to try to reprogram it so it feels and drives like an F1?

          "Bicycles are an unstable system". Yes, but speed will make the wheels function as gyroscopes and it gets stable, as for inputs, they are not reliable "under the influence", I am told.

          1. Kubla Cant Silver badge

            Re: Want to try to reprogram it so it feels and drives like an F1?

            Off-topic, but I think the "bicycle wheels as gyroscopes" theory is mistaken*.

            A cyclist essentially balances by steering the bike to keep the point(s) of support below the centre of gravity (when cornering, there are also centrifugal effects). It's harder to balance at low speeds because it takes longer for the wheels to move back underneath you. In this situation most riders also move their upper body over the point of support. The classic example of this is when standing on the pedals to climb a steep hill.

            If you can't steer the bike, it's very hard to balance it. Everyone can balance with their feet off the pedals, and most people can balance with no hands on the handlebars. Both at once is usually a recipe for disaster.

            * Stand your bike upside down and crank the pedals as fast as you can to get the back wheel spinning. Then tilt it. You'll get an idea how small the gyroscopic force is - especially if you have small or lightweight wheels.

            1. Prst. V.Jeltz Silver badge

              Re: Want to try to reprogram it so it feels and drives like an F1?

              "Off-topic, but I think the "bicycle wheels as gyroscopes" theory is mistaken*.

              me too

              "crank the pedals as fast as you can to get the back wheel spinning."

              just think, they are spinning a lot faster than when you are dawdling along at 5 mph , perfecly stably , at that point there is near zero gyro force

            2. Michael Wojcik Silver badge

              Re: Want to try to reprogram it so it feels and drives like an F1?

              I think the "bicycle wheels as gyroscopes" theory is mistaken

              Yep. There's been quite a lot of research into this, including various physicists both doing the calculations and creating modified bicycles which have minimal gyroscopic effects.

              It's not the cyclist either; bicycles will tend to remain standing while they're moving even if they're unoccupied. This has also been demonstrated in various ways, and you can do a simple home test by standing next to a bike and giving it a good shove forward using the handlebars. If you push it straight, it should stay upright - though it may turn - for a decent distance.

              It seems that bicycles are stable while moving because as they start to lean to one side or the other (i.e. as their vertical symmetry breaks) they naturally steer into the lean and turn. (This is due to the fact that bicycles are designed so the front wheel contacts the ground behind the steering axis.) That converts the lean into centripetal acceleration rather than linear.

              See e.g. this and this.

      3. rsole

        Re: Want to try to reprogram it so it feels and drives like an F1?

        Did someone mention the space shuttle, essentially a brick falling elegently with the assistance of software.

        1. W.S.Gosset Silver badge

          Re: Want to try to reprogram it so it feels and drives like an F1?

          Or the old 50s/60s' test-pilots' saying:

          "If you put a big enough engine on a BRICK, it will fly."

      4. ridley

        Re: Want to try to reprogram it so it feels and drives like an F1?

        Almost all modern military fighter aircraft are purposely designed to be unstable and want to fly backwards it is only the fly by wire system that allows them to fly. Doing this allows them to be extremely fast at maneuvering.

        On the other hand the F117 was just plain un aerodynamic and had to be flown by computer, hence the Woblin Goblin nickname.

        1. Prst. V.Jeltz Silver badge

          Re: Want to try to reprogram it so it feels and drives like an F1?

          first one i heard of that *needed* computers because of its shape was the stealth bomber, which u can tell just by looking at it , youd never fly it manually

      5. Reliance

        Re: Want to try to reprogram it so it feels and drives like an F1?

        There is no way you can make software fool physics and force physics to turn an unstable system into a stable one.

        The US F-117, for example, is an unstable aerodynamic platform run by software that makes it act stably. There are others.

        1. Anonymous Coward
          Anonymous Coward

          Re: Want to try to reprogram it so it feels and drives like an F1?

          "The US F-117, for example, is an unstable aerodynamic platform run by software that makes it act stably. There are others."

          But they haven't changed the physics. Although they've apparently fooled the people in the procurement departments at various War Ministry offices.

    1. ChrisC Silver badge

      Re: Car analogy

      ""...you could program it so the car would feel the same to drive as it did with the original engine and wheels."

      Maybe I or we could but it appears to me no company can, the best they can do is the feeling of a video game."

      Many moons ago watching a fly on the wall series at the Empire Test Pilot School following already qualified pilots earning their test pilot wings, ISTR one episode showed us a heavily modified bizjet (I want to say a Dominie, but I could be wrong) which could be configured to fly with the handling characteristics of something else. And more recently I was reading the excellent "Into The Black" book about the development of the Space Shuttle, which also made reference to use of a modified bizjet to allow the pilots to evaluate the shuttle handling characteristics during the glide down to landing.

      So whilst it clearly wouldn't be feasible to use a mixture of modified flight control software and aerodynamic mods to give a large lumbering aircraft the same handling characteristics as a much smaller nimbler aircraft, it does seem that by starting with something more capable than the aircraft you're trying to emulate, you can artificially slug its performance to approximate the desired characteristics sufficiently well to be of genuine use in roles where the accuracy of the approximated characteristics is rather important.

      Which is essentially what Boeing were trying to do with MCAS - artificially reducing the effects of the additional lift generated at certain points in the flight envelope to match the characteristics of the older 737s. And that's all well and good so long as it all works exactly as intended. It might even be OK if it stops working so long as the pilots are aware of what problems can arise when it goes wrong and how to work around the problem. What's clearly not even remotely OK to anyone outside of the Boeing board, is to introduce such a system without feeling the need to let the pilots know about it.

      1. Anonymous Coward
        Anonymous Coward

        Re: Car analogy, software emulation

        Using software to simulate the real world can work great but the last few I've used have done such a poor job of it I question my own memory of it being done right.

        The last car such car I drove had drive by wire for throttle and steering and electric assist and control for brakes, even the parking brake was electric and only worked under very specific conditions. Engine rpm often had little relation to throttle position and I managed to have engine RPM decreasing while throttle position was increasing, and vice versa, Steering was completely insulated from the car and road. Thanks to poor tires as well as electric steering was unable to predict when lost of traction was about to occur even on asphalt. With many cheap cars today brakes will not allow threshold braking and there is often no easy way to turn off or adjust traction control or ABS.

        On more than one occasion I have found myself fighting software for control of more than one vehicle.

        Of course a friend has pointed out that their Panamera feels great and has none of those problems. Which shows software can be as good as, and even much better than, mechanical systems if enough time and money is spent. Then again I'm pretty sure his model had a problem with steering software.

        Since I do not get to use those cars that get software right I'm left feeling like the some pilots of the 737's in question.

        IMO Boeing should not use software to hide or enhance flying abilities but that isn't what the buyers want and like cars it's the average buyer who decide such things.

        1. Mark 85 Silver badge

          Re: Car analogy, software emulation

          Upvoted because you have described the problem exactly. I suspect that some point, the airlines would like "pilots" that are basically meatbag window dressing. Hire basically a computer operator who has no knowledge of the guts and mechanics of flying.

          There's a lot of similarities betwixt flying and driving. Seat of the pants and feedback from the controls are big part of it. Once everything goes full computer control and human "takeover" controls aren't needed then fine, but until that happens we need the control and the feedback the vehicle gives us.

          1. eldakka Silver badge

            Re: Car analogy, software emulation

            I suspect that some point, the airlines would like "pilots" that are basically meatbag window dressing. Hire basically a computer operator who has no knowledge of the guts and mechanics of flying.

            That's true of any and all professions that require skilled (read: well-paid rather than minimum wage-slaves) staff.

            IT have been doing it, trying to 'dumb-down' systems so that organisations don't need highly technical, expensive, staff to set stuff up, configure it, tune it, keep it running smoothly. They'd rather have a room full of unskilled staff following flow-charts and panicking when the task goes off-script.

            Airlines have done it over time, at one time aircraft required a flight engineer(s) in addition to pilot/copilot to manage and monitor all the complicated systems that go into making an aircraft fly. They've managed to get rid of those by having more automation. Next on the chopping block will be getting rid of one of the pilots. That is, making the system 100% automated, thus only needing a single pilot to deal with emergencies, who could also be the chief steward for the flight as most of the time they won't be needed in the cockpit at all for the entire flight.

            Hospitals are trying to reduce the number of doctors and nurses needed per patient, by using computer expert systems to help with diagnosis.

            Factories try for more automation to get reduce even the minimum wage-level staff they need.

            It is all about cutting costs. And the most obvious cost to cut is the meatbags that require wages, and breaks, food, sanitary facilities, moan and whine to management or unions, complain and take the company to court when they get underpaid or other payroll shenanigans, can be whistle blowers to expose unethical (if not downright illegal) business activities, etc.

            1. Anonymous Coward
              Anonymous Coward

              Re: Car analogy, software emulation

              "IT have been doing it, trying to 'dumb-down' systems so that organisations don't need highly technical, expensive, staff to set stuff up, configure it, tune it, keep it running smoothly. They'd rather have a room full of unskilled staff following flow-charts and panicking when the task goes off-script."

              IT has been automating tasks and removing jobs for decades. IT people forgetting that they are supposed to be adding value and efficiency to an organisation OR an organisation assuming that IT adds no value or efficiency to an organisation often results in outsourcing and the effects you describe.

              The question is where the problem really lies and if the course can be changed before reaching the situation described above - in many cases, the answer is find a better organisation...

              For Boeing, losing one aircraft was "clearly" operator error and failing to follow the documented procedures and the automation remained unchanged. The second aircraft changed that and the changes in how Boeing certifies planes in the future will be interesting. But the automation will remain.

            2. Anonymous Coward
              Anonymous Coward

              Re: true of any and all professions that require skilled

              Showing the connections like that deserve a 10X thumbs up.

              It's all about cutting costs, increasing profits, and ensuring the rich get richer while the poor get poorer.

              And we now have decades of stats showing just that, the rich ever richer, while the middle, the skilled, gets less for more education, effort and time. Countries with rights and laws are made to compete with the most corrupt.

              Everything is connected and those connections can be seen by looking at a couple of 737 crashes.

            3. kiwimuso
              Facepalm

              Re: Car analogy, software emulation

              "It is all about cutting costs. And the most obvious cost to cut is the meat-bags that require wages, and breaks, food, ....."

              Perhaps 'management' should think about automating themselves then, as they are quite clearly the most expensive meat-bags on the company payroll.

              It would solve the "golden handshakes" cost to the company as well. You know the ones where the CEO fucks up, the company loses vast amounts of money and maybe the share price drops like a stone and when they finally let the CEO go they still get paid a handsome payoff, because "it's in their contract"!

              Most other staff just get fired if they screw up.

              Icon, 'cause it seems obvious to me.>>>>>>

          2. veti Silver badge

            Re: Car analogy, software emulation

            There's a joke that's been going around in aviation circles for a while now, that the ideal flight crew on a modern passenger jet is one pilot and one dog. The pilot is there to feed the dog, and the dog is there to bite the pilot if he tries to touch the controls.

        2. Prst. V.Jeltz Silver badge

          Re: Car analogy, software emulation

          "The last car such car I drove had drive by wire for throttle"

          virtually every new car does now , and has for some time

          1. Anonymous Coward
            Anonymous Coward

            Re: Car analogy, software emulation

            Indeed, so very true.

            Drive by wire, throttle, steering and brakes and now CVT's are so common and for so long that there is a generation of drivers who have no idea of what is happening while driving. They have been completely insulated from the car and road their whole life.

            The same generation feels that they should survive every crash, that crashing shouldn't be so feared. At worst crashing should only mean a reset and a new car. Software and tech has people believing that the laws of physics and mechanics can largely be ignored, and they are not wrong.

            That enables most drivers to drive far faster and in poorer conditions than they could otherwise.Expecting anything else would be similar to expecting pilots to fly airplanes.

            BTW there are cars that will allow access to the software to improve road feel but they have yet to give the feedback of equally expensive mechanical systems. For those that want that feedback, particularly in a RWD front engined car with proper balance there are increasingly few options.

            Shame that because electric powered cars can be, IME, even better when it comes to balance, control, feel and feedback.

  1. Commswonk

    Simply Ghastly...

    Whatever happened to the stick shalers of old that used to warn pilots of an imminent stall? Dispensed with as "old hat" by the look of it.

    I am more than a little alarmed that given the number of other flight characteristics being monitored by both pilots and software (forward air speed, rate of climb, attitude, altitude) this MCAS seems to have been fully autonomous in that it wasn't paying any heed to other inputs, and having decided on a course of action kept putting the nose down despite the pilot(s) applying manual corrections. It seems to have been functioning entirely on its own without reference to anything else.

    I know that it is in woefully bad taste to make any sort of joke about it, but it wasn't designed by an E*** M*** was it?

    Maybe it's the way the article was written but to me it signals very long term trouble for Boeing.

    1. Anonymous Coward
      Anonymous Coward

      Re: Stick shakers

      Stick shakers remind me of automatic spark advance for automobiles, a step in the progression towards having no operator at all. If an operator really needs that technology maybe we should reconsider if there is a need for the operator or just more technology?

    2. Alister

      Re: Simply Ghastly...

      The stick shaker is still included - and may in fact have been a contributory factor in the Ethiopian incident, as it could have added another distraction for the pilots at a critical period.

      1. ilmari

        Re: Simply Ghastly...

        That is a very good observation!

        Stick shaker and stall warning comes on. Pilots think plane is nearing stall.

        Nose dips down - yep, definitely stalling for real!

        But how can the airplane stall at 300 knots? The airspeed indicator must be wrong.

        They manage to recover, try to guess their airspeed, make sure they're going fast enough.

        Still the plane repeatedly stalls.

        Icing? In Africa? But it would also explain airspeed indicator fault.

        Are spoilers and flaps stuck? Troubleshoot the hydraulics, get them to ret*crash"

        ... And they failed to notice the trim was moving on its own, and didn't run the right checklist to deal with that.

        1. W.S.Gosset Silver badge

          Re: Simply Ghastly...

          This is probably the best description yet of the actual events&thought-processes in the cockpit at the time.

          It's for situations like this that the Flight Engineer used to be a mandated part of hte flight crew. Take a lot of the engineering aspects off the flightcrew when the shit hits the fan, so that the flightcrew can focus on the aeroplane, not on what its machinery is trying to do.

        2. paulll

          Re: Simply Ghastly...

          "Stick shaker and stall warning comes on. Pilots think plane is nearing stall.

          Nose dips down - yep, definitely stalling for real!"

          More likely:

          Stick shaker and stall warning comes on. Pilots think wtf? Why does this thing think it's stalling?

          Nose dips down -wtf? What the hell's it doing? It's not responding either. Wait I guess I can trim it out. Phew.

          10 seconds later, MCAS cycles again. But a bit lower. Repeat.

          Besides the obvious oversight with the AoA inputs, it's mind-boggling that this thing would engage while sinking at low altitude. I imagine it's supposed to work in a low altitude climb as part of its normal function and disengage in a low altitude sink but only accounts for instantaneous vertical acceleration...which would be an oops. (Once you get low enough it becomes pointless to try to recover lift by pitching down, it makes more sense to try for a potentially survivable belly-landing, rather than go nose-first and explode.)

    3. upsidedowncreature

      Re: Simply Ghastly...

      What is an E*** M***?

      Edit: Duh, never mind, the coffee kicked in and I figured it out.

    4. MMR

      Re: Simply Ghastly...

      It is a bad taste joke in a way. E*** M*** is not a one man company and T**** employs an army of programmers. Perhaps the same type of programmers who are able to write software for S****X which accidentally is another E*** M*** venture.

    5. Anonymous Coward
      Anonymous Coward

      Re: Simply Ghastly...

      Stick shaker still exists and works on the Max. Check out the preliminary report of flight 610. One of the line graphs you can see is the stick shaker - and it's on from the start of the flight until the end. Which no doubt added another source of confusion to the pilots.

    6. sanmigueelbeer Silver badge
      Holmes

      Re: Simply Ghastly...

      Whatever happened to the stick shalers of old that used to warn pilots of an imminent stall?

      The trigger points to the stick shaker is no longer just stall.

      In the Lion Air incident, the stick shaker was activated the moment the aircraft was airborne until it crash. In previous two (or three) flights, the pilots also observed the same thing. Except, those pilots "accidentally" (or mistakenly) disabled the MCAS and made it.

      Another thing about the stick shaker is how company policy defines it. In the initial Lion Air investigation recommendation, the pilot has the "final say" if the plane is deemed "flyable" or not. But when the stick shaker got activated immediately after takeoff that alone is grounds for any pilot to invoke that the flight is not "flyable" and turned around. Apparently, this policy is not "standard" in Lion Air. However, even if this was standard, it wouldn't make any difference because the aircrew didn't know how to disable MCAS.

      software update for the troubled Boeing 737 Max airliners is coming "soon".

      I hope Boeing did extensive testing of this software in a LIVE aircraft.

      So good news and bad news:

      Bad news: Operators and passengers are going to be caught between a rock and hard place. Ever operator will have to take the risk and load this software. Immediately.

      Good news: Due to the two incidents, it would seem that every pilot (will) know how to disable MCAS. Maybe some pilots will just disable MCAS before take-off.

      1. W.S.Gosset Silver badge

        Re: Simply Ghastly...

        > In previous two (or three) flights, the pilots also observed the same thing. Except, those pilots "accidentally" (or mistakenly) disabled the MCAS and made it.

        If you're thinking of the 2 Yank pilots: minor correction:

        they disabled the AutoPilot, and the problem went away.

        .

        Which then raises the questoin of the link between the autopilot and MCAS...

        1. sanmigueelbeer Silver badge

          Re: Simply Ghastly...

          If you're thinking of the 2 Yank pilots: minor correction:

          I'm talking about the pilots to the two previous Lion Air flight before the fatal one.

      2. imanidiot Silver badge

        Re: Simply Ghastly...

        There is no way to disable JUST MCAS. Pilots can only disable ALL electrical trimming (including the manual trim switches on the control yoke) which leaves them with only the cable driven manual trim system using a handcrank on the trimwheel in the cockpit. That is a workout to say the least (IIRC 120 turns for full stroke from full down trim to full up trim)

        But pilots might now know what MCAS is and why their plane keeps trimming nose down in 10 second bursts. However, if this information is not in the flight manual it remains to be seen they think of it during a time of crisis. It's come out yesterday that the Lion Air pilots where frantically searching in the manual to find an explanation and couldn't find anything. In both of these accidents the pilots didn't have time to decide to turn back. Once the wheels left the ground they were immediately in full emergency mode trying to find out why their aircraft was trying to kill them. I can understand why they kept climbing. Altitude is life in aviation. It gives you time and options.

  2. Groaning Ninny

    Following a previous poster to a different topic (Amazon cargo plane crashing in US), I started following the views on PPRuNe. The Ethiopian flight is discussed here:

    https://www.pprune.org/rumours-news/619272-ethiopian-airliner-down-africa.html

    Very interesting, and also quite possibly used as a primary source for various news sites.....

  3. Kennelly

    To save you visiting the full report for this info - the fig 5 charts extract spans a time period of approx 13 mins, the pilots were fighting the auto-adjustment for control over a period of seven minutes

    1. el kabong

      Software as a panacea, they used it before to sell us clean Diesel engines

      Now they are using it to sell us air travel that is both cheap and safe, it's magic.

      Is there anything software cannot do?

      1. Yet Another Anonymous coward Silver badge

        Re: Software as a panacea, they used it before to sell us clean Diesel engines

        Yes it can make flying amazingly safe.

        Of course you could go back to the days of piston engines and carburetors and pilots taking star sightings and hoping to find land before they hit cloud.

    2. Anonymous Coward
      Anonymous Coward

      The question I want answered is does a full 5 degree horizontal stabilizer trim provide so much nose-down moment, full elevator deflection is unable to correct it?

      If so, people at Boeing need to be going to prison for a long time.

      1. Anonymous Coward
        Anonymous Coward

        Because the more I think about it, depending on the slope of the CL/alpha curve for the wing and the horizontal stabilizer respectively, if the nose dips a little, causing alpha on the wing to drop, the wing will be providing less lift and less nose up moment, resulting in an unstable accelerating nose-down dive.

      2. Anonymous Coward
        Anonymous Coward

        That seems to be the issue, that MCAS is capable of pushing the nose down beyond the limits of the pilot being able to counter it. I'm reminded of how the Thinkpads (in the days before trackpads) had a little red 'nub' in the middle of the keyboard to control the mouse - occasionally the mouse would start to "drift", you'd correct it with your finger, confusing the laptop into drifting it to counteract your movements. Only lifting your finger off would allow it to realise and stop moving.

        It's also perhaps an overly simplistic interpretation, but I'm reading that the MCAS can be over-ridden but that it can in certain circumstances start adding trim again to an already-MCAS-trim level (effectively adding 4x as much stabiliser trim than that envisaged by Boeing)

        1. Anonymous Coward
          Anonymous Coward

          I think Air France crash that had the two pilots inputting opposite correction, one nose up in panic, one nose down down to stop a stall. Each one over compensated for the other, and in the end the aircraft fell from the sky. :(

          1. Anonymous Coward
            IT Angle

            Thumb down?

            Ok, I may have oversimplified that one. But similar happened, as the feedback on the stick did not clearly show the nose up was from the other co-pilot. I'm happy to be corrected though!

            1. SImon Hobson Silver badge

              Re: Thumb down?

              as the feedback on the stick did not clearly show the nose up was from the other co-pilot

              That is about it. One pilot realised what was happening but could not feel that the other pilot was holding the stick fully back - made worse by the way Airbus have put the sticks (basically a gaming joystick) in the cockpit sidewalls so one pilot would have to look across past the other to see what they are doing.

              In a traditional "mechanical" system, the two sticks would be mechanically linked - so one plot would be able to feel what the other was doing. It was eventually realised what was happening, but not before they ran out of altitude in which to recover the situation.

  4. Thought About IT

    Finger of blame

    If, as seems to be the case, the algorithm relied only on the device measuring angle of attack, when there are completely independent devices measuring airspeed and rate of climb which could have been used to check for faults, it's a choice between gross negligence by the designers and cost cutting by the accountants. In either case, someone should go to prison.

    1. A.P. Veening Silver badge

      Re: Finger of blame

      "In either case, someone should go to prison."

      After a severe flogging.

      NB: I object to capital punishment as there always is the possibility of finding an innocent guilty, I don't object to corporal punishment that leaves no more permanent injury than a scar.

    2. Anonymous Coward
      Anonymous Coward

      Re: Finger of blame

      Whistle blowers alleged that after Boeing merged with MD the MD culture (that produced the DC-10 within a tight budget) emerged over the Boeing quality control attitude. The B-787 was criticised as built to an inflexible deadline and budget, with safety implications.

    3. Anonymous Coward
      Anonymous Coward

      Re: Finger of blame

      What is the Boeing 737MAX manoeuvring characteristics augmentation system:

      https://theaircurrent.com/aviation-safety/what-is-the-boeing-737-max-maneuvering-characteristics-augmentation-system-mcas-jt610/

    4. imanidiot Silver badge

      Re: Finger of blame

      But how do you know your IAS and rate of climb indicators are correct? Not to mention the fact that AoA is almost completely independent of both airspeed and rate of climb. You can stall an aircraft at any speed at whatever rate of climb if you want to.

Page:

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Biting the hand that feeds IT © 1998–2021