Announcement

Collapse
No announcement yet.

Breaking news: Ethiopian Airlines flight has crashed on way to Nairobi

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by Gabriel View Post
    No. I'll try to explain, but let's start forgetting about the elevator, as I understand it has nothing to do with what happens with the trim (rather the opposite).

    The horizontal tail has 2 movable parts. The "fixed" (not so fixed) stabilizer in the front and the elevator in the back. The elevator is controlled with the control column. The elevator is hinged in the front, hinged to the back of the stabilizer, so when it moves the front part of the elevator stays in place and the back moves up and down. Under normal conditions they are hydraulically actuated, so aerodynamic forces cannot move it. But it is easy to visualize that if you deflect the elevator up (to push the tail down and the nose up), there will be aerodynamic forces that will tend to return it to the equilibrium position. Very much like a weather vane.

    The "fixed" stabilizer is really not fixed. It is hinged in the back. So, if you could move it freely, the back part will stay in place and the front of the stabilizer moves up and down. It can be intuitively seen how, if you start from the equilibrium position and displace the front of the stabilizer a little bit say up, and let it go, it will not tend to return to the equilibrium but rather move more up. It's like a weather vane facing back (this is pretty much what happened with the Alaska MD80 accident).

    The stabilizer is controlled by a jackscrew that goes through a bolt in the front of the stabilizer. When the jackscrew rotates in one direction or the other, it moves the nut up or down and the elevator with it. If the jackscrew doesn't turn, the stabilizer won't move, so you will not have the unstable behavior described above. I mean, the stabilizer will try to move farther away from the equilibrium, but the jack screw won't let it. How? Because the threads of the nut will make pressure against the thread of the jackscrew. The stronger the pressure, also the stronger the friction. And when is the pressure stronger? When the loft on the stabilizer is stronger, and that happens the farther away the trim is from the equilibrium position and the faster the airspeed (because the lift is proportional to the angle of attack, of the stabilizer in this case, and the speed squared).

    The trim wheel doesn't control the elevator. It controls the stabilizer. They are mechanically connected to the jackscrew. Rotating the wheels rotates the jackscrew.
    If the stabilizer is very out of trim (very far from the equilibrium position) and the airspeed is high enough, the friction between the jackscrew and the nut can be very high and it can be almost impossible to manually rotate the jackscrew using by manually rotating the trim wheel.
    Thank you that makes sense. So the elevator is hydraulic which means power assisted. The stabilizer trim is pure mechanical linkage and the forces will increase as speed increases and the controls move further into bias to the point where the pilot can't move them... not good if that is the only recovery available to the pilots. So, the jack screw motor is power by the switches and the hypothesis is now that they could not move the trim wheel due to forces, re-engaged the trim controls in the hope that the switches would be able to move the stabilizer back to neutral.

    In that scenario, it becomes 100% Boeing's fault.

    Comment


    • Originally posted by Schwartz View Post
      Thank you that makes sense. So the elevator is hydraulic which means power assisted. The stabilizer trim is pure mechanical linkage and the forces will increase as speed increases and the controls move further into bias to the point where the pilot can't move them... not good if that is the only recovery available to the pilots. So, the jack screw motor is power by the switches and the hypothesis is now that they could not move the trim wheel due to forces, re-engaged the trim controls in the hope that the switches would be able to move the stabilizer back to neutral.

      In that scenario, it becomes 100% Boeing's fault.
      Actually, from what I understand, the pilot can grab the spinning wheel and stop it, which means that the electric motor actually has less force than the pilot's hand. Would the jack screw motor even be able to overcome the forces?

      Comment


      • Originally posted by vaztr View Post
        Gabe - the video is 'no longer available'???
        Strange tings going on. I can confirm that the video WAS available. I saw most of it. And the link came from the Leeham news article. The article said that they had been working with Mentour in the sim and that them made a video, and at the end of the article there was a link to the video. Now the article still says that they worked in the sim with Mentour but there is no mention to a video, the link is gone, and the video doesn't exist anymore in the Mentour channel. A guy called Scott, who I don't have the slightest idea who he is, repeatedly said in the comment that the video was removed at the request of Mentour's employer (Ryanair), who seems to have quite a few of MAX in orders.

        Now I wish I had downloaded the video while I could. I am sure someone MUST have done it, but I could not find any references or re-posts.

        So, I will try to make a sort of transcript of the Mentour's video, which is in fact a transcript of my memory of it, some 10 hours later. Words in brackets are my comments.

        The video was titled something like "Pilots got it right, but it went wrong".

        [Video starts as always with Mentour sitting in his couch]

        This is a video that we prepared in collaboration with Leeham News because we both are upset with the bashing of the pilots. There is a lot of rant in the news and social media asking how it is possible that the pilots of Ethiopian fell in the trap after everybody, and especially 737-MAX pilots, knew what happened with Lion Air. After that accident, Boeing disclosed the MCAS and published a service bulletin instructing to treat a MCAS malfunction as a trim runway and follow the trim runaway procedures. So why didn't they follow the trim runaway procedure and just use the cutout switches to disconnect the trim? It is a fair question.
        This video shows a possible scenario of how the pilots might have followed the trim runaway procedure and still lose it. We still don't know what happened, but if in the next days the investigators report that the cutout switch was found turned off, remember this video.

        You first need to understand some things about the 737 [he didn't say MAX]. The 737 has a lot of sensors outside the plane. It has a pitot tube on the left side and another one on the right side. The pitot tube is used to measure how strong the air is hitting it. There are also static ports on the left and right sides, that measure the atmospheric pressure outside of the airplane. And there are also angle-of-attack vanes, one in each side, the measures the angle that the air hits the airplane and the wings. These sensors are used to display several parameters in the flight instruments. The left sensors are used for the left instruments which are the Captain's, and the right-hand sensors are used for the right-side instruments, the first officer's.

        Now, in older planes [older 737s included] the airspeed was calculated based only on the data from the pitot and static ports. In modern planes [including many 737's for sure the NG but possibly even earlier] the angle of attack is taken into account too because if the air hits the pitot at an angle it doesn't feel the same pressure [that's true, however the difference is just a small "fine tuning", it worked and still works pretty well just with the pitot and static]. So if one of the angle of attack vanes fails and sends incorrect information, the speed between the left and right airspeed indicators will differ [just a little bit] and we'll have a speed disagree alarm.

        Since the Air France accident, a lot of focus ha been put in the importance of following the airspeed disagree procedures to keep the plane flying safely without knowing your speed [just look at the stby instrument, in this case the stby instrument must have matched one on the airspeed indicators, this is a speed disagree but should NOT be an unreliable airspeed situation]. In the case of the 737, there are 2 conditions: Flaps up and flaps down. In this case the condition would have been flaps up. For that we need to set a pitch of 4 degrees nose up and 75% of N1 on the engines. This will keep the plane flying within the envelope everywhere from sea level to the max cruise altitude, but in different conditions. 75% N1 makes much more thrust at low altitudes that at high altitudes, so at high altitudes the plane will likely slow down to a slower speed and establish in a smooth descent, while at low altitudes it will climb and, more importantly, increase the speed. You'll see why this is important.

        So, airspeed disagree, the pilot sets 4 degrees pith and 75% N1 and the plane starts to accelerate. The trim wheel starts to move. This is not unusual, since the plane is accelerating, the trim moves to compensate for that [yes, that's the speed trim function, which I suppose would not work in a speed disagree situation]. But the pilot starts to feel it strange. The controls are heavy and he finds himself fighting to keep the nose up in 4 degrees, he trims up with the thumb switch, and it improves, but then the plane trims down again and it worsens again, so he trims up again and so on, and the plane is accelerating so in every cycle the controls feel heavier and heavier.

        The focus of the crew starts to shift from the speed disagree to the control issue. They start to evaluate, are you trimming? Do you see what it's doing? And eventually they decide that they have a trim runaway, so they go through the memory items. Hold the control firmly, autopilot off but it is already off, autothrust off, evaluate, the uncommanded trim continues, so trim cutout switches, off. At this moment the trim stops actuating, but the plane can be quite out of trim already, we lost the thumb switches, and the speed in increasing. As speed increases, you need more and more force on the yoke to keep the nose from going down. You need to use the trim wheel to re-trim. But the trim also becomes harder. The Boeing 737 manual [he didn't say MAX] says that the trim wheel can become very heavy at high speed [I don't remember if he also said "and at high trim deflections", but it is at high trim deflections. If the plane is not grossly out of trim, the trim wheels can be turned manually to trim the plane at any speed]. It can reach the point where it is impossible for the pilot to keep pulling on the yoke hard enough, and to manually turn the trim wheel. So let's see how it looks and feels like in the cockpit.

        [I had several reservations when Mentour said that last paragraph, reservations that were confirmed and I will expand with the sim part]

        [Mentour is now in the left seat of a 737 SIM that they did not say it was a MAX and I am sure it was not, rather it was an NG, there is another pilot in the right seat, the wide-angle camera is in the middle behind both pilots, you can see the wind screen, the instrument panel, the center console, both pilots and the control columns. The size and resolution is not enough to read things like altitude and airspeed, but you can tell if the speed is increasing or decreasing, and you can also tell the approximate vertical speed by the position of the vertical speed needle, the attitude indicator is visible but with the attitude range during this flight it looked all the time almost level, it is daylight IMC and only clouds textures are visible out of the windshield throughout the simulation]

        [Chime]
        FO: Speed disagree
        CAP: I agree. Speed disagree memory items please.
        FO: Autopilot disengage
        CAP: Disengaged
        FO: Flight director, disengage (and he disengages it)
        FO: Autothrust disengage
        CAP: Disengaged
        [the trim wheel starts moving forward and back, forward and back without pause, and continues until the trim cutout switches are turned off]
        FO: 4 degrees nose up, 75% N1
        CAP: 4 degrees, 75%
        FO: You are descending
        CAP: Yes, it's a but heavy to keep the nose up.
        [FO looks at the trim wheel for some seconds]
        FO: DO you have problems with the controls?
        CAP: Well, it trims down, I can trim up but as soon as I release it it trims down again.
        FO: It looks like a trim runway.
        CAP: I agree, trim runaway memory items please.
        FO: Hold controls firmly
        CAP: Yes, already doing it.
        FO: Autopilot disengage
        CAP: It is disengaged
        FO: Autothrust disengage
        CAP: It is disengaged
        FO: If the problem is not solved, trim cutout switches to cutout.
        CAP: Let's evaluate a second... When I trim up the trim responds, but as soon as I release it it trims down again. And it is getting heavier

        [1- Why don't you trim more nose up if it is getting heavier and you do have thumb switch authority? 2- This is NOT a realistic MCAS event, with the MCAS, when you use the trim switch in either direction the MCAS will STOP immediately its input and resume ONLY AFTER 5 seconds after you released the thumb trim switch. In this simulation, the trim wheel would spin back in response to the nose up thumb switch and IMMEDIATELY spin forward again. 5 seconds is not a lot but it gives you time to re-trim the plane, especially since the 5 seconds will not even start to count until you are done re-trimming the plane]

        FO: So let's proceed with the cutout switches?
        CAP: Proceed with the cutout switches.
        FO: [switches the cutout switches off]
        CAP: Now I am pulling buck pretty hard already, and it is getting heavier as the speed increases. How much trim do we have?
        FO: Four units nose down.
        CAP: Four units nose down? It could be much worse. So now, we would use the wheel trim to trim back up, but for the sake of the exercise, since it could have be worse, trim down please. Guys don't do this at home, it's only for the sake of the exercise, we should trim up, not down.

        [I want to mention that 4 units may not be a lot but it didn't need to be that much either, since he had authority over the trim switch and that was overriding the trim runaway, as it should, he could have trimmed the plane to a more comfortable condition before using the trim cutout switches. Nothing would have prevented him for applying say 15 seconds of continuous nose-up trim and the trim would have responded and the MCAS or whatever was causing the runaway would not have interfered with that. In the Lion Air crash, during a few minutes the pilot was trimming up in opposition to the MCAS that would trim down each time 5 seconds after the last thumb switch intervention, managing to keep the trim on average more or less in the same position, and they could have done even better by trimming up a but more]

        FO: [pulls the trim wheel crank lever out and starts cranking forward i.e. nose-down trim, for a couple of minutes]
        CAP: It is getting harder and harder.
        CAP: [at some point stops grabbing the yoke from the horns and instead passes his arms over the yoke interlacing both hands behind the column, as if he was hugging the yoke, to be able to make more force]
        FO: [Keeps turning the trim wheel with more and more effort, the wheel is evidently getting harder and harder quite fast, due to a combination of increasing stabilizer deflection as result of the spinning and the airspeed increasing due to the 4 deg / 75% N1 / low altitude condition.]
        CAP: I am using almost all my strength to hold it. And we still have a couple more points to go to reach the full nose-down trim.
        FO: [Struggles to keep turning the wheel, and at a point he is unable to keep moving it]
        CAP: Ok, start trimming up now.
        FO: [Tries during several seconds but is unable, while Mentour fights harder and harder to keep the nose up. I cannot see the altitude or the airspeed, but the vertical speed is a bit down, perhaps some 1000 fpm down].
        CAP: [looks at the camera or rather at a person next to the camera] Press the red button. Press the red button now! [not sure if he was referring to stopping the recording or stopping the simulator].

        [The video returns to Mentour sitting in the couch, and there is where I stopped watching it. Because I had to leave. My plan was to finish it later, which failed]


        There are a few reservations that I have. Many of them I already mentioned, others not yet, but I will summarize them below.

        But first, I want to acknowledge that, despite my reservations, the situation described in the video is quite critical, that my reservations are done in hindsight and from the comfort and safety of my armchair, and that some of the things that I mention could have been done are not in the procedures. So keeping all that in mind:

        - The speed disagree should have been a simple speed disagree and not an unreliable speed event. The difference is that in a simple speed disagree you still can tell which instrument is at fault. How? By looking at the standby instrument that MUST have matched one of the airspeed indicators. Why? Three scenarios: a) The standby airspeed indicator uses AoA data from the faulty AoA indicator, then it matches the airspeed indicator on that side. b) The standby airspeed indicator uses AoA data from the good AoA indicator, then it matches the airspeed indicator on that side. c) The standby airspeed indicator is "old style" and uses only pitot and static data, with no adjustment for AoA, then it matches the airspeed indicator on the "good AoA vane" side, because the AoA was actually low and that is where corrections are not needed.
        The Lion Air crew of the flight before the accident quickly realized that which speed indicator was incorrect.

        - Even disregarding the previous point, the degree of disagreement between the left and right airspeed indicators would have been quite small 10 or 20 knots, enough to call it a disagree, but not enough that you cannot take any of them as good and fly the plane.

        - If the airspeed disagree was the result of an AoA indicator that was faulty since the beginning, the disagreement would have started at rotation. They had plenty of time to figure it out before the MCAS started to interfere and divert heir attention from the speed disagree issue. The MCAS would have started after flaps retraction.

        - With the MCAS acting, there is nothing that prevents you from trimming nose-up as much as you want, and there is nothing that prevents you from stopping the MCAS immediately every time it starts acting. For both things you use the thumb switch.

        - The MCAS would not start acting again immediately after you release the trim thumb switch as seen in the video, but 5 seconds later.

        - There was no need to be so grossly out of trim by when you disconnect the trim cutout switches. Note that if you have the plane trimmed for positive lift, even if slightly positive and unable to balance the wight (let's say trimmed for 0.2 Gs at the present speed), that is a quite out-of-trim plane however, as speed increases, it will be easier, not harder, because that small but positive lift will be greater and greater. You need to have the plane trimmed for negative Gs to become worse as speed increases.

        - A couple of minutes spinning the trim wheel nose down for the sake of the exercise?

        - Note that in the Lion Air crash, the problem was not a hard-to-turn trim wheel or a harder elevator as peed increased. The main problem was that they were doing ok by trimming up with the thumb switch in response to the MCAS trimming down, until they stopped doing it, then they let the MCAS trim way down and as speed increased, yes, they could not pull back hard enough, but they still had thumb trim switch authority that they didn't use.

        - Ok, so the captain cannot hold the column back anymore, the FO cannot turn the trim wheel, at what point does the captain says "stop trying and failing to move this wheel and helps me pull up"? The speed was obviously increasing way a lot. They had 2 airspeed indicators in agreement and another one almost in agreement, the controls are getting harder and I can't prevent the plane from going down? First help me pull up, we need to get away from the ground, then let's reduce the speed, throttles to idle, nose forget the 4 degrees, let's go for 15. When the speed goes down, the controls will be much lighter and the trim wheel too.

        - A last resource could be to re-activate the trim cutout switches. The runaway will start again but so will your authority over the thumb switch and your capability to override and temporarily stop the runaway every time that it starts again. Anything is better than crashing.

        - Now, before you insult me, go a bit up and re-read the acknowledgement before these bullets.

        And yet, I still hope that something like this happened. Maybe not exactly like this, but something different than in Lion Air, something where the lessons learned from there would not have sufficed to ave the day. I would not like the pilots to have fallen in a trap from which the escape was readily available and should have been very well known after the Lion Air crash.

        --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
        --- Defend what you say with arguments, not by imposing your credentials ---

        Comment


        • Originally posted by Schwartz View Post
          Actually, from what I understand, the pilot can grab the spinning wheel and stop it, which means that the electric motor actually has less force than the pilot's hand. Would the jack screw motor even be able to overcome the forces?
          This is exactly my doubt too. I believe that when you grab the wheels with your hand you are NOT exactly stopping the motors. The motors will still spin but some clutch will slip. But the question still remains. We would need to clearly understand how the trim system works.

          --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
          --- Defend what you say with arguments, not by imposing your credentials ---

          Comment


          • Originally posted by Evan View Post
            Jesus. So, another stealth factor in which a provisional recovery technique for manually-unattainable high-speed trim recovery in the NG is contradicted by MCAS in the Max... and no one thought of it?
            Not sure what part of that would be exclusive of the MCAS/MAX and not be the same in a simple continuous trim runaway in any previous 737 version, like in the deleted Mentour video.

            --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
            --- Defend what you say with arguments, not by imposing your credentials ---

            Comment


            • Originally posted by Gabriel
              In the case of the 737, there are 2 conditions: Flaps up and flaps down. In this case the condition would have been flaps up. For that we need to set a pitch of 4 degrees nose up and 75% of N1 on the engines. This will keep the plane flying within the envelope everywhere from sea level to the max cruise altitude, but in different conditions. 75% N1 makes much more thrust at low altitudes that at high altitudes, so at high altitudes the plane will likely slow down to a slower speed and establish in a smooth descent, while at low altitudes it will climb and, more importantly, increase the speed.
              Why do they not use procedures based on altitude? The A320 procedure varies by being above or below 10,000ft. Also, the procedure varies again below thrust reduction altitude (5deg and TOGA). If the disagree occurs at rotation, this is the memory item I would expect an A320 pilot to use.

              With UAS, I would expect the initial procedure to be: level-off/establish a constant safe speed/do not reconfigure.

              Comment


              • Originally posted by Gabriel View Post
                Not sure what part of that would be exclusive of the MCAS/MAX and not be the same in a simple continuous trim runaway in any previous 737 version, like in the deleted Mentour video.
                The stealth factor would be exclusive. If pilots on the NG cannot move the trim wheel manually, they might momentarily re-engage the trim motor to trim up. On the MAX, doing so would result in an unexpected nose-down MCAS trim activation, thus doing so would be fatal.

                There is a 737 sim session of an NG trim runaway that I think you posted here. The PNF goes to manual trim but can't move the trim wheel. The PF then goes to assist him and, with both of them on the trim wheel, move it fairly easily. The manual trim system uses a spool that is supposed to provide mechanical advantage needed to move the stabilizer within the expected envelope.

                But, if the situation is as described, where the trim wheel cannot move the jackscrew in certain out-of-trim recovery conditions within the envelope, do we need to reconsidered the NG certification as well?

                Comment


                • Pilots followed Boeing's procedures but could not stop the plane from pushing nose down: https://www.bbc.co.uk/news/business-47812225

                  Comment


                  • Originally posted by Evan View Post
                    ...A better solution might be to focus more on training pilots to deal with their own human limitations. ... .

                    -
                    ... They were almost certainly confused however, and quite understandably. ...

                    ...result from a weak understanding of the automation (as was the case here).
                    seems like you were wrong.
                    Pilots "repeatedly" tried recommended procedures to prevent the crash of flight ET302, a preliminary report finds.

                    Comment


                    • Considerations and 'real world' heavy pilots

                      Originally posted by Gabriel View Post
                      Strange tings going on. I can confirm that the video WAS available. I saw most of it. And the link came from the Leeham news article. The article said that they had been working with Mentour in the sim and that them made a video, and at the end of the article there was a link to the video. Now the article still says that they worked in the sim with Mentour but there is no mention to a video, the link is gone, and the video doesn't exist anymore in the Mentour channel. A guy called Scott, who I don't have the slightest idea who he is, repeatedly said in the comment that the video was removed at the request of Mentour's employer (Ryanair), who seems to have quite a few of MAX in orders.

                      Now I wish I had downloaded the video while I could. I am sure someone MUST have done it, but I could not find any references or re-posts.

                      So, I will try to make a sort of transcript of the Mentour's video, which is in fact a transcript of my memory of it, some 10 hours later. Words in brackets are my comments.

                      The video was titled something like "Pilots got it right, but it went wrong".

                      [Video starts as always with Mentour sitting in his couch]

                      This is a video that we prepared in collaboration with Leeham News because we both are upset with the bashing of the pilots. There is a lot of rant in the news and social media asking how it is possible that the pilots of Ethiopian fell in the trap after everybody, and especially 737-MAX pilots, knew what happened with Lion Air. After that accident, Boeing disclosed the MCAS and published a service bulletin instructing to treat a MCAS malfunction as a trim runway and follow the trim runaway procedures. So why didn't they follow the trim runaway procedure and just use the cutout switches to disconnect the trim? It is a fair question.
                      This video shows a possible scenario of how the pilots might have followed the trim runaway procedure and still lose it. We still don't know what happened, but if in the next days the investigators report that the cutout switch was found turned off, remember this video.

                      You first need to understand some things about the 737 [he didn't say MAX]. The 737 has a lot of sensors outside the plane. It has a pitot tube on the left side and another one on the right side. The pitot tube is used to measure how strong the air is hitting it. There are also static ports on the left and right sides, that measure the atmospheric pressure outside of the airplane. And there are also angle-of-attack vanes, one in each side, the measures the angle that the air hits the airplane and the wings. These sensors are used to display several parameters in the flight instruments. The left sensors are used for the left instruments which are the Captain's, and the right-hand sensors are used for the right-side instruments, the first officer's.

                      Now, in older planes [older 737s included] the airspeed was calculated based only on the data from the pitot and static ports. In modern planes [including many 737's for sure the NG but possibly even earlier] the angle of attack is taken into account too because if the air hits the pitot at an angle it doesn't feel the same pressure [that's true, however the difference is just a small "fine tuning", it worked and still works pretty well just with the pitot and static]. So if one of the angle of attack vanes fails and sends incorrect information, the speed between the left and right airspeed indicators will differ [just a little bit] and we'll have a speed disagree alarm.

                      Since the Air France accident, a lot of focus ha been put in the importance of following the airspeed disagree procedures to keep the plane flying safely without knowing your speed [just look at the stby instrument, in this case the stby instrument must have matched one on the airspeed indicators, this is a speed disagree but should NOT be an unreliable airspeed situation]. In the case of the 737, there are 2 conditions: Flaps up and flaps down. In this case the condition would have been flaps up. For that we need to set a pitch of 4 degrees nose up and 75% of N1 on the engines. This will keep the plane flying within the envelope everywhere from sea level to the max cruise altitude, but in different conditions. 75% N1 makes much more thrust at low altitudes that at high altitudes, so at high altitudes the plane will likely slow down to a slower speed and establish in a smooth descent, while at low altitudes it will climb and, more importantly, increase the speed. You'll see why this is important.

                      So, airspeed disagree, the pilot sets 4 degrees pith and 75% N1 and the plane starts to accelerate. The trim wheel starts to move. This is not unusual, since the plane is accelerating, the trim moves to compensate for that [yes, that's the speed trim function, which I suppose would not work in a speed disagree situation]. But the pilot starts to feel it strange. The controls are heavy and he finds himself fighting to keep the nose up in 4 degrees, he trims up with the thumb switch, and it improves, but then the plane trims down again and it worsens again, so he trims up again and so on, and the plane is accelerating so in every cycle the controls feel heavier and heavier.

                      The focus of the crew starts to shift from the speed disagree to the control issue. They start to evaluate, are you trimming? Do you see what it's doing? And eventually they decide that they have a trim runaway, so they go through the memory items. Hold the control firmly, autopilot off but it is already off, autothrust off, evaluate, the uncommanded trim continues, so trim cutout switches, off. At this moment the trim stops actuating, but the plane can be quite out of trim already, we lost the thumb switches, and the speed in increasing. As speed increases, you need more and more force on the yoke to keep the nose from going down. You need to use the trim wheel to re-trim. But the trim also becomes harder. The Boeing 737 manual [he didn't say MAX] says that the trim wheel can become very heavy at high speed [I don't remember if he also said "and at high trim deflections", but it is at high trim deflections. If the plane is not grossly out of trim, the trim wheels can be turned manually to trim the plane at any speed]. It can reach the point where it is impossible for the pilot to keep pulling on the yoke hard enough, and to manually turn the trim wheel. So let's see how it looks and feels like in the cockpit.

                      [I had several reservations when Mentour said that last paragraph, reservations that were confirmed and I will expand with the sim part]

                      [Mentour is now in the left seat of a 737 SIM that they did not say it was a MAX and I am sure it was not, rather it was an NG, there is another pilot in the right seat, the wide-angle camera is in the middle behind both pilots, you can see the wind screen, the instrument panel, the center console, both pilots and the control columns. The size and resolution is not enough to read things like altitude and airspeed, but you can tell if the speed is increasing or decreasing, and you can also tell the approximate vertical speed by the position of the vertical speed needle, the attitude indicator is visible but with the attitude range during this flight it looked all the time almost level, it is daylight IMC and only clouds textures are visible out of the windshield throughout the simulation]

                      [Chime]
                      FO: Speed disagree
                      CAP: I agree. Speed disagree memory items please.
                      FO: Autopilot disengage
                      CAP: Disengaged
                      FO: Flight director, disengage (and he disengages it)
                      FO: Autothrust disengage
                      CAP: Disengaged
                      [the trim wheel starts moving forward and back, forward and back without pause, and continues until the trim cutout switches are turned off]
                      FO: 4 degrees nose up, 75% N1
                      CAP: 4 degrees, 75%
                      FO: You are descending
                      CAP: Yes, it's a but heavy to keep the nose up.
                      [FO looks at the trim wheel for some seconds]
                      FO: DO you have problems with the controls?
                      CAP: Well, it trims down, I can trim up but as soon as I release it it trims down again.
                      FO: It looks like a trim runway.
                      CAP: I agree, trim runaway memory items please.
                      FO: Hold controls firmly
                      CAP: Yes, already doing it.
                      FO: Autopilot disengage
                      CAP: It is disengaged
                      FO: Autothrust disengage
                      CAP: It is disengaged
                      FO: If the problem is not solved, trim cutout switches to cutout.
                      CAP: Let's evaluate a second... When I trim up the trim responds, but as soon as I release it it trims down again. And it is getting heavier

                      [1- Why don't you trim more nose up if it is getting heavier and you do have thumb switch authority? 2- This is NOT a realistic MCAS event, with the MCAS, when you use the trim switch in either direction the MCAS will STOP immediately its input and resume ONLY AFTER 5 seconds after you released the thumb trim switch. In this simulation, the trim wheel would spin back in response to the nose up thumb switch and IMMEDIATELY spin forward again. 5 seconds is not a lot but it gives you time to re-trim the plane, especially since the 5 seconds will not even start to count until you are done re-trimming the plane]

                      FO: So let's proceed with the cutout switches?
                      CAP: Proceed with the cutout switches.
                      FO: [switches the cutout switches off]
                      CAP: Now I am pulling buck pretty hard already, and it is getting heavier as the speed increases. How much trim do we have?
                      FO: Four units nose down.
                      CAP: Four units nose down? It could be much worse. So now, we would use the wheel trim to trim back up, but for the sake of the exercise, since it could have be worse, trim down please. Guys don't do this at home, it's only for the sake of the exercise, we should trim up, not down.

                      [I want to mention that 4 units may not be a lot but it didn't need to be that much either, since he had authority over the trim switch and that was overriding the trim runaway, as it should, he could have trimmed the plane to a more comfortable condition before using the trim cutout switches. Nothing would have prevented him for applying say 15 seconds of continuous nose-up trim and the trim would have responded and the MCAS or whatever was causing the runaway would not have interfered with that. In the Lion Air crash, during a few minutes the pilot was trimming up in opposition to the MCAS that would trim down each time 5 seconds after the last thumb switch intervention, managing to keep the trim on average more or less in the same position, and they could have done even better by trimming up a but more]

                      FO: [pulls the trim wheel crank lever out and starts cranking forward i.e. nose-down trim, for a couple of minutes]
                      CAP: It is getting harder and harder.
                      CAP: [at some point stops grabbing the yoke from the horns and instead passes his arms over the yoke interlacing both hands behind the column, as if he was hugging the yoke, to be able to make more force]
                      FO: [Keeps turning the trim wheel with more and more effort, the wheel is evidently getting harder and harder quite fast, due to a combination of increasing stabilizer deflection as result of the spinning and the airspeed increasing due to the 4 deg / 75% N1 / low altitude condition.]
                      CAP: I am using almost all my strength to hold it. And we still have a couple more points to go to reach the full nose-down trim.
                      FO: [Struggles to keep turning the wheel, and at a point he is unable to keep moving it]
                      CAP: Ok, start trimming up now.
                      FO: [Tries during several seconds but is unable, while Mentour fights harder and harder to keep the nose up. I cannot see the altitude or the airspeed, but the vertical speed is a bit down, perhaps some 1000 fpm down].
                      CAP: [looks at the camera or rather at a person next to the camera] Press the red button. Press the red button now! [not sure if he was referring to stopping the recording or stopping the simulator].

                      [The video returns to Mentour sitting in the couch, and there is where I stopped watching it. Because I had to leave. My plan was to finish it later, which failed]

                      There are a few reservations that I have. Many of them I already mentioned, others not yet, but I will summarize them below.

                      But first, I want to acknowledge that, despite my reservations, the situation described in the video is quite critical, that my reservations are done in hindsight and from the comfort and safety of my armchair, and that some of the things that I mention could have been done are not in the procedures. So keeping all that in mind:

                      - The speed disagree should have been a simple speed disagree and not an unreliable speed event. The difference is that in a simple speed disagree you still can tell which instrument is at fault. How? By looking at the standby instrument that MUST have matched one of the airspeed indicators. Why? Three scenarios: a) The standby airspeed indicator uses AoA data from the faulty AoA indicator, then it matches the airspeed indicator on that side. b) The standby airspeed indicator uses AoA data from the good AoA indicator, then it matches the airspeed indicator on that side. c) The standby airspeed indicator is "old style" and uses only pitot and static data, with no adjustment for AoA, then it matches the airspeed indicator on the "good AoA vane" side, because the AoA was actually low and that is where corrections are not needed.
                      The Lion Air crew of the flight before the accident quickly realized that which speed indicator was incorrect.

                      - Even disregarding the previous point, the degree of disagreement between the left and right airspeed indicators would have been quite small 10 or 20 knots, enough to call it a disagree, but not enough that you cannot take any of them as good and fly the plane.

                      - If the airspeed disagree was the result of an AoA indicator that was faulty since the beginning, the disagreement would have started at rotation. They had plenty of time to figure it out before the MCAS started to interfere and divert heir attention from the speed disagree issue. The MCAS would have started after flaps retraction.

                      - With the MCAS acting, there is nothing that prevents you from trimming nose-up as much as you want, and there is nothing that prevents you from stopping the MCAS immediately every time it starts acting. For both things you use the thumb switch.

                      - The MCAS would not start acting again immediately after you release the trim thumb switch as seen in the video, but 5 seconds later.

                      - There was no need to be so grossly out of trim by when you disconnect the trim cutout switches. Note that if you have the plane trimmed for positive lift, even if slightly positive and unable to balance the wight (let's say trimmed for 0.2 Gs at the present speed), that is a quite out-of-trim plane however, as speed increases, it will be easier, not harder, because that small but positive lift will be greater and greater. You need to have the plane trimmed for negative Gs to become worse as speed increases.

                      - A couple of minutes spinning the trim wheel nose down for the sake of the exercise?

                      - Note that in the Lion Air crash, the problem was not a hard-to-turn trim wheel or a harder elevator as peed increased. The main problem was that they were doing ok by trimming up with the thumb switch in response to the MCAS trimming down, until they stopped doing it, then they let the MCAS trim way down and as speed increased, yes, they could not pull back hard enough, but they still had thumb trim switch authority that they didn't use.

                      - Ok, so the captain cannot hold the column back anymore, the FO cannot turn the trim wheel, at what point does the captain says "stop trying and failing to move this wheel and helps me pull up"? The speed was obviously increasing way a lot. They had 2 airspeed indicators in agreement and another one almost in agreement, the controls are getting harder and I can't prevent the plane from going down? First help me pull up, we need to get away from the ground, then let's reduce the speed, throttles to idle, nose forget the 4 degrees, let's go for 15. When the speed goes down, the controls will be much lighter and the trim wheel too.

                      - A last resource could be to re-activate the trim cutout switches. The runaway will start again but so will your authority over the thumb switch and your capability to override and temporarily stop the runaway every time that it starts again. Anything is better than crashing.

                      - Now, before you insult me, go a bit up and re-read the acknowledgement before these bullets.

                      And yet, I still hope that something like this happened. Maybe not exactly like this, but something different than in Lion Air, something where the lessons learned from there would not have sufficed to ave the day. I would not like the pilots to have fallen in a trap from which the escape was readily available and should have been very well known after the Lion Air crash.
                      I am new here, however, have been following this forum for many years. This is a superb, considered post. There is so much considered and well thought out input from the many here. Evan, 3WE and many many more (forgive me if I forget your usernames)

                      I am sorry to say that the worst (imho) input comes from the armchair pilots of 'real' aircraft - ie - boeingbobby etc - (sorry bobby!) - Armchair - ? - yes, because that is the vast majority of time they spend in the air being fat dumb and happy. So rapid to blame the pilots, so rapid to scoff at 'low hours' so 'definite' in their biased opinion and 'I would sort that out without QRH etc etc' - what a terrible attitude and total lack of CRM and adaptation to modern systems. One could postulate that these are neanderthal comments from serial followers of a year or twos experience repeated MANY years without adaptation.

                      It would seem that the MCAS has far greater authority in movement and angle than the trim ability available to the pilots via trim switches. One could postulate a case where - as it seems to be - that the ability of a system of automatic trimming allows a far greater angle that the pilots can possibly make and at a rapid speed. In the event of a TR (WHEN diagnosed!) the recoverability due to the excess angle of the stabiliser is greater than the available trim to the pilots, and the lack of ability - due to limitations in pilot input trim, and load, can and apparently has, led to a non recoverable situation.
                      The 737 MAX is (IMHO) looks like a duck, quacks like a duck , but certainly cannot walk like a duck without falling over on it's beak without a hunters whistle.

                      It is a hard lesson to learn that the #'older'# so called, more 'experienced' heavy pilots posting here are the least knowledgeable and are so set in their ways as to mislead readers due to their self climbing on their own pedestal.

                      A huge thanks for the input from well thought out posters that do not have 12 months experience repeated for 15 years.

                      SA.

                      Comment


                      • After just a few minutes, with the plane still nose down, the Swedish 737 training pilot is exerting all his might to hold the control column, locking his upper arms around it. Meanwhile, on his right, the first officer tries vainly to turn the stabilizer wheel, barely able to budge it by the end.

                        If this had been a real flight, these two very competent 737 pilots would have been all but lost.

                        The Swedish pilot says at the start of the video that he’s posting it both as a cautionary safety alert but also to undercut the narrative among some pilots, especially Americans, that the Indonesian and Ethiopian flight crews must have been incompetent and couldn’t “just fly the airplane.”
                        There was a fairy, her name was 'Nuff' - Fairy Nuff.
                        especially Americans

                        Comment


                        • Originally posted by SmoothAir View Post
                          It would seem that the MCAS has far greater authority in movement and angle than the trim ability available to the pilots via trim switches. One could postulate a case where - as it seems to be - that the ability of a system of automatic trimming allows a far greater angle that the pilots can possibly make and at a rapid speed. In the event of a TR (WHEN diagnosed!) the recoverability due to the excess angle of the stabiliser is greater than the available trim to the pilots, and the lack of ability - due to limitations in pilot input trim, and load, can and apparently has, led to a non recoverable situation.
                          The 737 MAX is (IMHO) looks like a duck, quacks like a duck , but certainly cannot walk like a duck without falling over on it's beak without a hunters whistle.
                          I think it's fair to say that any flight control surface (stabs included) must remain controllable throughout their operational range throughout the certified operating envelope of the aircraft. This is still all very speculative and unclear to me, but, should it become apparent that the B737 does not meet this requirement in pitch trim, a redesign and mandatory retrofit is called for. That said, I suspect we are going to learn a lot more about this issue in the coming days and weeks.

                          Comment


                          • Originally posted by Evan View Post
                            I think it's fair to say that any flight control surface (stabs included) must remain controllable throughout their operational range throughout the certified operating envelope of the aircraft. This is still all very speculative and unclear to me, but, should it become apparent that the B737 does not meet this requirement in pitch trim, a redesign and mandatory retrofit is called for. That said, I suspect we are going to learn a lot more about this issue in the coming days and weeks.
                            Agreed, it appear to me that the MCAS has full authority to move the stab to the final limits, whereas, the pilot trim can only move to preset limits. WTF ?
                            This would 'appear' to be a situation where the pilot's ability to retaliate against mcas would be akin to arm wrestling with a guy with a larger arm, more strength and the ability to move more in up and down variations. Once the situation is 'diagnosed' the ability to recover is lost. Good night Vienna. When you hit the 'cut off' switches - the result is down to speed and angle of stab ! - both potentially MCAS related leading to a non recoverable situation when following the bible. The MCAS would appear to be, in its current form, the 'Jese Criz' bolt on the trike microlight. - are Boeing going to produce a guide that tells the pilots to count down in seconds when the appropriate time is to hit the 'kill' switches ? - WITH - never mind without ! - erroneous readings ?. This is NOT protective automation, this is literally a 'kill all' switch scenario. The MAX should be grounded until the engines are moved backwards AND the wings / gear are redesigned - never mind extended mounts and nose gear 8" lift ! - and then adding a bit of code to compensate for a now inherently unstable platform. It is a LEMON. The FAA and Boeing are going to suffer 'bigly' 'sic' will it ever pass EU regs ? doubtful.. The DH 'Comet' revisited. You have a total muppet in charge of the 'free world' (cough cough) - cannot even say origin - but says 'orange' like his skin.. he cannot even appoint a senior figure to the FAA - but, can give HUGE clearances in a nepotistic way. It does not bode well. !... the @FREE@ world - ps - that does NOT include 'trump land' hill billies - wants a solution that is viable - not political. It will be a great test. ! .. He has opened a Pandora's box....... Airbus ! - you are only going to reap dividends from this fiasco. - and rightly so. IT is NOT the way to run a railroad. - or a 'wall' or 'fence' or 'bit of wire' - sheesh.

                            Comment


                            • Originally posted by SmoothAir View Post
                              I am new here, however, have been following this forum for many years. This is a superb, considered post. There is so much considered and well thought out input from the many here. Evan, 3WE and many many more (forgive me if I forget your usernames)

                              I am sorry to say that the worst (imho) input comes from the armchair pilots of 'real' aircraft - ie - boeingbobby etc - (sorry bobby!) - Armchair - ? - yes, because that is the vast majority of time they spend in the air being fat dumb and happy. So rapid to blame the pilots, so rapid to scoff at 'low hours' so 'definite' in their biased opinion and 'I would sort that out without QRH etc etc' - what a terrible attitude and total lack of CRM and adaptation to modern systems. One could postulate that these are neanderthal comments from serial followers of a year or twos experience repeated MANY years without adaptation.

                              It would seem that the MCAS has far greater authority in movement and angle than the trim ability available to the pilots via trim switches. One could postulate a case where - as it seems to be - that the ability of a system of automatic trimming allows a far greater angle that the pilots can possibly make and at a rapid speed. In the event of a TR (WHEN diagnosed!) the recoverability due to the excess angle of the stabiliser is greater than the available trim to the pilots, and the lack of ability - due to limitations in pilot input trim, and load, can and apparently has, led to a non recoverable situation.
                              The 737 MAX is (IMHO) looks like a duck, quacks like a duck , but certainly cannot walk like a duck without falling over on it's beak without a hunters whistle.

                              It is a hard lesson to learn that the #'older'# so called, more 'experienced' heavy pilots posting here are the least knowledgeable and are so set in their ways as to mislead readers due to their self climbing on their own pedestal.

                              A huge thanks for the input from well thought out posters that do not have 12 months experience repeated for 15 years.

                              SA.
                              Rest assured that the heavy pilot in question has received a stern talking-to. Your input is much appreciated.

                              Comment


                              • Originally posted by Evan View Post
                                The stealth factor would be exclusive. If pilots on the NG cannot move the trim wheel manually, they might momentarily re-engage the trim motor to trim up. On the MAX, doing so would result in an unexpected nose-down MCAS trim activation, thus doing so would be fatal.
                                I disagree.

                                First, if you have a failure (not MCAS) that makes continuous nose-down trim inputs (except when using the thumb switch), you will have the same situation and even worse since it will not wait 5 seconds after your last thumb switch input. This can happen in any 737 and it is what happens in the now deleted Mentour video... in an NG sim, not a MAX.


                                Second, you are desperately needing to trim nose up, the cutout switch are off because you disconnected it because you had a nose-down trim runaway, you are fighting to turn the trim wheel and not being able to do it, so you decide to reconnect the cutout switches to be able to use the nose-up thumb switch, reconnect the cutout switches, and not use the thumb trim? Because if you do use the thumb switch then the trim will go up, not down, regardless of the MCAS or runaway situation.

                                There is a 737 sim session of an NG trim runaway that I think you posted here. The PNF goes to manual trim but can't move the trim wheel. The PF then goes to assist him and, with both of them on the trim wheel, move it fairly easily.
                                It was the previous Mentour video on trim runaway. And the PNF could move the trim, he just was annoyingly slow. It looked to me more like a matter of motor coordination than lack of force, when the PF "helps" the PNF it looked like the PNF was more a drag than a help, And then the PF keeps using the wheel by his own without effort (acknowledged, it would have been not so out of trim at the moment so the friction would have been less).

                                --- Judge what is said by the merits of what is said, not by the credentials of who said it. ---
                                --- Defend what you say with arguments, not by imposing your credentials ---

                                Comment

                                Working...
                                X