Tweet
Initial report just out. Will take some time to read and analyze...
-
--- 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 --- -
Juan Browne of blancolirio has posted a video walking through the initial report already. It is 24 minutes and he pretty much completely only talks about the report. The reports indicates about 9 seconds worth of nose down trim movement from 4.6 to 2.1 units. This is an exact specification of the MCAS, to move 2 units for about 9 seconds. It is its calling card. That is the initial sign of the problems. One interesting take from this too is he points out the 200 hour pilot made the right calls, which were agreed upon by the captain. But there are other data points that can not be deciphered if it was pilot input or MCAS input. But most of the data can be easily determined if it was human input or done by the planes systems.
Anyways, way too much for me to go over in this post, have a watch, it is very in depth and looks over the whole accident sequence.
Comment
-
I posted somewhere way back in this thread about a 707 (I think) crash that involved a dive with full nose down elevator and trim, where the trim could no longer move the stabilizer upward because the opposing force needed to overcome the aerodynamic force would have compromised the supporting airframe structure, and thus the jackscrew clutch was designed to slip as a safety feature rather than fail the airframe. Maybe there is a similar protection here as well.Originally posted by Gabriel View PostComment
-
Quickly scanned it. Two things caught my eye...Originally posted by Gabriel View Post
- Thrust is near 100% N1 and remains there despite sustained overspeed warning (tunnelling?)
- The Left AoA value suddenly returns to normal, then fluctuates. This seems to correlate with the transition from climb to final steep descent/dive and the vertical acceleration plot. Was the AoA vane moving by acceleration force instead of aerodynamic force?
It appears to me that the fatal error here was leaving the thrust at TOGA, despite the overspeed warning. The a/c, now exceeding VMO, beyond its operating speed envelope, could no longer be trimmed manually due to aerodynamic force.Comment
-
and as is his wont, evan finds a way to blame the pilots. the fatal error here was boeing's bloodthirst for money and desire to not cede market share to airbus who simply beat them to the punch.Originally posted by Evan View Post
FUCK!
a whole shit-ton of boeing people need to spend the rest of their lives in prison...nay, in 4th world hard-labor camps.Comment
-
It is unusual (in my experience) that the Table of Contents highlights only paragraph 1.6.2 (Flight Control System) in all capital letters.Originally posted by Gabriel View PostComment
-
Yeah, I have heard several professionals (not just the link I posted) say the pilots acted accordingly and that the MCAS error/malfunction is a hard beast to tame for even the most experienced. You see where I live, not Seattle itself but Boeing brings their planes to my city on tests sometimes. The 787 flew right over my house when it was being tested many times. 737's fly over before they are delivered on pre delivery test flights all the time. One of my favorite vacations is to the Museum of Flight in Seattle. But right now I am not looking to Boeing as a badge of excellence or a point to be proud as far as patriotism for my state goes. I always was prideful they were the leading force in aviation by most accounts. Cue the 787 problems. Hey they were not too bad, hey they fixed it, just a little bump in the road (or turbulence since this is aviation speak). But this is not something I am going to get over. Not just me, far from just me. Sadly this could be bad for the economy of Seattle, and it will trickle down here to Spokane and Moses Lake too. Moses Lake was being eyed for the 797 production. Will they be able to go forward with the 797? Will airlines want to even buy it? Lord knows the brand name is now associated with this fiasco and the 787 isn't a clean slate either.Originally posted by TeeVee View Post
I know it sounds funny but I feel betrayed by Boeing even. You can laugh and say thats ridiculous. Go ahead. They have betrayed the common bounds of safety and as such have betrayed the trust the common flyer should put into their newest planes when booking a flight on them, or the pilot that has to figure out how the hell to safely land because another error has happened praying for their life. I don't care if I exaggerate, Boeing didn't exaggerate their safety measures enough and lives were lost, thats the true shameful exaggeration.Comment
-
After the stab trim is disabled, the aircraft still pitches down intermittently. The expert in this video finds this strange. The expert does not understand why the stab trim has been reactivated causing the final and fatal pitch down.Originally posted by KGEG View PostComment
-
My conclusions after reading and analyzing the preliminary report. I am mixing facts (or things that I consider facts) and opinion. I am putting opinions in parenthesis:
- The MCAS actuates as designed (as horrible as designed).
- The trim switches actuated as designed.
- The electric trim motor was able to move the stabilizer in both directions even when the trim was already way off and the speed was way high, overspeed in fact, and the manual trim wheel was so hard that it was impossible to turn manually (this basically kills my concern that in extreme airspeed and out-of-trim conditions the trim may become so hard that even the electric motor would not be able to turn the jackscrew).
- The throttles were left in the take-off position since the start of the take off roll and were never moved afterwards.
- The left AoA sensor had an instantaneous failure a couple of seconds after lift-off that made it go instantly from a normal AoA indication, identical to that of the right AoA sensor, to the ridiculous value of 74 degrees and remain totally fixed there for the remainder of the flight except some fluctuations during the final dive. (The reason for the sudden trip of the left AoA indication may have been a bird strike that moved and bent the vane in a way that got stuck in that ridiculous position or an electronic issue, for the fluctuations in the final dive I don't have any hypothesis).
- The huge AoA disagree generated a small speed disagree (which remained small throughout the flight) and an initially small altitude disagree which however became very big as the airspeed increased throughout the flight. By comparing with the radio alt, it is evident that the right altitude was correct and the left altitude was wrong (I don't know why the altitude difference becomes so high, old altimeters worked quite well without compensating for AoA. On the other hand, I see how, at 300+ knots, 74 degrees of AoA, if they had been real, can create huge static pressure errors. I don't understand how Boeing doesn't have simple monitoring algorithms to rule out obviously erroneous values. 74 degrees of AoA at 250+ knots is simply impossible. There would be no more plane and no more AoA by then, the plane would have been shredded to pieces by aerodynamic forces).
- (It is almost certain that the standby speed and altitude would have matched those of the right side instrument, allowing the pilots to quickly recognize which instruments to trust)
- That erroneous left AoA immediately generated a false stick-shaker on the left side whih remains until the final dive. There is no mention in the report about any reaction or verbal communication of either pilot regarding the stickshaker.
- With the plane reaching 400 ft AGL, they try to activate the autopilot twice, but the autopilot would not engage and the autopilot disengage aural warning would sound in both occasions. (Activating the AP at 400 ft is standard procedure but I don't think it should be done if you have the stickshaker shaking, not at least until you have deemed the stickshaker false, and then activate the autopilot on the right side. The reason why the autopilot would not engage was likely that they attempted to engage the left-side autopilot and it would not engage with an AoA indication of 74 degrees. If they had activated the right autopilot it would probably have worked and the MCAS would have never activated since it only works in manual flight. During the approach, if they had at least some flaps extended by when they disconnected the autopilot, the MCAS would not have activated either since it activates only with flaps retracted, and if the flaps had not been extended yet the MCAS would have likely activated but it would have been easier to control since the speed would have been slower and the thrust probably about idle).
- In the next seconds they make a few nose-up and nose-down manual trim inputs with the thumb switch and the trim responds accordingly.
- (The plane seems to be climbing slowly than normal and accelerating faster than normal. They initially rotated to about 18 degrees, which is a normal initial climb attitude, but after the stickshaker activation the pitch is lowered and held at 8 degrees for several seconds and then lowered again to 4 degrees. Why? Are they following an unreliable airspeed procedure? Are they concerned with the stall due a stickshaker activation? In doesn't look like wither of that since they attempted to engage the AP twice and they are about to do it a 3rd time).
- Then the try and manage to connect the AP. The left AP, according to the report, which I find surprising since the left AoA is still 74 degrees and the left stickshaker is still shaking.
- The AP makes several trim inputs, completely normal for the AP, and the trim reacts accordingly.
- About 20 seconds after engaging the autopilot, they move the flaps handle to the full retracted position. It would take the flaps some 20 seconds to fully retract.
- About 30 seconds after egging the autopilot and 10 seconds after moving the flaps lever to the fully retracted position, with the flaps still retracting, the AP disengages. There is no mention whether it disengaged by itself or if it was disengaged by the pilots.
- The flaps reach the fully retracted position. So far, between the pilot and the automatic trim inputs, the trim has moved all the time very close around the 5-units mark. Basically neither the pilot inputs nor the automatic inputs have moved the trim significantly in either direction so far. But this is about to change.
- As soon as the flaps fully retract, there is a long automatic nose-down trim input. There is no thumb switch intervention during this automatic activation, the trim moves about 2.5 units nose-down from about 4.5 units to about 2 units (full trim down is 0 units). That must have been the MCAS, for the 1st time in the flight.
- A few seconds after the MCAS stops by itself, the pilots apply briefly nose-up thumb switch bringing the trim about 0.5 units up to the 2.5 units position. (Why they stopped there, I don't know. The plane was evidently still quite out of trim, the MCAS actuation had been much longer than the opposite thumb trim actuation)
- 5 seconds after they pilots finish their nose-up thumb switch input, there is another long (albeit slightly shorter) automatic nose-down input. The trim goes down some 2 units nose-down to about 0.5 units, almost the full nose-down position.
- The pilots apply nose-up thumb switch (and that seems to cut the MCAS motion a bit shorter than the previous full time actuation). This time they keep the thumb switch input for about the same length of time that the MCAS had been actuating and the stabilizer returns to about the 2.5 units position. In the mean time, the airspeed increased to about 320 knots (overspeed is 330 knots).
- (The airplane must have felt terribly out of trim with 2.5 units and 320 knots, again, why did they stop trimming up?)
- The FO calls trim cutout, the captain agrees and the FO calls trim cutout (surely confirming the action, there is no recording of the actual trim cutout switch position but the next event confirms that they had been moved to cutout).
- 5 seconds after the pilot finished the last nose-up trim thumb switch command, there is a third automatic nose-down command that lasts about the same than the 1st one. However, there is no trim motion related with this command. That must have been the MCAS trying to apply nose-down trim again but the trim not responding due to the cutout switches having been moved to the cutout position.
- The plane starts to climb more rapidly and the speed stabilizes at about the overspeed onset.
- During the next 2 minutes and a half, the recorded position of the trim increases it's nose-down position slightly by just 0.2 units. The report contains no explanation. (I believe it might be due to play and elasticity of the mechanism and structure, with the force on the stabilizer and resisted by the jack screw increasing as the airspeed increases).
- After the trim cutout calls, there are several conversations in the cockpit about the captain asking for help to pull up (confirmed by records of simultaneous nose-up force in both control columns mentioned in the report but not shown in the FDR graphs), the captain asking the FO to tell this or that to the ATC, and the FO making those ATC communications. For a while, there is no mention of trim for several seconds and no indication that there was any attempt to use the trim wheel.
- About 1 minute and 10 seconds after the cutout calls, the captain asks the FO if the trim is functional and the FO replays that it isn't (this may be referring the the thumb switch), then the FO offers to try manually (this may be referring to the trim wheel) and the captain agrees. A few seconds later the FO says that the manual trim is not working either (possible meaning that he was unable to turn the trim wheel because it was too hard). The speed by then had increased to about 360 knots.
- For more than 1 minute there are several conversations in the cockpit, all if them unrelated with the trim. Again, talks about help to pull up, conversations with ATC, changes in the autopilot settings, a master caution that sounded in relation to the left AoA vane anti ice...
- About 1 minute and 20 seconds after the cutout calls, the pilot once again asks help to pull up and says that the pitch is not enough.
- While there is no conversation or parameter recorded, at that moment someone must have re-engaged the trim cutout switches beck to the operative position because immediately after...
- There are 2 very briefs nose-up trim thumb switch clicks and the trim DOES RESPOND by moving up 0.2 units.
- (If they reconnected the the trim cutout switches o be able to apply nose up trim, why on Earth they did just 2 super-short clicks????)
- (Sure enough), 5 seconds after the second click we have a long automatic nose-down trim input (hello MCAS again) that moves the trim to almost the full nose-down position. There was no recorded attempt to interrupt the MCAS with the thumb switch or to trim back up afterwards. (Why why why!!!!???)
- The airplane had achieved 7000 ft over the ground (i.e. they were not so low), but with a minus 3G, minus 40 degrees pitch dive at 400 knots, 20 seconds later they were a hole in the ground.
My conclusions:
- I will let Evan and TeeVee discharge his rage on Boeing's design, and I will agree. I want to focus on the pilot's actions GIVEN the airplane design and what was known form the Ethiopian crash.
- The MCAS and trim system worked as bad as it is designed to but not worse than that. For example, these bad things speculated in the forum did NOT happen:
--- MCAS acting in conditions other than manual flight with flaps retracted.
--- MCAS acting when the thumb switch was actuated.
--- MCAS not stopping when the thumb switch was actuated.
--- MCAS acting with the trim cutout switches moved to the cutout position.
--- The electric motor not being able to move the stabilizer in extreme-force conditions.
--- Limitations on the authority of range of the trim when using the thumb switches.
--- The stabilizer failing moving in response to the trim thumb switch (with the cutout switch in the operational position).
- For unknown reasons, the pilots did not re-trim the plane before resorting to the cutout switches.
- For unknown reasons, for more than 1 minute, when the speed was already high but not so high, the pilots did not seem to attempt to operate the trim wheel manually after disconnecting the cutout switch, even when the plane was severely out of trim. By when they attempted it, the plane was going to fast and the wheel was too hard to move manually.
- When the situation became desperate, the re-connected the cutout switches. The only logical explanation for that would be that they would attempt to use the thumb switch to re-trim the plane. Yet:
- They did not apply nose-up trim except for 2 supper-brief clicks on the thumb switch. The stabilized did respond accordingly moving an insignificant little bit nose-up.
- When the MCAS started to trim down again, they did not attempt to use the thumb switch to stop it. The nose-down trim automatic actuation would have been super evident not only by the turning of the trim wheels but also by the increased force on the control column (with which they amazingly managed to move the control column further back) and the yet uncontrollable quick nose-down pitch.
- With both pilots pulling hard on the control column together and being unable to control the pitch-down, they didn't attempt to use the thumb switch to trim up.
- Any of the previous 6 bullets would have saved the flight and the lives. (Also a sound design would have, but I leave that to Evan and TeeVee)
- All of the previous 6 bullets were known from the Lion Air crash
(the part that was not known was that it could be impossible to turn the trim wheel manually in extreme conditions)
--- 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
-
After the stab trim is deactivated (i.e. the trim cutout switches are put in the cutout position) the stabilizer doesn't move any more except, the report mentions, it moved barely 0.2 units nose-down in 2 minutes and 30 seconds. No mention of intermitency. I suspect this can be reduction of free play and flexibility in the mechanism and structure as the speed kept increasing and hence the aerodynamic forces on the stabilizer kept increasing.Originally posted by birdguts View Post
I think it is quite clear. They were desperately trying to keep the nose from going down. The captain asks for help from the FO to pull up and the FO helps but the captain says that the pitch is not enough. They desperately needed to trim up and they had already tried the trim wheel manually only to find it impossible to move. They needed to revert to the trim switch. For me the real unanswered question is, why didn't they revert to the trim switch then, after re-engaging the trim cutout switches? They payed for that mistake with their lives.The expert does not understand why the stab trim has been reactivated causing the final and fatal pitch down.
--- 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
-
From a comment in the AvHerald article:
I am not in that position. I am more in the middle. The design was horribly flawed. The possibility of a bird strike disabling a single AoA vane is a clear possibility, and that triggering not only airspeed disagree, altitude disagree and stikshaker on the affected side but also large uncommanded (although stoppable and reversible) trim inputs is just crazy.New procedure:
By (anonymous) on Thursday, Apr 4th 2019 21:43Z
Stick shaker active ———> Autopilot engage
EGPWS master caution —-> Change configuration immediately
Flight control problem ——-> Contact ATC immediately
Stabilizer trim runaway ——> Re-engage cutout switches ASAP
Overspeed warning. ———>. Keep T/O thrust
Facts from Lion Air. ———>. Ignore, no need to inform yourself
Just blame EVERY crash to the airplane, don‘t ever,ever question the actions or performance of the flight crew. Never!!
Btw, Beirut Crash of ET 737-800 is also fun to read.
Oh i am sorry, EACC corrected the official report...it was a bomb LOL
But I always want to think, in every situation and regardless how bad the design or how bad the circumstances (including previous pilot errors that led to the delicate condition), what could the pilots have done THEN to stop the chain of events and save the lives. I asked myself this question in a lot of accidents, and in most of them the answer was... they could (note I am not saying should) have been done much. And this accident is no exception.
Regarding the last sarcastic comment about the ET accident in Beirut, this is the context (from Wikipedia):
The final investigation report released by the Lebanese Ministry of Public Works and Transport, presented on 17 January 2012, stated that "the probable causes of the accident were the flight crew's mismanagement of the aircraft's speed, altitude, headings and attitude through inconsistent flight control inputs resulting in a loss of control and their failure to abide by CRM [Crew Resource Management] principles of mutual support and calling deviations."
Ethiopian Airlines stated the airline "strongly refutes" the report, and that it "was biased, lacking evidence, incomplete and did not present the full account of the accident". The airline wrote in a press release issued the same day as the investigation report that the halting of flight data and cockpit voice recording at 1,300 feet, the disappearance of the aircraft from radar at that time, and the eyewitness reports of a fireball "clearly indicate that the aircraft disintegrated in the air due to explosion, which could have been caused by a shoot-down, sabotage, or lightning strike."
--- 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
-
From the AvHerald just-updated article:
I hate intentional ambiguity, lack of openness, and obscurantism.On Apr 5th 2019 Boeing reported a second "relatively minor" software problem unrelated to MCAS was identified and is being fixed. This fix delays the submission of the MCAS fix to the FAA, that was intended to take place until end of March and is now going to happen in the coming weeks.
The software problem affects flaps and other related flight control hardware. Boeing informed the FAA about the issue, the FAA rated the problem critical to flight safety and have ordered to fix that problem.
--- 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
-
Is it possible they were afraid to reduce the thrust because of their struggle to keep the nose up? There is a pitch-up effect when you set high thrust, right?Originally posted by Evan View PostComment
-
In nearly 100% of ALL accidents of EVERY type, someone will easily work out the steps that would have avoided the accident to begin with. When I was cutoff by a taxi in 2012 while on my enduro bike, had I leaned hard, turned and jumped the curb before attempting to brake instead of attempting to brake only to find that my knobby tires didn’t offer enough friction on asphalt to stop in time, I wouldn’t have 21 screws and 3 plates in my ankle. So I guess I am to blame. Never mind the taxi who changed lanes illegally and then stopped to block a lane of traffic illegally.
It’s easy to armchair these things after the fact, when one has all the time in the universe to think about what was happening and what should be done. It’s easy to say, memory items, qrh, crh and all the other acronyms you care to.
I watched Juan brown’s last video, and strangely absent was any placement of blame on pilot action or inaction. Maybe he realizes that humanity of every pilot—something evan seems incapable of...
But if you’re gonna armchair, preface your genius with the absolute truth: Boeing, in a GROSSLY negligent manner, place these (and others) pilots in a position they should NEVER have been in. All for the sake of $$$ and making shareholders happy. Boeing, and NO ONE ELSE, well maybe some turds at the faa, are 100% the cause of the 364 deaths, as well as the enormous loss of revenue airlines around the world have been and will continue to endure.Comment
-
In the days since the Ethiopian crash, and after its initial fall on March 11, the Boeing share price has been fairly resilient, trading narrowly around $370/380. What do we conclude from that ?Comment
Comment