|
safeteepee #5039
Do not disagree with anything in your post. But do you think that it is possible to use electric trim to counteract MCAS before using the cut-out switches? I feel there is something being missed here. Clearly, it is intended by Boeing that you can, and that is the information that has been provided. Given, two crews appear to have failed to apply sufficient nose-up trim, I have a niggling doubt it is the whole story. I want a reason they failed to apply sufficient nose-up other than they were stupid. I realise there may not be a reason... |
Rogue Boeing 737 Max planes ‘with minds of their own’ | 60 Minutes Australia
Compelling interviews with Chris Brady ( The Boeing 737 Technical Site ), Dennis Tajer, Peter Lemme, David Learmount and Dominic Gates. 43 minutes... |
Threat and Error Management
PART 1
In the interest in moving the conversation along, and in particular to help transition this thread from a rearview, reactive mode to a more forward-looking proactive one, I would like to tee up a subject that is hopefully familiar to many aviators - the Threat and Error Management Model. I have some more thoughts to share that build upon this model, so it would be helpful to have some familiarity with it. Briefly, the TEM model examines aviation safety through a lens that assumes that pilots will always be faced with threats (known and unknown) and errors. It is assumes that there are no perfect aircraft, environments or people, but it tries to devise a resilient system that is capable of identifying and mitigating threats and trapping errors. Rather than cut and paste a full description of the TEM model, I would ask you to look at the links below: Introduction to Threat and Error Management Wikipedia Threat and Error Management Part of the TEM model is the concept of barriers. Barriers are those things that can be put in place to mitigate threats and trap errors. Some people refer this to the "Swiss Cheese" model because it assumes that no barrier is perfect either. Even though there will be holes in each barrier, the concept is to have enough of the right type of barriers so the "holes" do not line up and lead to either an undesired aircraft state or worse, an incident or accident. . https://cimg6.ibsrv.net/gimg/pprune....554adf94e.jpeg The MAX accidents can be analyzed using the TEM model to identify not only the particular threats and errors, but also whether there were sufficient barriers and/or why the existing barriers did not ultimately prevent these accidents. Armed with this information, then the goal is to identify how those barriers can be improved to prevent future incidents and accidents. |
Originally Posted by Takwis
(Post 10464762)
What does TBC-19, the Boeing Flight Crew Operations Manual Bulletin, say you should do when there is an AOA disagreement?
Not that I agree, but they make no mention of an unreliable airspeed checklist...they jump right into fixing the trim problem. A careful reading of TBC-19 basically states that a bad AOA input can create unwanted stab trim movement, and the proper response to the unwanted stab movement is to execute the Runaway Stabilizer Trim checklist. I might add that functionally while experiencing this malfunction in the air, we do not refer to this document (or its equivalent). We refer to the appropriate non-normal checklist. |
Originally Posted by wheelsright
(Post 10464773)
safeteepee #5039
Do not disagree with anything in your post. But do you think that it is possible to use electric trim to counteract MCAS before using the cut-out switches? I feel there is something being missed here. Clearly, it is intended by Boeing that you can, and that is the information that has been provided. Given, two crews appear to have failed to apply sufficient nose-up trim, I have a niggling doubt it is the whole story. I want a reason they failed to apply sufficient nose-up other than they were stupid. I realise there may not be a reason... Other than 'deer in headlights' loosing it I see a few possible factors: 1: Pilot accustomed to short blips not comprehending amount truly needed, this fits Lion Air when the FO was ineffective at the end when PIC handed over control while the Captain was mostly successful. 2: Trimming by column feel not position: May seem that AC is closer to trimmed than it is. In other words if you have been pulling really hard then just slightly hard may feel like close to trimmed, I am sure the pilot did not want to over trim given all the alarms. This might explain ET first retrim at 05:4-:15. Unfortunately we don't have the column force graph, just position. 3: Some as yet to be revealed flaw that interferes with pilot trim inputs; one possibility is biomechanical factors related to actuation switches after prolonged pulling. This is unlikely but could explain the final seconds of both accidents. Hopefully the final reports will fully address this question. |
737 Driver, TEM incorporates the whole system, not just pilots being faced with issues. So, in this case one could view MCAS trim runaway as the threat with AOA failures, inadequate training, crew responses etc as errors. Without going over the apportionment argument again (that’s one for the lawyers), MCAS brought a new TEM ‘flow’ that was not properly thought through. The slightly worrying thing is a similar flow could apply to any trim runaway; in hindsight the number of layers in place may be too few or, more likely, some of the layers have rather large holes. TEM works, as long as everyone involved recognises the importance their role in the system. Some thousands of posts back I postulated that the overall risk may have been (inadvertently) ‘shuffled’ too far in the direction of the crews. |
Originally Posted by MurphyWasRight
(Post 10464802)
3: Some as yet to be revealed flaw that interferes with pilot trim inputs; one possibility is biomechanical factors related to actuation switches after prolonged pulling. This is unlikely but could explain the final seconds of both accidents. |
Originally Posted by Cows getting bigger
(Post 10464808)
737 Driver, TEM incorporates the whole system, not just pilots being faced with issues. So, in this case one could view MCAS trim runaway as the threat with AOA failures, inadequate training, crew responses etc as errors. |
driver,
the flow of if/then logic in the system is lost with legacy and lack of a comprehensive logic flow chart. As an example, the accident in DXB, when the pilots pressed TOGA, but then the ac bounced on the runway. The pilots had no idea of the logic switch in the system, and what was disabled with weight on wheels. The mantra, press TOGA, pull back...within 10 seconds the ac impacted the rwy. The legacy commands, the if/thens, are lost in the FMS programming. Only on incidents, does the legacy and the myriad of if/then scenarios emerge. MCAS was a poorly applied band-aid to a much larger problem. At least 3 different crews found the holes in the model, 2 of which, cant speak to the issues. |
What, exactly, is the use of an "A of A DISAGREE" with no A of A indication to back it up? I can't see it tells you anything useful at all, in fact it merely adds an unquantifiable and unanswerable question into the mix. Had it been fitted it would have required a QRH action to go with it. What could tgat possibly tell you to do about it?
Add another A of A sensor and a triage system and it becomes another matter of course, but it seems unnecessarily harsh to criticise Boeing for not incorporating procedures that can only confuse and not help. |
Originally Posted by wonkazoo
(Post 10464833)
The Boeing whistleblower who is working with the FBI confirmed that making the MCAS flight control system reliant on only one sensor was a deliberate decision to avoid the need for expensive Level D or flight simulator training.
“MCAS was designed using data from only one of the sensors because we knew the FAA would not have certified a two-sensor system without Level D Training…” https://www.youtube.com/watch?v=Qytf...ature=youtu.be at 35:30 Boeing is dead, at least the Boeing we all knew and loved. It’s sadly time to accept that reality. |
Originally Posted by PJ2
(Post 10464861)
Was this video done in a B737MAX simulator? If so, who's simulator, and if not, we need to know how an NG simulator re-produces MCAS behaviour such that the demonstration is a valid one.
|
Originally Posted by PJ2
(Post 10464861)
Was this video done in a B737MAX simulator? If so, who's simulator, and if not, we need to know how an NG simulator re-produces MCAS behaviour such that the demonstration is a valid one.
Forget the demonstration- that was so much theater. Look at what the Boeing employee said- that is the important bit, which is why I posted it. Regards, dce |
Originally Posted by wonkazoo
(Post 10456057)
Reading the back and forth on the conundrum of “Pilots completely at fault for being stupid” as set against “Pilots completely not at fault because Boeing designed a $hit system,” and I’ve decided to take a stand. Apologies in advance as this is long, but trust me- it is worth it.
Of all of you out there who have posted here- whether with a desire to blame the pilots entirely or the opposite, how many of you have actually experienced an imminent, binary and life-threatening emergency in your airplane?? One that is so explicit you will either get it right or you will die?? And you have maybe 30 seconds to make that choice. My guess is the list of aviators who can answer honestly that they have been at that threshold of death is very very small. I am on that list. And I survived, despite making bad choices along the way. I share this story because I want people to understand once and for all that while it is fine to offer that the pilots could have done better (they totally could have) the root cause of the MAX crashes was one of design, and human failures only built on that edifice to achieve the final outcome. I also share this story because I want to try to explain to everyone here, in terms we can all understand, what it is really like when your known world explodes and you have to improvise in order to survive. In June of 1996 I was in a very high performance unlimited category biplane named the Goshawk. (N345RM) I had departed Livermore CA several minutes earlier and was headed to a legal practice box adjacent to the Tracy airport. While over the Altamont hills at an indicated altitude of 4000MSL I began warming up by pulling to a 45 degree upline and doing snap rolls to the right. I did this once or twice. On the third attempt, once again at approximately 4000+MSL I initiated the snap roll to the right and hit hard left rudder as the wings returned to level to stop the autorotation. When I did this the left rudder pedal/bar shot away from my foot instead of providing actual resistance. The left rudder cable had snapped. The airplane (which was by design dynamically unstable) paused its rotation for a moment and then began again violently to the right, probably at about 360 degrees per second. And here’s where the chair-jockeys don’t get it. I probably went two or three full revolutions before my mind could accept what I already knew had happened. I immediately pulled power, but the aircraft was already entering a nose-down spin- at a rotational rate of at least 360 degrees per second. The ROD of a spinning aerobatic biplane is pretty steep, probably on the 1500-2000FPM range. I checked my altimeter, saw I was descending through 4,000 feet and decided to try to recover the airplane before bailing out. AND HERE IS THE IMPORTANT POINT: BECAUSE I REFUSED TO ACCEPT THAT THE AIRPLANE WAS COMPLETELY EFFED I would nearly die. My mind knew before then, as it knows now, that if you put a Pitts-like airplane into an autorotational state the only thing that is going to get it out is opposite yaw. With no rudder THERE CANNOT BE ANY OPPOSITE YAW!! I had thousands of hours in similar aircraft, I was an unlimited category competition aerobatic pilot and instructor, and yet when faced with the obvious I could not process it quickly enough, despite having the evidence staring me squarely in the eye, to react quickly enough to prevent me from nearly dying. So I frittered away precious moments trying to use opposite yaw via ailerons, shots of engine thrust, hell I might have even prayed, I don’t know. What I know now is I could have done better. What I also know now, and somehow managed to forget then, was that I was over the Altamont. When I saw 4000’MSL and thought “OK, I’ve got time to play with this” the reality was I was over a hill- that was 2134’ high. Tracy- just 20 miles away and where I was headed sits at 193’ MSL. In my mind, because I was stupid overwhelmed, or just unable to process everything being thrown at me I had maybe 3500-4000’ to play with. So I could spend 30 seconds fighting the airplane to try to recover it before I had to bail. In reality I had less than 2000’ before I would be dead. I spent probably ˝ to ľ of the real time I had to get out of the airplane in it- fighting to try to save it, and I did this by deliberately ignoring what I already knew (I had lost rudder control completely) what I should have known (I was over the Altamont) and what I should have accepted (I had to go- the Goshawk was not going to survive this, the only real question was would I??) I obviously did reach the (already foregone but stubbornly ignored by me) conclusion that the airplane was unrecoverable and decided to bail out- which is an interesting concept in a stable spinning airplane. I undid my harnesses as I had practiced, and I fought my way out of the airplane- pinned against the left side of the cockpit coaming by the rotational g-forces before eventually getting enough of my upper body into the slipstream that I was basically yanked out of the airplane. I was falling in a fetal position, thought about waiting to pull the ripcord, said eff-it and pulled, and after the shocking introduction to my first and (so far) only canopy opening was struck by the sound of the airplane smacking into the ground just a second or two later. Future calculation efforts would show that my chute opened between 134 and 200 feet above the ground, which at that rate of descent equaled a couple of seconds at best. Surviving that incident has given me some small window of insight into what happens when your comfy world devolves in seconds into one where you know you are about to die. The biggest lesson, and the greatest ego-killer was simple: I didn’t respond nearly as I would have hoped I would. It took me countless seconds to register the fact of the failure. I knew as soon as the pedal fired away from my foot what had happened. But my mind simply refused to accept that reality for some short period of time. The second error was equally simple: I thought I was the hero pilot (Neil Williams etc…) who would bring my crippled plane back to the airport, thereby saving the day. That thought nearly cost me my life, as I wasted precious seconds performing an absolutely useless dance of fancy “airmanship” that did nothing but allow my airplane to bring me closer and closer to the ground with every moment. And now to the main point of this entirely too-long post: For those of you who suppose you will see everything clearly and “FTFA” when your own fatal opportunity presents itself please hear me when I say this: YOU WILL NOT!! The question that will determine your survival is how quickly will you move past that initial shock and be able to function properly again. In my case it was a single (albeit fatal) failure. I was extremely well trained, averse to panic-driven responses, and well-able to handle the emergency I had been presented with. Yet I wasted probably a full minute in an airplane I had no business being in any longer. In the 737 crashes it was a cascade of failures. My own- very rare life experience tells me that those pilots had little chance given the stressors they were working under, as would the rest of you. These are not the words of someone who doesn’t know what it’s like. I’ve been there. I lived. So please trust me when I tell you that your vaunted talents will wither to nothing if someday you are in this unfortunate position. At best you will be semi-functional, at worst you will be functionally useless. What you will not be, in any context, is a hero who defies these realities. Final note: This isn’t about placing blame on anyone. Boeing designed an airplane with a crap system that had random and unmonitored control over the single most important control surface of the aircraft. The FAA paved the way for certification of the airplane, and once in the hands of pilots that airplane not once, but twice flew itself into the ground. (The pilots didn’t- it was MCAS that did, and that’s an important fact to take note of…) You can blame the pilots all you want, but it was the airplane itself that had a failure mode that required the pilots to be perfect or die. Boeing had years to create a functioning system that would not put the pilots in this position and they failed to create one. So the two (six really) pilots were left to defend themselves against an airplane that was trying to kill them. Four failed in that endeavor, and they have my utmost respect and gratitude. Only those who have walked the path and survived can understand the fine line between winning and dying- which is why I have posted this ridiculously long post tonight. Sorry for the sermon, just tired of reading the constant back and forth about who we should blame. Link to the Final on my incident: Well despite being a member for years I haven't reached the vaunted 10-post threshold for posting URLs. Search "NTSB June 17, 1996 N345RM" for the final report. Regards, dce //M |
Threat and Error Management
Part 2
Continuing the Threat and Error Management (TEM) discussion.... Necessary preamble.... Yes, there were many other factors leading to this accident, many of which were outside the control of the crew. The TEM model can be applied to those pieces of the puzzle, but my interest is in what can be learned and applied at the professional flight crew level. What are the takeaways so that pilots can avoid a similar situation in the future? Using the TEM model to analyze the Ethiopian accident, we can look at the factors that were directly bearing on the crew. We can start by asking "What were the threats?" Threats can be external or internal (inside or outside the crew's direct control). External threats are things such as weather, terrain, language barriers, external pressure to meet schedule, etc). Internal threats could include such things as fatigue, distraction, and crew experience. Threats can also be known and unknown (but not necessarily unknowable). A known threat might be an inoperative aircraft system that has been placarded. Unknown threats which could still be anticipated would include such things as possible aircraft malfunctions during the flight, a pop-up TCAS alert, or a sudden call by tower to execute a go-around from low altitude. Once identified, the next step is to attempt to mitigate the threat using the tools available to the flight crew. For ET302, environmental threats might include field conditions (high altitude) and high terrain in the general area. Specific detail on the crew pairing and schedule has not been released, but we can probably categorized a low-time First Officer as a potential threat. If the crew had not flown together much, lack of familiarity with each other could have been a threat. Insufficient rest could have created a fatigue issue. It has been suggested that there may have been a steep authority gradient gradient at Ethiopian which would have discouraged a First Officer from correcting a Captain. If so, this would be a threat, but perhaps an unappreciated one if this authority gradient was deeply embedded in the airline culture. The primary unknown threat was the pending AOA malfunction. Another possible unknown threat was a lack of specific systems knowledge as it related to MCAS and the subsequent trim problem. For each of the identified (known) threats, what could have the crew done to mitigate them? Primarily, by actually identifying the threats and briefing any appropriate procedures - forewarned is forearmed. For example, one thing I do whenever I fly with a new First Officer is that I explicitly state that anyone on the flight deck can make an error and it was the job of the other pilot to correct those errors. I want my FO's to feel free to speak up. Slowing down and methodically using checklists and flows is another good mitigation strategy. In the more extreme cases, simply refusing to depart until the situation has been satisfactorily changed for the better may be the most appropriate mitigation strategy. The next question in this analysis is, "What were the errors?" Errors are caused by human actions or inaction that increase the likelihood of an adverse event. The difference between an error and a threat is that an error can, with careful attention, be quickly identified and crew members can find prompt solutions to the error. This is sometimes known as "trapping" the error. The impact of an error can, therefore, be quickly reduced if properly managed. If not managed correctly, an untrapped error can lead to an undesired aircraft state or create a new threat (known or unknown). Examples of errors include procedural errors, perception errors, miscommunications, and violations of SOP's (intentional or unintentional). The crew errors seem to fall in the following categories: perception errors (not picking up on obvious cues), procedural errors, CRM errors, and basic airmanship errors. Most of these have been extensively discussed, so I won't repeat them here. The next question is what were the barriers that should have enabled the crew to trap any error before they led to an undesired aircraft state? Traditional aviation barriers include policies and procedures, checklists, CRM, knowledge and aircraft handling skills, as well as external resources (ATC, maintenance, etc). As we look at this accident through the TEM lens, I think there are two very important questions: Why did the existing barriers fail? What happens when a barrier actually becomes a threat? |
Originally Posted by MurphyWasRight
(Post 10464802)
It has pretty well been established on this thread that pilot electric trim will work in all conditions (does not stall under load) and interrupts MCAS if active as shown by ET trace at ~05:40:27. This pilot trim input was possibly then interrupted by cutout switches.
Other than 'deer in headlights' loosing it I see a few possible factors: 1: Pilot accustomed to short blips not comprehending amount truly needed, this fits Lion Air when the FO was ineffective at the end when PIC handed over control while the Captain was mostly successful. 2: Trimming by column feel not position: May seem that AC is closer to trimmed than it is. In other words if you have been pulling really hard then just slightly hard may feel like close to trimmed, I am sure the pilot did not want to over trim given all the alarms. This might explain ET first retrim at 05:4-:15. Unfortunately we don't have the column force graph, just position. 3: Some as yet to be revealed flaw that interferes with pilot trim inputs; one possibility is biomechanical factors related to actuation switches after prolonged pulling. This is unlikely but could explain the final seconds of both accidents. Hopefully the final reports will fully address this question. Consider that if the airline policy was to use the autopilot to the max (sorry about the pun), their entire career of flying a Boeing product would be to takeoff, climb to 400 ft then engage the autopilot. For this short bit of flying there is no requirement to trim if the stabilizer trim was set properly. During any changes in speed or configuration the autopilot would automatically trim the aircraft. The aircraft would stay on autopilot until short final (1000 AGL or less), the autopilot would be disconnected and, assuming that no configuration or speed changes occur, the aircraft would not have to be manually trimmed. In speaking with friends flying outside of “the western world” this is exactly what happens. In fact, if the flight data analysis (which, in some cases is analyzed after every flight at some carriers) shows manual flying, the Captain gets queried about why. If it happens too often they get docked pay. To those of us that insist on doing some hand flying of our jets, this might seem preposterous however that is what is going on in many parts of the world. I would be interested in hearing if others are under this practice or have colleagues that are. |
Originally Posted by L39 Guy
(Post 10464894)
Another factor is that given the dependence on automation these pilots may have never trimmed this aircraft or any others in their airline career. Consider that if the airline policy was to use the autopilot to the max (sorry about the pun), their entire career of flying a Boeing product would be to takeoff, climb to 400 ft then engage the autopilot. For this short bit of flying there is no requirement to trim if the stabilizer trim was set properly. During any changes in speed or configuration the autopilot would automatically trim the aircraft. The aircraft would stay on autopilot until short final (1000 AGL or less), the autopilot would be disconnected and, assuming that no configuration or speed changes occur, the aircraft would not have to be manually trimmed. In speaking with friends flying outside of “the western world” this is exactly what happens. In fact, if the flight data analysis (which, in some cases is analyzed after every flight at some carriers) shows manual flying, the Captain gets queried about why. If it happens too often they get docked pay. To those of us that insist on doing some hand flying of our jets, this might seem preposterous however that is what is going on in many parts of the world. I would be interested in hearing if others are under this practice or have colleagues that are. I can just see it now.... My next session is going have some spot training involving hand flying while operating the electric trim switches up and down . |
Not a pilot - software engineer that started my career at a small company that produced hardware-in-the-loop testing and certification solutions for major aerospace and defense organizations. I experienced first-hand the sort of pressures and poor communications/management between entities that leads to audits/test results/certifications be pushed through too fast and in incomplete forms and to be honest it made me wonder if we wouldn't start to see failures like this tragedy unfold in the near future.
From the information available my gut feeling tells me that the systems design on the MAX was intentional, not an instance of well-meaning engineers making a mistake. There's the change of the rate of trim applied made late in the testing stages (from 0.6 units to 2.5 units per time period?) and the seeming lack of any sort of sanity checks in the MCAS system that just seems impossible to miss. There are very basic things that can be done in software even beyond bringing in more sensors to fall back on such as checking the values that come before (is it ever possible for AOA to jump over from 14 to 75 degrees in under a second?). This is something that should have been glaringly obvious to all involved in the process and if there was genuinely no bad intentions on Boeing's part then what this suggests is serious organizational rot. In either case I would hesitate to put confidence in other aspects of the design if this system is such a mess. I've yet to see a satisfactory explanation behind the changes to the trim cutout switches. Sure, I've read that they are always used at the same time but how does that justify neutering the behavior but leaving the two switches except for different labels? In what world does it make sense to get rid of the ability to cutout the plane's automation (STS, MCAS, probably others that I'm not familiar with) while still allowing the pilot to enter trim commands using the stab motor? What does this improve upon the original design of the switches? A change like this involves multiple engineering departments in order to implement and everything is documented and cross checked along the way. I'm very curious as to what the justification is here. And the mysterious short blips of trim shortly before MCAS dealt it's final blow... From following along in this thread and others I've come to the knowledge that trimming away pressure in the control column is one of the most basic aspects of flying that there is. The failures to get completely back into trim and then the final blips suggest to me that something else is wrong. From what I see other pilots saying it sounds like the equivalent of someone in the path of an oncoming semi-truck applying very slight turns of the steering wheel to get out of the way (sorry for the clumsy metaphor). If airline training has taken such a deep dive in quality over the years as to lead to pilots that don't have supposed basic airmanship skills, doesn't this imply that we should expect to see a steep increase in the amount of pilot errors leading to accidents? So far they have been confined to two hulls of the same make and model within the first years of it entering into flight... If you read all of that then thanks for taking the time to consider the viewpoint of someone from outside the profession. I've become somewhat obsessed with this MAX fiasco because just like it highlights the importance of airmanship to some pilots here, to me it highlights the importance of systems design and good engineering practices in the software and technology world. It is cheap and easy to alter products by manipulating lines of code but we must keep in mind that the impact that it has on the word is just as real as that of the other more "material" engineering professions. |
Originally Posted by Lost in Saigon
(Post 10464918)
As airline pilots our recurrent training often emphasizes or incorporates items that were the causes of recent airline accidents or incidents:
I can just see it now.... My next session is going have some spot training involving hand flying while operating the electric trim switches up and down . |
wheelsright, ‘But do you think that it is possible to use electric trim to counteract MCAS before using the cut-out switches?’ Overall, and biased by the accident outcome, I doubt that nose up trim would have been effective, particularly when considering time of recognition, realisation, and action. Continued debate considers if trim would / was enabled to override MCAS. Then, the erroneous rate / power of MCAS design would out perform trim by 9 sec down to 3 sec up. Also, there is debate and interpretation if trim is restricted due to design (EASA questions) and further if the aerodynamic forces restrict nose up motion - tail / elevator interaction. Then the above must consider the pilots perception of the situation, the need for trim and the extent of trim required in very stressful, surprising, and demanding conditions. All that must be confined to history. Whatever is changed in the MCAS operation it must not be able to mis trim the aircraft to extreme. As for judgement of crew performance, consider https://www.pacdeff.com/pdfs/Errors%...n%20Making.pdf Crews either fail to understand the situation, HF - thus choose the incorrect procedure Or with correct understanding, the incorrect procedure is chosen - HF (or procedure is not available - documentation, training, novel failures) |
| All times are GMT. The time now is 22:29. |
Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.