I’m not sure where the misconception comes from about simulator training being equal to the aircraft being a new type certificate. The DC-9/MD-80 series/B717 are all the same DC-9 type certificate despite spanning many decades, longer than the 737 even. Pilots receive a DC-9 type certificate to fly any variant, however, for obvious reasons, the FAA requires what’s known as differences training in a simulator to transition between the different aircraft.
I can’t possibly imagine that the FAA is or any aviation governing body is going to jump from iPad distance learning to full on new type certificate. B737 type rated pilots will receive 2-3 sessions in the MAX simulator and be sent on their way flying 737NG and MAX variants and the aircraft will go on to have a normal safety record.
The more concerning aspect of this entire situation is finding where the system broke down to prevent this additional training to begin with and how do we fix the breaks in that system.
They will have to fix the simulator first. As I understand it they can't currently simulate the handling of the aircraft under MCAS failure modes. If they correct the simulator and require pilots to train for the handling of MCAS failures then I have to problem with the reintroduction of the max on the same type certificate. From what we are hearing in the press it sounds like Boeing is still pressing for nothing more that a dumb iPad lesson. If the FAA falls for that they will loose all credibility in my eyes.
EU has its own oversight that has, at least at some point in the past, been pretty much copying FAA word-for-word. Other countries obviously have their own, but also leaned on FAA for its influence. Yet another aspect of soft power of the US.
"I don't know that understanding this system would've changed the outcome on this. In a million miles, you're going to maybe fly this airplane, maybe once you're going to see this, ever. So we try not to overload the crews with information that's unnecessary so they actually know the information we believe is important," the Sinnett said.
Boeing 757 and 767 share a common type rating as well. I believe that differences training is required to move from one to the other, but I’m not 100% sure.
How would you accurately simulate the force needed to turn the trim wheel under aerodynamic loads that vary with elevator position? It sounds like a very tricky mechanical setup.
This previous story (https://news.ycombinator.com/item?id=19902068) implies that the simulators try to simulate the forces, and in a particular simulation even the simulator crew wasn't able to move the wheel:
"What the U.S. crew found was eye-opening. Keeping the aircraft level required significant aft-column pressure by the captain, and aerodynamic forces prevented the first officer from moving the trim wheel a full turn. They resorted to a little-known procedure to regain control."
The way I would approach it (based on mechantronics I have done for robots) is to back up the control wheel with an encoded motor. (Position/rotation encoder on the tail shaft.) The encoder will give you wheel position, the motor allows you to apply arbitrary torque to the wheel to simulate the feedback forces felt by the pilot. Typically, that would be done by measuring motor current and writing a control loop that controlled for target wheel position (based on encoder) and target maximum motor current (as a proxy for torque). Then the simulator would give out the simulated results of wheel position and torque, which the control loop would attempt to follow with its control law.
Alternatively, you could put an actual torque sensor between the motor and the control wheel, which might buy you some accuracy at considerably more expense. For a flight simulator, expense is unlikely to be an issue -- I tend not to have that luxury, and typically don't need that much accuracy anyway.
These kind of control systems are all over modern aircraft, much less simulators. Any new grad control systems engineer should be able to knock something like this out.
>
The way I would approach it (based on mechantronics I have done for robots) is to back up the control wheel with an encoded motor. (Position/rotation encoder on the tail shaft.) The encoder will give you wheel position, the motor allows you to apply arbitrary torque to the wheel to simulate the feedback forces felt by the pilot. Typically, that would be done by measuring motor current and writing a control loop that controlled for target wheel position (based on encoder) and target maximum motor current (as a proxy for torque). Then the simulator would give out the simulated results of wheel position and torque, which the control loop would attempt to follow with its control law.
Or you just buy a COTS servo that does all that for you. Just saying, hardly a need to reinvent these particular wheels...
Yeah, good point. I'm sure there are servos that have the required specs. I'm just not in the habit of looking at catalogs where the prices are that high :) Maybe I should switch to building flight simulators so that I can order the good stuff :)
If pilots are trained in how to recognize the emergency and what to do, and they are instructed that in an actual emergency the forces on the trim wheel (presumably what they are referring to) will vary, they will respond appropriately. Simulators can't exactly replicate extreme conditions, upsets and whatnot.
If you're training for trim runaway it would be helpful to simulate the full forces but I'm not sure if it reflects a worrying lack of preparedness.
The most important fix is that the MCAS should not be able to put the trim in an extreme position, the fact that it did is quite worrying.
This particular article seems much ado about very little.
Root problem seems to be: Software puts trim in an extreme position. It should never do that.
Aviation Week article: if you are in an electric trim runaway, you need to turn off electric trim, then in order to turn the trim wheel manually, you need to nose down to reduce aerodynamic forces. Seems necessary to train for which it suggests was not the case.
NYT article: Simulator doesn't simulate the forces on the trim wheel correctly (although Aviation Week suggests they were able to simulate something useful)
The NYT article really seems like the least of the problems.
Why does MCAS have to take control of trim at all? Would it not be sufficient to alert the pilot of an imminent stall situation so they can adjust the angle of attack themselves? Is it because doing so would put the MAX too far afield from the old 737 such that it would require simulator training?
There's an airworthiness rule requiring monotonically increasing yoke backpressure as the plane approaches a stall. It appears that the MAX violates this rule aerodynamically due to extra lift at high power generated by the high and forward nacelles, combined with the yoke being mechanically coupled. It can get easier to induce a stall as the plane approaches critical AoA.
This isn't directly the same thing as saying the MAX will stall itself: if you aren't pulling back on the yoke near critical AoA then you don't stall. It's just easier to stall with the yoke than regulations say it must be. MCAS "fixes" that handling issue.
I don't think an audible warning would be sufficient to turn unairworthy behavior into airworthy, so if all the assumptions above are correct, that's why it has to use trim -- or a stick pusher, but perhaps that wouldn't have enough control authority and also I'm not sure the MAX has a stick pusher, as opposed to just a stick shaker. It's a very mechanical cockpit, in general the forces you feel are coupled to aerodynamics, in stark contrast to an Airbus (or even more modern Boeing airframes).
In the working case, MCAS presumably triggers at a particular AoA, trims nose down (but is this a fixed amount and what is it?) which in effect requires more stick back pressure to maintain the angle of attack. Thing is, it seems like MCAS, again in the working case, has a trigger AoA and will incrementally nose down until it goes below some defined angle of attack - which is not really a stick force moderator alone. It's acting as a kind of AoA guardian.
Yes. MCAS is designed to make the Max handle like previous 737s when manually flying at high angle of attack. Without MCAS the handling would be different enough to require simulator training.
> A good software assurance program in a flight environment should also be ensuring simulators are of high enough fidelity for testing and training.
Actually, "must" not "should". Aeronautics certification standards require tools such as simulators to be qualified for their intended use. Similar requirement have been introduced in recent release of railway standards.
I agree with you're correcting my imprecise wording, but it still seems like there's an issue with the certification process if this particular "must" didn't get built in.
I guess my question is: why didn't the QA flag get raised on this or if it did, why wasn't it given any credence?
I'm not trying to armchair quarterback this, but I'm legitimately curious if this issue is one the certification process is expected to catch
Certification process is generic enough to apply it to different types of aircrafts. There is no specific requirements for different subsystems introduced by manufacturer. Simulation fidelity is tested against real world data but it is not possible to collect data for every condition that might occur during flight. You may refer to FAA FSS Level-D QTG document for further reference. [1]
Considering how fast and strong the trim wheel is spinning under MCAS control the simulator may be unsafe to use. Touching a wheel spinning so fast is outright dangerous.
The trim wheel spins like that anyway under normal operation, both due to pilot input and while on autopilot.
More importantly, you don't grab the trim wheel while it's moving... If the aircraft is operating normally, you use the trim switch on the yoke.
The only time you'd spin the trim wheel by hand is if you've had to cutout the electric trim motor... and once that's done it's not going to move on its own anyway.
They teach pilots a specific way to grab the trim wheel for that scenario as to not break their wrists. It still can (and occasionally does) cause abrasive wounds to the hands however.
> "The system is explicitly designed to override pilot action and prevent them from regaining control authority, in order to avoid an inadvertent stall."
If it's not an anti-stall system what would you characterize it as?
Good question. Not sure the "flight envelope protection" description in the Wikipedia article is good.
If I'm not mistaken, MCAS was put in to fulfil certification requirements stipulated in 14 CFR § 25.173 - Static longitudinal stability, and § 25.175 Demonstration of static longitudinal stability, which states that "The stick force curve must have a stable slope". The MAX has a curved slope in some regions of the flight envelope, due to the fact that the nacelles generate lift in front of the centre of gravity. (It is not clear to me whether the slope of the stick force at high AoA in a MAX without MCAS goes negative, or just decreases (while staying positive)).
Note the Wikipedia article you quote also says:
> It is thus distinct from an anti-stall device, such as stick pusher which physically moves the pilot's control column forward when the airplane is approaching a stall.[5]
See also here [1]:
> Numerous reports have incorrectly said that MCAS is a stall prevention, or stall recovery, system. It is a Maneuvering Characteristics Augmentation System and would not be expected to activate during any normal flight conditions.
Or here:
>> It does seem MCAS is closely related to protecting the plane from entering a stall scenario.
> Only in the sense that it is giving the pilots the feedback, through the stick force, about how close the plane is to a stall, that the pilots are used to from previous 737 models. The point of MCAS is that without it, the stick force feedback as a function of angle of attack would be different from what the pilots were used to, so they might misjudge how close to a stall they were.
> In a tense meeting with the American Airlines pilots union after the crash, a Boeing vice president, Mike Sinnett, said he was confident that pilots were equipped to deal with problems, according to an audio recording review by The New York Times. A top Boeing test pilot, Craig Bomben, agreed, saying, “I don’t know that understanding the system would have changed the outcome of this.”
No he is agreeing that the pilots were as well equipped as they could have been, because he believes understanding the system wouldn't have made a difference anyways.
Hmm, I see. So the Boeing VP was claiming that a pilot following procedures properly would've saved the plane, and (at that time, after the first crash) the test pilot (also from Boeing) agreed. Makes sense now, thanks.
And then a second crash occurred where they followed Boeing's procedure, it didn't work, were forced to diverge and then crashed.
Boeing's procedure only works if you catch it quickly. If you let MCAS trim the aircraft too much it is difficult to recover, particularly at low altitude.
I don't think it was a sarcastic response, and I don't think GP meant it that way, either.
Remember, Boeing at the time had not disclosed MCAS to the pilot community. Both speakers are from Boeing, and the first one says that the accident pilots, even though they did not manage to fix the problem, were well equipped (by standard Boeing procedures for stab trim runaway) to do so. The second speaker agrees, saying that knowing about MCAS would not have made any difference.
> The second speaker agrees, saying that knowing about MCAS would not have made any difference.
The article directly quotes him as saying, "I don’t know that understanding the system would have changed the outcome of this." The "this" refers to a deadly crash in which everyone died. He is saying that if a pilot knew about MCAS, the plane still would have crashed. He said this in response to a claim that pilots didn't need to know about MCAS. That is disagreeing by using sarcasm. The Times interprets it though as agreeing and doesn't see the sarcasm.
To summarize:
Claim: Pilots don't need to know about MCAS.
Response: Correct, even if they knew, given the system's flawed design, everyone still would have died in this case, so, technically, knowing about it wouldn't have helped.
Now maybe that's not what he meant. But it looks like he did. I don't believe that the context of the conversation supports an interpretation that he was agreeing that sufficiently skilled pilots would not have crashed whether or not they knew about MCAS given that he refers to "the outcome of this" not changing, the "outcome of this" being the crash.
What part if that is sarcasm? If anything he's saying a failed AOT sensor means a fatal crash regardless of whether you know why or not. Which is alarming and terrifying.
Given that he's a Boeing test pilot (and in the context of agreeing with the VP), his take would probably be that pilots should have followed the good old Stab Trim Runaway checklist, and all would've been fine (whether or not you know about MCAS).
I agree with you that the test pilot's remarks were not sarcastic. However, you write "he's saying a failed AOT sensor means a fatal crash regardless of whether you know why or not", and I disagree with that. He's saying a failed AoA sensor invoking MCAS can be fixed like any other Stab Trim Runaway, regardless of what you know about MCAS.
Except that it doesn't manifest like a classic Stabilizer Trim Runaway.
The classic Stabilizer Trim Runaway is a continuous uncommanded actuation of the trim mechanism on a particular direction.
Catastrophic MCAS failure manifests as a series of discrete, amplifying activations, which are far easier for a pilot to dismiss as the normal operation of the Auto-Trim, and speed trim systems.
The is is a manifestly different pattern to look for, and was only caught by one air crew (on the penultimate flight of the Lion Air aircraft) by the assistance of a third, uninvolved pilot along for the ride.
I understand how one could see Boeing's internal test pilot as being reasonable, but keep in mind who signs his paycheck, and future prospects for him if the company gets found liable.
Perspective, and understanding of what different parties have at stake is essential in evaluating what is actually being communicated, and why. Words have long tails in the aftermath of a crisis like this.
Yes, agreed with everything. I don't think Boeing's test pilot's dismissal (basically: they had their checklists, and knowing about MCAS wouldn't have made a difference) was reasonable, but neither do I think it was sarcastic.
Craig Bomben is full of shit. If they had told pilots how MCAS worked, almost surely one of the tens of thousands of 737 pilots worldwide would have had a WTF moment and sounded the alarm about it being able to apply trim beyond the ability of a human to overcome while depending on a single sensor.
> Boeing has maintained that simulator training is not necessary for the 737 Max and regulators do not require it, but many airlines bought the multimillion-dollar machines to give their pilots more practice.
Boeing is only arguing this because if they say simulator training is necessary, then the plane must be recertified. Their executive MBA analyst and stock watcher types under no circumstances will tolerate recertification, therefore regardless of facts, their position must be that simulator training is not necessary. Even though it obviously is. Key point to observing this and everything else about this debacle is that Boeing has no concern whatsoever for safety, professionalism or lives, and is only interested in their own profit, executive bonuses, and stock dividend gains and will do absolutely anything to protect those, even though it inevitably is going to lead to another crash and more deaths. At which point they will again start by doing everything they can to blame the pilot and the culture of whatever country he hails from, as well as paying shills to post nonsense and attack pilots, engineers and others who question their narrative publicly.
It isn't enough to just say a statement when you want to make a point. It's usually expected, especially here on HN, to put some effort into demonstrating why you believe that statement is true, something which his linked comment doesn't do. It doesn't need to be anything elaborate, even references to other comments mentioning things FAA has done would be more than enough to make the point, which gives people something factual or "real" to respond to or look into.
I don't think there's anything wrong with his statement per se. You don't always need to write a treatise whenever you want to join a discussion or throw in your two cents. Just that when he says he's made the point before, it'd be more helpful to link to something more substantial than the same statement in a different place.
As an Engineering firm, your job is to tell the customer, "I'm sorry, I cannot give you approval to do that." The idea being that you, the Engineering Authority, are bound by professional ethics, and a commitment to the public interest.and equipped with the specialized knowledge required to determine the outcomes of a design before it is actually rolled out.
It is never justifiable to make an unsafe bridge. Airplanes are no different. No matter the size of the money bag.
The trouble with making this a crime is then you strongly incentivize people to deny, delay, obfuscate, etc., when what you really want is full cooperation. The full cooperation "no fault" approach has led to aviation being incredibly safe.
"Anything you say can and will be used against you" means you will never get to the truth of what caused an accident, and it will be much less likely to ever get fixed.
Furthermore, if you criminalize mistakes, you'll have a hard time getting top quality people to work in the industry.
> It's hard to believe that any action by Boeing has furthered the investigation in this case.
If that's indeed the case, it is likely motivated by fear. Any admission of error leaves Boeing open to massive lawsuits and the regular calls for criminal prosecution. This entire issue has played out in the media with daily newspaper articles for months. It's quite unlike any other airliner accident investigation I recall.
Look at how the CEO has handled this mess. It is not the work of someone approaching the problem as an engineer. He is approaching it as a politician, with much heart felt meaningless responses, blame deflection, ignorance of undesireable facts. An engineer's approach would be to identify the problem through logical, rational thought, generate a few choice viable solutions, test each solution for effectiveness, and implement the best combination of them.
And I'd say it is common knowledge by this point that Boeing is ruled by their shareholders' bottom line more than any other metric. So yes, Boeing attempted a business move to counteract the success of Airbuss's new plane by speeding through the design of the Max. Recertification would take too long, losing them market share, so they rode the line as close as they could to avoid recertification and minimize cost. The rushed design meant failures were more likely, the lack of recertification meant the pilots in the cockpit didn't have knowledge to properly deal with these failures, and now we have a lot of dead people. There's not much gray area to this story.
I was definitely one who thought it should be so simple to follow a standard stab trim runaway and made several comments to it too. But the more it drags on the more I'm learning from the data that the procedure to correct is just bullshit.
Boeing is now in unsaveable waters for this situation and all doubt is being removed. I still am upset that any airline was able to in good faith say "nah we don't want all the safety options not worth it" but here it doesn't seem it would have helped much the more data comes out.
Since I’m in your old boat (so to speak), what additional evidence suggests that the Ethiopian flight was not saveable by a crew who flew the airplane away under the stick shaker, reduced thrust, and then fixed the trim at altitude? (I’m not saying that is easy or even “par”, but it does seem like they had a salvageable situation until they turned the stab trim back on after correctly putting it in cutout and then left it on and let it run the trim nearly full nose down.)
https://news.ycombinator.com/item?id=19902068
Even the simulator proves that it was impossible to save the plane using Boeing suggested procedures.
Now imagine doing that while you know that you are going to die soon and there is your answer.
The content on that link suggests that it is in fact possible to save the airplane (and on the fatal flight, the Ethiopian Air crew got the airplane climbing and I believe exceeded 8000’ before losing control as they fought the bad hand they were dealt).
Possible to save the airplane doesn’t mean that it’s expected, but it’s quite a bit farther from “the simulator proves that it was impossible”.
So the extra detail can be found in the flight data from the flights FDR, and CVR, and is laid out in the crash report.
If you combine that with a bit of knowledge with regards to the basic physics involved, and complications introduced by not fully briefing pilots on the system, it becomes obvious the crew was fighting a losing battle without the information we have now.
The prelim report, and summary of the event can be found at the link above.
To lay things out, ET302 ended up having an AoA sensor malfunction.
One of the Appendicies includes a page from the flight manual, there are several pages mentioning the possibility of runaway stabilizer, along with a list of other symptoms that may be experienced due to a failed AoA sensor input to the Flight Control Computer.
There is also the page on which the Stabilizer Trim Runaway checklist is printed, which also specifically characterizes a Stabilizer Trim Runaway as an uncommanded continuous stabilizer trim activation event with mucho emphasis.
The page containing the closest thing to a hint as to the existence of MCAS (page 29), gives a small blurb that the flight computer does something, and that maybe you should be ready to run through the Stab Trim Runaway checklist.
Now. Look at the list of associated symptoms.
IAS Airspeed disagree is one of those items.
There is a memory item for 737 for airspeed unreliable, which if run through before the Stab Trim Runaway can greatly exacerbate the difficulty of recovering from the out of trim condition.
Which checklist gets remembered first then, can have a drastic effect on whether you can recover the plane, and the checklist to solve the problem that MCAS caused, if committed to memory entirely, doesn't even characterize the entry point for the checklist in a manner consistent with MCAS malfunction.
This is without even taking into account what actually happened, mind. I'm just pointing out documentation/communication failures that could contribute to a pilot not making the connection in time, while trying to parse the cacophony of alerts to ascertain what his plane is trying to tell him.
This is coupled with his autopilot being unable to hold things stable so he can troubleshoot, his FEEL DIFF PRESS light indicates the haptic feedback provided by the computer may not be appropriate to actual conditions, so the finely tuned muscle memory that pilots can rely on as a recall aid may be inaccurate or downright dangerous.
In short, nothing about this situation feels right, the pilot who has had it ground into them to trust their instruments over their own senses is actually dropped into a situation where almost all of their instruments are lying, because the computer is correcting actual readings with garbage AoA data.
Again, we've not even touched on what actually happened. This is just enumerating what could have happened, which constitutes possibilities that a pilot would have to navigate in that moment to successfully come to a conclusion about what was going on. Without the luxury of priming by knowing that's what you're going into the simulator to practice today, and with the additional entropy and complacency that comes from having flown uneventful flights 100's of times before.
Now, take into account that a highly experienced pilot, clued into the possibility of the Stabilizer runaway, did manage to troubleshoot, but was left in a bad spot.
The torque required to actuate against the airstream at the speeds the flight had achieved were such that they needed the electrical trim motor to overcome the resistance. So they turned it back on, and re trimmed. However, they were unaware they would need to keep signalling with the trim switches to keep the MCAS system from kicking in 5 seconds later.
This was a system, snuck in, under the radar, and documented in a completely inconsistent manner to the catastrophic outcomes that could arise from a malfunction.
I'm not a legit Aeronautics Engineer. Just a polymath with a penchant for running down and understanding things without having the benefit of having someone explain them to me in good faith. (Immensely important skill for auditing purposes).
I don't want you to just take my word for anything. Go back through most of my posts on 737 MAX topics if you'd like, and do your own research.
-The relationship between Boeing and it's regulatory framework, was of so much concern, the Office of the Inspector General was calling it out.
-The A320neo taking them by surprise.
-The work culture at Boeing
-the reprioritization of shareholder value over product quality
All of these to any Quality Engineer only leads to one variety of outcome. It's a really pervasive pattern.
When I'm tackling an engineering problem, that is how I do it. Not with ignorance of facts that I don't like, or heartfelt sentiments that don't solve the problem. I sit down and science the shit out of it. I use data, model the problem, create plausible solutions, math the metrics from those to form differentiation. I do this on a weekly basis. What I listed isn't supposed to be a specific methodology that one uses to accomplish any engineering task; it's a very general representation for how to approach a problem from an engineering perspective.
I have no idea what you mean by "teams with proper objectives... not a guarantee". Yes the CEO of Boeing is not a team, he is the team leader; he has all the resources he could possibly need to accomplish this task, and I think that his biggest problem is that his motivations are shit, which is why he is playing the politician instead of kicking his company into high gear to keep any more of the planes from falling out of the sky. And nothing is guaranteed, ever... Are you trolling me? Was that a troll comment? Can't make sense of what you're trying to say, other than telling me I'm wrong.
Edit: changed "most" to "very" to sound less pompous.
> Not with ignorance of facts that I don't like, or heartfelt sentiments that don't solve the problem. I sit down and science the shit out of it.
This isn't a TV show. Being an engineer or having an engineering background doesn't turn you into a machine. Nobody works in a perfectly rational vacuum like you describe.
Engineers solve technical problems, but how you solve those problems is a process that usually involves many people with defined responsibilities and procedures to produce an outcome, dependent on the proper resources and objectives. This process is needed precisely because of the fallibility of humans, and even then it can still fail, which is why this problem happened in the first place.
The role of a CEO is specifically not one of solving technical problems so this isn't applicable anyway, but this Boeing disaster is an outcome of both poor engineering and business practices.
Yes, it's not a TV show, which is why I used my own real world example. How do you approach an engineering problem? You've complained about my methodology, but haven't offered any alternative... Rational approaches are how an irrational person can solve real world problems.
I realize the role of the CEO, and I was specifically responding to the many claims that "he is an engineer". He isn't approaching this problem like an engineer, as I've said twice now, he's approaching it politically. That may be the status quo for CEO behaviour, but it is a strong deviation from an engineering approach. I also stated that this problem was caused in the first place by Boeing's business practices hamstringing their engineering practices. They have many, many amazing engineers at Boeing, many of whom voiced objections to the rush jobs and regulation gerrymandering that Boeing's business side (run specifically by the CEO) used to get this aircraft flying before it was safe. The outcome of this is very visible.
I'm likewise skeptical. I've done a number of projects for Boeing, and their engineers are some of the most cautious and careful customers I have ever had. Automotive tier-one customers need it done on time and under budget, no matter the shortcuts required to make that happen, but if the junior janitor at Boeing sees something he feels is maybe suboptimal we'll halt the project and fully study the concern. I keep my schedule clear for 2 months after an aerospace project is supposed to be done, because it probably won't be, and it will be better for it.
I just can't square these experiences with this news. Perhaps design and manufacturing are more siloed than I understand, or perhaps there really is an MBA culture running requirements and certification separate from the engineering of the physical components, but in my personal experience Boeing isn't a place that cuts corners.
I've done a number of projects for Boeing, and their engineers are some of the most cautious and careful customers I have ever had.
And if you've been following the news you'll see the engineers had less and less say. Going back to the 787 program — Boeing tried to outsource a ton of the engineering and that resulted in minimal oversight and the type being grounded for a bit.
Keep in mind Boeing is a company where the previous CEO (McNerney) bragged about being able to make the employees cower in fear and pushed for dramatic cost reduction above all else.
I'm curious how long ago you did your projects for Boeing though, as there seems to be an appreciable shift in culture.
Welcome to HN, where 95% of the population has an absolute conviction about the knowledge that they gained in 5 minutes of internet searches, where every problem is obvious and easy to solve, and where a basic knowledge of computers allows one to opine about everything from politics to sex to optimal routing options for oil pipelines.
most of the people posting here about this topic don’t know a damned thing about it, they’re just regurgitating something they read on the 2nd page of google search results.
The fact is that most HN comments are full of shit.
In the same way that a good caricature or satire exaggerates to some degree while capturing something true, I think this comment probably ascribes a bit more arrogance to the average HN comment than is actually there, but I laughed anyway because you phrased it really well and in some cases you're actually spot-on.
I think that "where a basic knowledge of computers allows one to opine about everything from politics to sex to optimal routing options for oil pipelines" maybe captures our entire historical era. Nice work and a hearty lol
Whenever these kinds of things happen, it usually doesn't seem to stem from the engineers themselves, but has far more to do with the organizational culture above them in management. I think we especially feel this in software, where people outside of software development have no idea how complex and hard it is to build these big systems and what can go wrong, and have crazy expectations and sell things that aren't possible to other people who also have no idea about software development with crazy expectations. Engineers, no matter how conscientious, can only do so much in the face of business requirements that are often out of touch with reality, where deaths or other problems are often considered a footnote or just another business expense.
In that, a senior executive talked about how the quality and culture at the SC office (where the 737 MAX was built) of Boeing was significantly inferior to the rigor and care that people otherwise expected of Boeing.
> Since the two fatal crashes of the Boeing 737 Max, airlines around the world have moved to buy flight simulators to train their pilots. They don’t always work. Boeing recently discovered that the simulators could not accurately replicate the difficult conditions
Typical article which does not address the entire context of the issue. In other words it isolates what a 737 simulator can't do without regard to what a typical simulator (for another aircraft or by even a different manufacturer) is or is not able to do. [1] The idea is to make the manufacture look like a total ignorant screw up in every way to create anger and compelling content.
[1] This is like a news story talking about what a company fails to do and then gets hacked without talking at all about how likely the same thing happens at another company.
Are you suggesting that most simulators get control forces wrong for things that can realistically occur within the recoverable flight envelope after one single sensor fails? I'd have thought that that is something you'd want to get right.
I can’t possibly imagine that the FAA is or any aviation governing body is going to jump from iPad distance learning to full on new type certificate. B737 type rated pilots will receive 2-3 sessions in the MAX simulator and be sent on their way flying 737NG and MAX variants and the aircraft will go on to have a normal safety record.
The more concerning aspect of this entire situation is finding where the system broke down to prevent this additional training to begin with and how do we fix the breaks in that system.