SoylentNews is people
SoylentNews
Boeing making new 737 MAX software updates to address computer issue:
Boeing Co (BA.N) said late on Tuesday it will make two new software updates to the 737 MAX's flight control computer as it works to win regulatory approval to resume flights after the jet was grounded following two fatal crashes in five months.
The planemaker confirmed to Reuters that one issue involves hypothetical faults in the flight control computer microprocessor, which could potentially lead to a loss of control known as a runaway stabilizer, while the other issue could potentially lead to disengagement of the autopilot feature during final approach. Boeing said the software updates will address both issues.
The Federal Aviation Administration said on Tuesday it is in contact with Boeing as it "continues its work on the automated flight control system on the 737 MAX. The manufacturer must demonstrate compliance with all certification standards."
The largest U.S. planemaker has been dealing with a number of software issues involving the plane that has been grounded since March 2019. Boeing halted production in January. Boeing said it does not expect the issues to impact its current forecast of a mid-year return to service for the plane. Boeing said the new software issues are not tied to a key anti-software system known as MCAS faulted in both fatal crashes.
(Score: 4, Insightful) by RS3 on Thursday April 09 2020, @08:32PM (11 children)
Just to be very clear, I am 99% in agreement with you on all points, but one small thing that makes a difference in many of the results: every plane handles differently- maybe even more so than the handling differences between various cars, trucks, tractors, etc.
Most commercial pilots just fly one brand / model plane, and as such, they get very used to the handling. So any significant differences would understandably require additional training.
Thrust pitch-up reaction dynamic is a thing on all planes. Again, as I wrote in my previous tome, I'd really like to talk to a big plane pilot. What I'd like to know is: does the MAX engine size and placement really make the MAX difficult to handle? Or is it just somewhat different from standard 737, and as such, FAA rules required additional pilot training? I'll answer that question:
It is an established fact that without MCAS, 737 pilots would have needed additional training to fly the MAX variant. From that we can logically deduce that the MAX would be okay to fly without MCAS- just that FAA wanted to be sure pilots understood how it differs from standard 737.
I've been following this story ever since the Lion Air (first) plane went down. There were many incidents with MAX planes previously, and the pilots were able to fight and win. In some cases they figured out to turn OFF the electric trim. That bugged me: to me, "trim" means a FINE adjustment. NOT a major full excursion elevator movement. There are many "black box" accounts of MAX planes going full nose-down. MCAS should NOT have that amount of control, and it doesn't need it.
The whole thing is a case of corner-cutting from the beginning.
And yes, I'm an engineer, and yes, I'll defend the engineers: we don't get to make these kinds of decisions. Ultimately the MBAs do, and generally engineers know what's coming from them- constant pressure to cheapen and corner-cut. Let pilots and engineers run Boeing and the problems will self resolve.
(Score: 1, Informative) by Anonymous Coward on Friday April 10 2020, @02:14AM (2 children)
The type rating would not transfer because the differences in performance appears to be just different enough between the aircraft. It is sort of like driving a Toyota Corolla LE for a living and then being switched to Toyota Camry XSE. Except they don't tell you about the fact that you need to give it 5 degrees more steering input as it pulls to the left more than you are used to nor about the electronic lane-keeping assist. In most normal situations, the lane assist will compensate for the extra pull when you don't realize it and most of the time you'll compensate for the pull without really realizing it. But when things go wrong, the assist can make the things worse than you think and the tendency to pull to the left can make emergency corrections incorrect if you aren't used to it.
BTW, thrust == pitch up is not universal across all planes with most pushers being the obvious examples of where it isn't.
(Score: 2) by RS3 on Friday April 10 2020, @04:27AM (1 child)
Awesome post. Wish I could credit you, mystery person.
I agree with what you've written, and I agree with your analogy. Over my driving lifetime, I've occasionally driven radically different vehicles. Sometimes the steering, braking, accelerator pedal, etc., are quite different. My job, as 2D pilot, is to learn and adapt. Great example- on my own car I recently mounted a great-looking slightly used tire that pulled pretty strongly. No matter which side I put it on, or which other tire I matched it with, it pulled pretty bad. Point being, I had to adapt, learn, compensate, and had no problems. I would hope a commercial pilot would be at least as good as I am. If they're not, then we have too many square pegs in round holes in this world. Sigh. I feel strongly that is true regardless...
And again, my other point about the MAX plane- that training was an option kind of proves the MAX is flyable without MCAS.
I might be a radical, but I would advocate testing drivers in simulators, subjecting them to fairly difficult situations and maybe not letting some people drive. I would hope a pilot could compensate for some unexpected pull. Heck, look at landing in heavy crosswinds. That's got to be harder than compensating for MAX engine placement and the thrust pitch issue.
You may know this, but when you turn a plane you lose altitude, so you have to pay attention and you may have to compensate with extra throttle, rudder, and maybe flaps if it's bad (losing altitude) enough. I've done that. But not the flaps part.
(Score: 0) by Anonymous Coward on Friday April 10 2020, @05:58AM
Don't get me wrong, it is totally flyable without the MCAS. The problem is that the performance on paper, as that is the only experience I have with the MAX, is different then the nonMAX. Because it is too different, especially near stalls, they had two choices: require a new rating or somehow get it to perform like the nonMAXes. They chose the latter, which isn't surprising since the former would have cost them more money, possibly more than this may end up costing them after all this brouhaha.
(Score: 3, Insightful) by bzipitidoo on Friday April 10 2020, @03:57AM (7 children)
It's a general problem with greedy, reality denying idiots being the final decision makers, having too much authority. That's why Fukushima melted down, and why Deepwater Horizon and the Exxon Valdez resulted in major oil spills, to name just a few industrial disasters. They were not freak accidents, not a matter of unforeseen problems. They were totally foreseeable, and foreseen. The people in charge cut too many corners. They ignored or silenced the engineers who tried to warn them. They didn't understand, and didn't care to understand, the risks they were taking, and acted as if the odds of disaster were less than 1%, when in fact the odds were quite high. Fukushima in particular was highly likely to be a disaster, something like over 90%, unless a lot of changes were made. The power plant was a ticking time bomb.
When enough corners are cut, reality whops us hard with a clue bat. You'd think we would be more careful about who we let into the driver's seat, and then, no matter who they are, we don't let them go wild. Until we improve on that, we'll keep seeing this kind of tragedy now and then, the totally preventable disaster that happened anyway because the daredevil, risk takers in charge got more reckless than ever, not trampling upon just one or two safety measures, but running over a whole host of them, guaranteeing trouble.
(Score: 2) by RS3 on Friday April 10 2020, @04:52AM (5 children)
Space Shuttle Challenger is surely one of the bigger examples in my mind. Maybe Titanic too. Somewhere I have a book of major disasters. Yes, in my professional career it's been many many losses (never bodily) due to non-technical-types overruling us tech-types. So much waste and loss. Stunning stupidity.
Fukushima is horrible. Not sure where to lay the blame though. It seems like they had disaster recovery plans in place, but didn't anticipate the extent of the potential problems. I don't know enough about who made the decisions there.
Agreed on Deepwater Horizon.
Not sure about Exxon Valdez. Wasn't that just human error? Like that Italian cruise ship Costa Concordia? Draft? Shoals? What that mean? /s
I'll tell you a somewhat inside story: the TMI (Three Mile Island) nuke disaster could have been avoided if they had spent some $ on a flow monitoring system that now most (if not all) nukes have. That disaster is very complicated, but if they had had the flow monitoring system in place, they would have known what was happening, including the stuck-open steam vent valve. Not sure if you know that story, but the workers deduced that the reactor vessel was overly full of water, when in fact the reactor core was becoming uncovered and going critical. I'm not willing to write much more about it here, but the point is that non-technical people made that decision- to save money and time.
(Score: 3, Interesting) by bzipitidoo on Friday April 10 2020, @02:29PM (4 children)
Exxon Valdez, and several other shipping disasters, are why oil tankers are now all double hulled. Similarly, the Titanic tragedy spurred a number of safety improvements, with the big one being thou shalt provide enough lifeboats for everyone. Double hulls are not infallible, of course, but do cut way down on oil spills. Costa Concordia was human error. The captain, who was noted for his recklessness, and therefore should never have been promoted to captain, finally had the odds catch up to him.
Blame for Fukushima rests squarely on the decision makers. It was not just one mistake, but a whole parade, to save money. First, they didn't build the wall high enough. They were warned it needed to be higher, but they didn't want to spend the money. Arguably, it should not have been located on the coast at all, but that call was made to have ready access to sea water, for emergency cooling purposes. Next, backup diesel generators were idiotically located in the basement, where they were guaranteed to be flooded if the wall was topped. The generators were not in working order anyway, as they had skimped on maintenance. Another issue is that most existing nuclear power plants have very antiquated designs, and lack a lot of safety features that have been invented more recently. Fukushima was no exception to that rule. Should have been decommissioned years before the accident, but again, money.
But let us not think that engineers and technical people are immune. I once worked at a small company in which the developers, for their convenience, insisted that the database not be password protected. The boss went along with that decision. The DBA protested, in vain. And one day, a few months later, it happened. One of the devs thought he was blanking the test database, but had pointed his little script at the production database by accident. Took out everything-- the company website, all our customer's data, all gone in an instant. DROP TABLE on every table. I was chatting with the DBA when it happened. He had turned to check something on the website, and it wasn't there, after having just read something else off the website hardly 15 seconds before. Then it was a mad scramble. We didn't know what the f** had happened. First checked that our browsers and Internet connections were working. Yes, they were. Had we just been hacked? I frantically started checking for online intruders, trying to see who else the systems reported as logged in, though I wasn't at all sure that wasn't futility. I saw nothing there, and nothing in the logs. All seemed in order. When the DBA reported that the data was gone, we narrowed our focus to trying to figure out what had happened with the database. About then, the dev responsible for the immediate error confessed. But the root cause was of course the refusal to protect access to the production database.
Our DBA began the long, hard work of restoring from backup. First, he discovered that due to lack of space, recently another dev had cut the frequency of the database backups way back. There was no daily backup any more, there was only weekly, and, of course, the most recent one was 6 days old. However, he had logged all the transactions, and they were still available. It took him a day to get the database restored to the point it could be used again, and support the company's website. Then it was 3 weeks to run all those transactions again. He managed a miracle, a near perfect restoration of the database. That was more than they deserved.
I definitely sympathize with security skeptics who decry password excess, but in this case, it was such a small thing to ask, to protect vitally important data. I got busy writing a much safer replacement script. No more pushing out to a target server, no. Instead, I made it so you logged into the server to be altered, and pulled in the alterations. Much harder, perhaps impossible, not to realize you are logged into the wrong server before doing a pull.
(Score: 2) by RS3 on Friday April 10 2020, @04:04PM (3 children)
Wow, thank you for all of that. I didn't know Fukushima was such a compromised design. I knew most of what you wrote. I could never figure out why the emergency generators were low in a building. Maybe it would have cost more to put them up higher due to stronger building framing? Maybe they wanted to keep them near the pumps they had to run?
Wow, what a mess with the lost database. Infrequent backups? All my adult life I've thought about this stuff on many levels. We tech-types are generally not strong forceful type-A types, and usually back down when higher-ups argue against spending money on safety (whether hardware, software, whatever). I've never understood why I need to be the champion of saving the company's future. I've seen several companies completely fold (after I was out of there). I still can't figure out what happens to mgt. types. They certainly don't seem to learn from business / economic history.
Years ago I was tasked with setting up a server, including a DB (forget which one, possibly DB2). The backup software was amazing- it constantly watched the filesystem and wrote any changes to tape on the fly. No delay.
You also reminded me of 2 developers at that company who spent months on a very complex C/Unix project (industrial controls). Super-brilliant guys. They had lots of notes, but weren't doing backups. Near the end (or so they had hoped) the one guy gives the Unix command to copy the disk to the tape drive... but... reversed it, so he wrote blank tape to the hard disk. Needless to say it took them another 3 weeks working day and night, 7 days, to recreate the whole project. Truly good souls- everyone felt so badly for them, but they regained their good spirits within a day or two as it was all so fresh in their minds.
DROP TABLE reminds me of that case where DB read/writes were part of a URL and in some school district a kid issued URLs with DROP TABLE in them. Oh my. Like with most such stories, the useful details never come out. Like, who did that programming? What company? How is it that we are pretty much forced into at least 13 years of "schooling" but these major life lessons are unlearned?
(Score: 3, Insightful) by bzipitidoo on Saturday April 11 2020, @01:35AM (1 child)
> I still can't figure out what happens to mgt. types.
I also wondered that.
It varies. They can fall from favor, and never recover. It can take more than one disaster. 2nd chances seem to be a bit more frequent for the management class. Even 3rd chances. But it's not many, and 3 or 4 small disasters or 1 big one can finish their careers. I've seen that happen. I know at least 2 ex-managers who were demoted, and finished the last decade or 2 of their working lives on the bottom rung, no one reporting to them, just glad to still have a job, swallowing the humiliation of it all. I also know of a few other ex-managers who would not take the demotion, instead storming out and away to another job where they thought their amazing talents would be better appreciated and utilized. But there's only so much of that any manager can do before it becomes too difficult to explain to yet another prospective employer why they keep changing jobs. And I know of yet another manager who stank it up and lost the contract, and yet was actually rewarded with a promotion to VP. I can only guess he must have had a lot of valuable contacts, and perhaps the move was something of a kick upstairs.
It also depends on how independently wealthy they are. The rich especially buy into the myth that management, particularly upper management of course, is the ultimate in careers, and won't give up trying to manage others no matter how bad they are at it. There are many elite private schools that cater to this thinking. They call it "leadership", because that sounds a lot sexier than "management". They offer a pretty good education, but not for the sake of education, no, education is but a tool, a means to an end. And what end might that be? Being in position to maximize the exploitation for their own selfish goals. Many aren't much interested in education as something pleasurable and virtuous in itself. One indication is that these schools cling to archaic and outmoded education methods that have been shown to be counterproductive, stuff such as shamings and even beatings for getting bad grades. Even if the schools themselves know better, they have to do it anyway because that's the way the parents want it. See, for example, the Robin Williams movie, Dead Poets Society.
These rich leaders can cock it up again and again and again, and super rich Daddy will bail them out almost every time. The terrible ones are masters at self-delusion about how great they are at management, and will have an endless list of tiresomely predictable excuses about why the latest disaster is not really their fault, it's all the fault of their lazy and stupid underlings, and bad luck, cutthroat competition, and the entire body public for not appreciating their brilliant work and lining up to buy whatever it is they were making. And it was a risky proposition anyway.
As to how someone is tapped for management in the first place, that too is riddled with problematic thinking. So often the loudmouthed ignoramus is mistaken as an aggressive go-getter. They undervalue technical knowledge and skills, and overvalue aggression and outright bullying. And you don't really think the upper class twits in charge are any good at discerning who will make a good manager, if they're no good at management themselves.
If you are wondering why so much management is crap, that's why. Not much merit involved in choosing managers. It's also been shown that voters actually don't like a candidate who is too good, too smart. They want their elected leaders to be only a little smarter than themselves. No doubt that preference is also reflected in management.
(Score: 0) by Anonymous Coward on Sunday April 12 2020, @03:14PM
Well I think lots more people will follow the guy confidently saying loudly "FOLLOW ME! I KNOW THE WAY!" even if he's wrong than follow some nerd saying "I think this could be one of the better paths given the little we know at the moment".
What can work is a confident leader sort of person who is humble and wise enough to listen to the smart people and ignore them at the right times - because sometimes you just have to make the call and take a leap (no choice sometimes).
(Score: 0) by Anonymous Coward on Sunday April 12 2020, @03:04PM
Higher spec works:
https://www.oregonlive.com/opinion/2012/08/how_tenacity_a_wall_saved_a_ja.html [oregonlive.com]
But costs more, sometimes not just in money:
How many would do that? I'm not brave enough to say I'd do the same thing. Keep in mind the "pesky" designer was already dead decades before the quake hit.
If they got lucky and there was no such quake for more decades, the 39 feet would have been good enough. If they got unluckier the 46 feet might not have been good enough (but I guess the other measures might have still prevented it becoming a nuclear disaster).
(Score: 2) by Thexalon on Saturday April 11 2020, @02:21PM
It's worth watching Chernobyl for a drama based entirely around the concept of management being stupid and bullying engineers into doing something that shouldn't have been done, only to watch it all go horribly wrong.
The only thing that stops a bad guy with a compiler is a good guy with a compiler.