Slashdot Mirror
Boeing Delays 737 Max Software Fix (arstechnica.com)
Boeing's promised software fix for its 737 Max planes involved in two deadly crashes since October has been pushed back several weeks after an internal review by engineers not connected to the aircraft raised additional safety questions. "The results of the 'non-advocate' review have not been revealed, but the Federal Aviation Administration confirmed on April 1 that the software needed additional work," reports Ars Technica. From the report: "The FAA expects to receive Boeing's final package of its software enhancement over the coming weeks for FAA approval," an FAA spokesperson said in a statement. "Time is needed for additional work by Boeing as the result of an ongoing review of the 737 MAX Flight Control System to ensure that Boeing has identified and appropriately addressed all pertinent issues." Just how far back the delivery of the MCAS patch has been pushed is uncertain. The New York Times reports that the update's schedule has been pushed back "several weeks." And after its delivery, an FAA spokesperson said, "the FAA will subject Boeing's completed submission to a rigorous safety review. The FAA will not approve the software for installation until the agency is satisfied with the submission."
This means it could be months before grounded Boeing 737 MAX aircraft are once again deemed airworthy. And that means more flight cancellations for airlines that have the aircraft in their inventory. Southwest Airlines, Boeing's largest 737 MAX customer, canceled all of its flights dependent on its 34 737 MAX aircraft through April 20 so far -- about 150 flights per day. And Boeing's delivery of new 737 MAX aircraft -- the company's best-seller -- has been indefinitely delayed.
This means it could be months before grounded Boeing 737 MAX aircraft are once again deemed airworthy. And that means more flight cancellations for airlines that have the aircraft in their inventory. Southwest Airlines, Boeing's largest 737 MAX customer, canceled all of its flights dependent on its 34 737 MAX aircraft through April 20 so far -- about 150 flights per day. And Boeing's delivery of new 737 MAX aircraft -- the company's best-seller -- has been indefinitely delayed.
right... and some idiots are out there panting over how much we need to ban driver controlled vehicles and we must move to AI/Autonomous vehicles ASAP!!!! This right here is why we shouldn't /sensor issue was worked out/tested and approved before the Dept of _____________ signed off on it. horse hockey.. no thank you.
can you imagine if your car was "grounded" for months while a safety computer control
If internal reviewers are brave enough to point out flaws with this huge amount of pressure, it must be a really bad mess. Or they actually have some engineers left that found a backbone and are unwilling to be responsible for hundreds of people killed, no matter what management wants.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Indeed. Not how anybody sane designs systems that can kill lots of people in accidents. The whole thing was completely botched, and the motivations was plain old-fashioned greed and arrogance.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
Forget about this. EU will definitely revenge for WV dieselgate, but everything else will stay where it is now.
Look at Intel's Meltdown and Spectrum processors vulnerabilities, all exciting processors produced since 2005 are basically must replaced as soon as possible. Do you see anybody to sue Intel? The answer is "no", so it is the same for Boeing.
I am more amazed by the fact that shitty social media companies like Snap-chat manage to make IPOs for 30Bln when Boeing capitalization is just 150 Bln. Is it Ok?
As someone that has worked in both functional safety and off-highway vehicles.
How the fuck did this ever make it into production. Why is a 'second sensor' an upsell?
When given the option to completely update the cockpit to the latest and greatest with digital displays.
They chose to replicate the old mechanical dials so the pilots couldn't be retrained.
The entire thing from start to finish was rushed. Mechanical design comes first. There is no 'try and develop software in parallel'. A clean software design depends on a good mechanical design.
The plane should have been a white board redesign, it should have been balanced such that a pilot could fly it stable with no avionics. This isn't a jet fighter.
But it was rushed because Europe invested in R&D and beat them to economy routes. How much money did Boeing C-suites make before 2011? During the 2009 crash there was a hiring spree by some companies because the market was flooded with cheap, good engineers that just got laid off. Companies invested in talent. Did Boeing?
People died because... Boeing sat on R&D from post WWII while making a ton of money so when Airbus released a good plane they scrambled to retrofit an old design by putting huge engines on an airframe causing it to pitch up but to appease its clients it added software to mimic the old plane behavior and tested it themselves and told the FAA they promise they did it right.
More or less.
Who will write down the loss the 150 flight per day canceled for months. I assume airlines are insured for that risk, but still, the cost will be close to the billion USD.
MAX hidden settings
From what I understand, the planned change involves adding one more trim sensor and leaving the pilot to notice a "disagreement" light in the middle of trying to keep a flying bucking bronco stable. It's almost like they're ASKING for another major crash.
The CEO should step down. On top of the crashes, the pay-for-safety thing is squarely on him: he either signs the final order, or if somebody else did it under his nose, he's incompetent for not watching the ship.
On a different note, what would happen if the Max is grounded for several months or years? Are there enough other planes to cover the load, or will rates go way up?
Table-ized A.I.
They added Clippy: "It looks like you are battling an aggressive autopilot. Would you like some help?"
Table-ized A.I.
Forget about this. EU will definitely revenge for WV dieselgate, but everything else will stay where it is now.
Look at Intel's Meltdown and Spectrum processors vulnerabilities, all exciting processors produced since 2005 are basically must replaced as soon as possible. Do you see anybody to sue Intel? The answer is "no", so it is the same for Boeing
Wait, what does West Virginia have to do with Diesel? I thought that was Coal Country and they were just waiting for Trump to bring back steam locomotives to make rail great again?
Except mcas was designed not as an anti-stall mechanism exclusively but rather a software solution to make this abomination of an aircraft fly move like the NG models. The design was so shitty from the start they had to install a new system and keep it quiet from pilots. Further more this is not just a simple trim runaway like pilots have been trained on, MCAS runs the trim a whole lot faster than standard trim inputs, do this during takeoff when there's a lot going on and you can quickly find yourself in an extremely bad situation. It's a shit plane that Boeing rushed to the market in order to compete with the Airbus A320Neo. This pile of shit should have never been drafted on paper much less frankenstined onto a 60+ year old design like those 707's you reference.
In any normal flying, the trim is just a help. So even if this system trims badly, it should just annoy the pilot that has to countermand it. But apparently that was not possible, and that seems to be the real design issue. Nothing to do with sensors.
It would be like having automatic lane sensors in a car. If they go wrong the driver should be able to just grab the wheel and override them. The driver should NOT have to read some checklist in the manual to figure out which buttons to press to disable the system all while the car is heading towards a tree.
Only 8?
Confucius say, "Find worm in apple - bad. Find half a worm - worse."
Boeing is about to learn how the cheap becomes the expensive.
In between the 2 Boeing crashes, another crash was avoided only because there just happened to be another pilot who was in the extra cockpit seat. He knew and understood this flaw, and was able to instruct the 'real' pilots on how to keep the plane from nosediving into the ground like the other 2 did. If not for this, random informed off-duty pilot being in that cockpit, we would have had a third tragedy.
How many hardware engineers does it take to change a light bulb?
None - we'll fix it in software.
How many software engineers does it take to change a light bulb?
None - we'll document a workaround.
How many tech writers does it take to change a light bulb?
None - the user can figure it out.
So in this case we have:
How many hardware engineers does it take to not crash an airplane with a faulty sensor?
None - we'll detect and avoid it in software.
How many software engineers does it take to not crash an airplane with a faulty sensor?
None - the pilots can be trained to disable the auto-trim mechanism.
How many trainers does it take to not crash an airplane with a faulty sensor?
None - the pilots can figure it out.
How many pilots who have no idea why the plane is reacting as it is does it take to crash an airplane with a faulty sensor?
None - the plane will do it for them.
(No, this isn't meant to be funny.)
Quattuor res in hoc mundo sanctae sunt: libri, liberi, libertas et liberalitas.
The reason aircraft software should always be able to be overriden by the pilot is that software can never be proven to work as specified.
In computability theory, the halting problem is the problem of determining, from a description of an arbitrary computer program and an input, whether the program will finish running (i.e., halt) or continue to run forever.
Alan Turing proved in 1936 that a general algorithm to solve the halting problem for all possible program-input pairs cannot exist.
Honestly the whole 737 scandal only raised a little concern. You make products for long enough one will crash and burn. Thinking that you can actually design, produce, and QA, and begin delivering a solution which lives depended in a matter of weeks, is a institutional cultural problem that will take a major effort to overhaul. Agile has its uses, this is a abomination of what looks to be a corporations attempt at it.
Agree, what a screw up it is, just today I read about the latest accident, that MCAS was engaged several times just before the crash even after it had been deactivated.
Always fond of Boeing, fully aware that nobody's perfect (Airbus also had their issues with overriding pilots decisions - yet not on such a scale), honestly have to say that this crack in their reputation cannot be easily repaired.
Finally about the way it was handled: FAA let Boeing engineers verify their own work, reports about MCAS problems were coming long before accidents, but it took 2 crashes and hundreds lost lives to do something about it - disgusting - and the whole reason for MCAS was to make the new plane feel like the old one, because they didn't bother to properly design it.
"He knew and understood this flaw"
That pilot did not know and understand the flaw. He knew that if you worked through the trim excursion checklist you eventually hit something that resolved the symptom.
You're correct. Had the jump seat pilot not been present, the Lion Air plane (the first crash) would have crashed the day before. If that happened, I strongly suspect it would not have crashed again the next day. :)
Yes, the AOA sensor was not repaired properly on the Lion Air plane, it was put back into service, and the plane crashed. Unlike the Ethiopian airlines plane, the pilots were not aware of the MCAS system.
The jump seat pilot properly identified a runaway stabilizer condition (caused by MCAS), and by following the procedure for runaway stabilizer (which was unchanged from previous 737 models), prevented the crash.
-=Lothsahn=-
That pilot did not know and understand the flaw. He knew that if you worked through the trim excursion checklist you eventually hit something that resolved the symptom.
EVERY pilot is supposed to understand that if you are experiencing a runaway stabilizer that disabling the electric trim system is how you stop it. That's part of their training. It's part of the training beginning with the first contact with an autopilot that has an electric trim system. It's part of the recurring training in the simulators as ATP-level pilots go through their corporate check rides. Before I was ever let loose in an airplane with such an autopilot the CFI made me show him every one of the half a dozen ways of disabling the autopilot, including pulling the circuit breaker for the trim system. And amazingly enough, it's part of the preflight checklist, too.
It's not a case of "eventually hit something", it's a part of the design of the system and part of the emergency procedures that every pilot ought to know.
The important question is not why the dead-head pilot knew it, but why the six other pilots involved did not. No, it's not Boeing's fault that they weren't trained on dealing with this failure, it was in the POH from the beginning.
It's pretty much a given Boeing will to lose this one big, and pay big. But that cost will be small compared to the customer defections, which have already started.
It's beyond me why anybody would think it's a good idea buy an obsolete deathtrap 737 in the first place.
When all you have is a hammer, every problem starts to look like a thumb.
EVERY pilot is supposed to understand that if you are experiencing a runaway stabilizer that disabling the electric trim system is how you stop it.
Fuck off. The trim should not have run away, on top of the whole sad fiasco starting with the stubby landing gear. Boeing put the flight crew in a position where they needed to solve a puzzle in one minute or die in a huge fireball. They didn't figure out for whatever reason and hundreds died. Boeing did that.
Listen, just fuck off asshole. Boeing will pay, but it won't bring those people back to life.
When all you have is a hammer, every problem starts to look like a thumb.
And all 737s should be junked. Short ones are ancient and long ones are deathtraps.
When all you have is a hammer, every problem starts to look like a thumb.
Fuck off. The trim should not have run away,
Thank you for your kind words. Of course the system should not have failed. But, being a physical device, designed by humans, built by humans, systems CAN fail. And knowing that the system CAN fail, the designers put in a method of disabling the effects of the failure. Screaming that the system should not have failed is simply childish and immature, like the rest of your language demonstrates.
If the stabilizer trim runs away and there is no way to stop it, then yes, there is a problem with the design and outrage would be justified. But since there is a documented and practiced method to disable such a failure, outrage at the manufacturer is misplaced.
on top of the whole sad fiasco starting with the stubby landing gear.
The landing gear has nothing to do with MCAS or a runaway stabilizer emergency procedure.
Boeing put the flight crew in a position where they needed to solve a puzzle in one minute or die in a huge fireball.
And Boeing provided a method of disabling the runaway stabilizer trim that takes just a few seconds to accomplish. Prior to following the standard emergency procedure, pulling back on the yoke takes just a second and would stop the descent.
If you are going to be outraged at how Boeing put these pilots in such a bad position, then you need to be outraged at EVERY aircraft manufacturer, because EVERY aircraft manufacturer of similar aircraft has an electric trim system that can run away.
You clearly are not a pilot, or not a pilot who has transitioned from anything more complicated that a C152 or other trainer. If you had, you would know about electric trim failures and how easy it is to stop them dead without dying yourself.