The fix for the 787s Lithium battery problems wasn't a whole lot better. Remember those exploding left and right and the 787 fleet grounded?

The fix wasn't to reengineer the large amount of electrical systems, lack of bleed air or use of different batteries.

The fix was to put them in a stronger container with a hole in it so they can vent outside when they inevitably blow up again.

That has worked well. So far.

Kind of - https://aeronauticsonline.com/united-787-battery-failure-pro...

I'm with Boeing in this fix. The problem is not with the "the large amount of electrical systems, lack of bleed air". This is how things evolved.

The main issue IIRC was manufacturing defects in some batteries that caused internal short circuits.

> The fix was to put them in a stronger container with a hole in it so they can vent outside

And this is an acceptable fix (beyond better QA/manufacturing practices of the battery). If the batteries fail the fault should be contained.

The article you linked explains the changes Boeing implemented.

Yes, batteries failed. The battery failure was contained correctly. Besides improving the reliability of the battery (which was addressed with some changes) that's what Boeing should have done (and did).

They're not going back to NiCa or Lead batteries

Also, unless you can guarantee the failure rate of the battery to be low enough, those incidents made it clear the battery needed to be better contained—even if you also decrease the failure rate.

That's not exactly true as per your link. They did make changes to the battery, as well as outside the battery.

Similar thing there - the model was delayed and thus likely high pressure and incentives internally to get on with it. According to the article, a series of fires didn't stop Boeing from allowing the plane to keep flying or to be sold, but FAA had to step in and ground it. Then, I can imagine, the cost of [FAA grounding] > [properly fixing it] means it got fixed. At least till the cost comparison tilted the other way.

That’s horrifying.

I’ve also heard that the 787 was a boondoggle.

The 787 is weird (and possibly unsafe) for at least two significant reasons.

Firstly, Boeing totally re-invented their approach to assembling planes. Previously they built most of the aireframe components and assembled them in their own plants. But the 787 was specced out during an ongoing Boeing/Airbus anti-trust dispute over government cross-subsidies that turned nasty (escalated to USA/EU level, mediated via the WTO, with sanctions threatened). For strategic reasons Boeing decided to outsource subassembly production to third parties and retain only overall control and final assembly (the goal was to have 51% of the work go through the EU, so that the EU would shoot themselves in the head if they applied sanctions against Boeing aircraft). (Similarly, Airbus did much the same: you can buy A320s—the 737 series' Airbus rival—built entirely in China, for example.)

Anyway, this resulted in huge cost overruns, delays as small suppliers grappled with trying to supply components of reliable quality and consistency in Boeing-scale quantities, and so on. On the other hand, it gave Boeing a shiny new 21st century supply chain which they're using on new projects, so the $30Bn wasn't entirely wasted on just one bird.

Secondly, Boeing went full-tilt for composites in place of aluminum as the main structural material.

Now, Airbus had been building composite airframes for decades at this point, but they got there incrementally, and not without fatal accidents. See for example the crash of American Airlines flight 587:

https://en.wikipedia.org/wiki/American_Airlines_Flight_587

It turns out that switching to all-composites from aluminum has some serious gotchas, including entirely different wear/fatigue/delamination issues.

Boeing has done surprisingly well up to this point with the 787 (although the early Ethiopian Airlines hull loss ground-side due to a runaway battery fire should be a warning—it was a shiny new plane and the battery fire burned through the top of the hull, turning it into a $200M write-off; an aluminum bird could probably have been patched and repaired). But we have no idea whether Boeing managed to avoid baked-in long term weaknesses in the airframe, of the kind that only show up 10-30 years into an aircraft's life. Such weaknesses can show up suddenly and kill an entire type within a matter of months (see for example the fate of the Vickers Valiant, a British strategic bomber of the 1950s that succumbed to cracked wing spars and was hastily scrapped).

Anyway: this is why I'm very dubious about flying on 787s, at least for another few years. Your mileage may vary, obviously ...

The 787 is not weird. It is an evolution of practices and learning from across the industry, academia and governmental organizations (NASA, ESA, etc) that culminated in a new product. The 100% composite structure has been completely flawless for the most part, with the largest issue being how to recycle the plane when it's done. Boeing has long outsourced the production of many components of all of its planes to various vendors, the big difference with the 787 was that they did this on a global scale.

The use of LiIon batteries was a logical next step as well. Manufacturing issues in the first sets of batteries required them to re-engineer the safety system around them, and the vendors that created the batteries now have stricter standards to adhere to.

None of these issues currently compromise the safety of the airframes.

It's easy to pile-on after incidents occur, but these are machines made by people. People make mistakes all of the time. You wouldn't not use a computer because it had bugs. The likelihood of you falling out of the sky in a modern airliner is still vanishingly small and is getting smaller every day.

>The 100% composite structure has been completely flawless for the most part

Doesn't refute GP's point. All the 787s flying now are basically new planes. If they have issues that only appear after 10 years of use, they will all be "completely flawless" until they start undergoing unplanned disassembly while on the air.

Right, we don't know what the long-term usage of these composites in commercial aircraft will result in, in terms of service-life. I'm sure Boeing have their projections, and I'm sure they're optimistic. I think the larger firms will sell off their aging stock to developing nations as they approach end of life. Those nations might be inclined to extend the life of their air frames past original specifications due to economic reasons.

We know the FAA rubber-stamps whatever Boeing claims at this point, so don't expect regulators from keeping your plastic plane from disintegrating mid-flight.

>American Airlines flight 587

The accident investigation found that the vertical stabilizer was stressed beyond its certified load limits by the pilot's rudder movements.

I can see why people were inclined to blame composites at the time of the crash. But 18 years have passed now, and there was never another incident where an A300 had a structural failure of the vertical stabilizer.

There's a good comment here: https://forums.jetphotos.com/showthread.php?50480-Airbus-Com...

> vertical stabilizer was stressed beyond its certified load limits by the pilot's rudder movements.

That's something you don't expect to be possible on an Airbus.

Why not? The Airbus A300 has fully mechanical linkage, and none of the fly-by-wire systems introduced later.

Precisely because Airbus was an early adopter of fly-by-wire systems and envelope protections.

I find your first point much more relevant than your second. Boeing had been using composites extensively in military aircraft since at least their work on the B-2, and had been using composites in smaller areas from well before then. That was all was before they purchased Northrop Gumman, which brought considerably more expertise in house.

According to your "weird fatigue / sudden catastrophic failures" theory, you should be very comfortable flying on a 787 today, and become less so as the airframes age. And definitely stop flying 787s the moment one breaks up midair.

This is not the case. The 737 MAX has different handling characteristics to a regular 737, but there is nothing inherently unsafe about those handling characteristics. The purpose of MCAS was just to prevent the need for pilots to recertify by masking the difference.

Without MCAS a MAX has flight characteristics that would not pass FAA certification.

I don't believe that's true. What's your source?

There are specific regulations on how the stick force must change in response to various flight conditions. Without MCAS the stick force curve of the 737-MAX does not satisfy those regulations.

Do an HN search on "MCAS stick force curve" (without the quotes)[1] and you can find some discussion of this with cites to and quotes from specific sections of the regulations.

[1] https://hn.algolia.com/?query=MCAS%20stick%20force%20curve&s...

There is no source in those comments for the claim that the 737-MAX would not satisfy these regulations without MCAS.

Note that the actual regulations are a tad more complex than the incomplete sentence quoted: https://books.google.co.uk/books?id=5tAhOC4KC8MC&pg=PA200&lp...

I'm one of the originators and principle repeaters of the theory, and the various facets that go into it are admittedly built up via synthesis of many different slices of life.

The principle source that moved me in the direction that the MAX may not be certifiable without MCAS was from this interview from the Royal Aeronautics Society with D. P. Davies, a well known test pilot for the U.K. ARB, and the past woes that Boeing has had to navigate with their designs to be flown in the U.K.

https://www.aerosociety.com/news/audio-the-d-p-davies-interv...

The 727 had difficulties getting certified for use in British airspace due to high lift device interactions that tended to cause undesired extra pitch sensitivity near critical AoA. The 727 was eventually certified under protest conditional to the addition of a stick pusher to remedy the behavior. The approach was preferred because it both made it harder to stall the plane, and made it readily apparent to the pilot that "thar be dragons here". This makes apparent to me, that at a minimum, it is favorable to design a system to cope with aberrant behavior in such a way that the pilot is immediately aware of the fact they are in a perilous part of the flight envelope.

The certification was controversial amongst test pilots because it was seen as as the beginning of a slippery slope that would eventually normalize airframes that don't actually present sound aerodynamic characteristics without excessive assistance from systems that are liable to break. Also note, the FAA never required Boeing to install stick-pushers, and several deep stalls did end up occurring.

Since then, many aircraft have become notorious for accidents/difficulties caused where this slippery slope approach to aerodynamics was embraced. See the MD-11 LSAS, and now the MAX 8. Both of these were 'hacks' put in place to make aerodynamic changes to the airframe, while minimizing retraining requirements. The sales points being that the new plane flew just like the old plane.

I'm not sure that MCAS is first and foremost an artificial feel system. I've categorized it as more of a "certification hack" in light of whistleblower testimony, and conclusions I've come to doing calculations and attempting to cobble together simulations myself.

I do know, however, that if the aerodynamics are so severe near critical AoA, that mere elevator actuation pressure adjustments are no longer adequate to prevent the plane from flinging itself past critical AoA at the merest nudge, and that the entire horizontal stabilizer needs to be moved to ensure the nose is brought back down in short order, I am not confident that the raw math to draw out the stick force curve will end up drawing a compliant curve.

And I know that it is not acceptable for a pilot familiar with a sluggishly handling aircraft to just hop into a jitterbug and be on their way without training; I have a hard time being sold that having the same plane go from "smooth and steady" to "put your nose in the air, and stall it like you just don't care" in the same civil transport airframe should be acceptable either. That seems a clear violation of the intent of that design directive.

If there's info/precedent/experience missing you can offer to refine or refute it, that'd be awesome. I'm just grabbing on to as many facts as I can run down, throwing them against reality as best I can find equations or regulation to describe it, and trying to see what sticks.

I'm more interested in finding the truth than my theory being right.

> the entire horizontal stabilizer needs to be moved to ensure the nose is brought back down in short order

Doesn't MCAS take a long time to significantly move the stabilizer? Like tens of seconds? It's hard to square needing to decrease AoA so dramatically with the actual mechanics of MCAS.

MCAS has a duty cycle of on 10 seconds, off five. It ramps up the degrees per activation up to a max of 2.5 degrees per activation (2.5 degrees per 10 seconds), all the way to max down trim if you let it.

Keep in mind, the entire horizontal tail plane moving has a heck of a lot more influence on attitude than the elevators.

If it were a stringed instrument, trim would be the tuning tabs at the top of the neck, the elevators would be the fine adjustment screws on the body of the instrument.

Thanks. I understand that the stabilizer is powerful. I'm wondering whether it's true that, when your plane is in that danger state of "flinging itself past critical AoA", 2.5 degrees over 15 seconds is sufficient to avoid or recover the wing stall in time to avoid terrain.

(I think the proposed software changes also reduce this control authority but I'm not sure how.)

You're right, I thought I read a post that had a citation, but I cannot find one.

> there is nothing inherently unsafe about those handling characteristics

this is not entirely true. applying throttle to the max causes a pitching momentum, something you don't want during a deep stall when control surfaces are totally unresponsive, because it would make the exit procedure (throttle up, nose naturally pitches down, gain airpseed and regain control) unsafe (throttle up, whops the plane is now pointing upward and going backward)

Applying throttle to any airplane with engines under the wings causes a pitching momentum. The pitching momentum is different in the MAX than other aircraft, but that doesn't make it less safe than anything else as long as a pilot is trained to understand it. The MAX is not naturally more likely to enter a stall regime than its predecessors.

well yes, but there's more to a big plane stall. wings washout makes them stall at the root first, so that the center of lift moves backward, slightly helping overcome the engine momentum

moving the engine forward increased such momentum

large planes are finely balanced machines, it doesn't take much to get outside their flying boundaries, wether authority or loading, especially because the trim on them has buttload more authority than the elevator controlled from the yoke

I don't think recovery from deep stalls is actually much of a safety issue with airliners. Are there any instances where an airliner has deep stalled and then successfully recovered? By definition, a deep stall is one where standard stall recovery procedures won't be effective.

> much of a safety issue with airliners

yeah because they tend to pitch down way before reaching a full wing stall, but the max pushes the partial stall deeper.

They lengthened the plane without redesigning the rest of the components to compensate. This changed fundamental aerodynamics on how it flies. The plane is dynamically unstable and that instability was masked in software.

>They lengthened the plane without redesigning the rest of the components to compensate.

Not exactly. It's the placement of the newer, larger engines that's largely responsible for the different handling characteristics.

>This changed fundamental aerodynamics on how it flies.

Yes.

>The plane is dynamically unstable.

No, it just has different handling characteristics in some parts of the flight envelope.

Dynamically unstable means that it will leave controlled flight if it does not receive control input, and that the departure from controlled flight will accelerate after it starts, therefore requiring constant computer-assisted control inputs. Pretty sure this is not the case, but it will have different and potentially dangerous behaviors for some speeds, angles of attack and thrust settings.

That's literally what happens to the 737 max airframe. Specifically, the angle of attack is prone to increasing at high angle of attacks.

That's not dynamic instability.

Are you going to provide any source for that?

http://nymag.com/intelligencer/2019/03/is-the-boeing-737-max...

If (as appears to be the case) you regard any random person on the internet as a source, then sure:

https://www.quora.com/Are-any-commercial-airliners-other-tha...

The sources I linked to:

http://nymag.com/intelligencer/2019/03/is-the-boeing-737-max...

https://spectrum.ieee.org/aerospace/aviation/how-the-boeing-...

These are international, reputable organizations unlike a Quora thread. That would be like me using reddit as a primary source. Get real. I am going to believe the reputable news sources, over random people on the Internet.

And two can play at that game:

https://www.quora.com/How-long-do-you-think-it-will-be-befor...

https://www.quora.com/Should-divergent-instability-be-allowe...

Can you actually get a 737 MAX in controlled flight, into a state where its angle of attack will keep increasing faster and faster without changing the control inputs, until it stalls? That would be dynamic instability, albeit only during certain situations.

You don't know there's nothing inherently unsafe about the handling characteristics, that's a Boeing talking point. You're not qualified to make that determination, and the regulators didn't bother to make it either.

> The purpose of MCAS was just to prevent the need for pilots to recertify by masking the difference

This is a Boeing talking-point as well. The real purpose of MCAS might be to make an otherwise wildly-unsafe plane flyable, and without it it's a disaster waiting to happen. Again, we don't actually know how safe the plane is without it.

I mean, sure, there might turn out to be safety issues that we don't yet know about. You could say that about any plane.

>The real purpose of MCAS might be to make an otherwise wildly-unsafe plane flyable

MCAS would never activate in a typical flight, so I don't think it can be that bad.

There's a simple and plausible explanation for why MCAS was added (the desire to maintain the same type rating). That explanation doesn't entail that the 737 MAX had fundamentally unsafe handling characteristics, only that its handling characteristics differed significantly from those of other 737s. I'll stick with that explanation absent evidence to the contrary.

> MCAS would never activate in a typical flight, so I don't think it can be that bad.

Also untrue. There have been plenty of flights where MCAS has activated (either American or Southwest released some data about the number of MCAS activations they had without crash incident). We (the public) don't have any data regarding a flight where MCAS would-have activated but was disabled. In the cases where we know MCAS was disabled, those aiframes are now destroyed.

> There's a simple and plausible explanation for why MCAS was added (the desire to maintain the same type rating).

The benign name "Maneuvering Characteristic Augmentation System" is in itself a Boeing-talking point. What it it was called the "Thrust-Input Anti-Stall System?". It's completely a Boeing talking point that why MCAS was added, it might not actually reflect reality. We don't have any incentive to trust Boeing or the FAA in this case, in fact, we have just the opposite, they appear to have been deliberately misleading.

> I'll stick with that explanation absent evidence to the contrary.

We already have evidence Boeing withheld the amount of authority it gave MCAS counter to what they told the FAA. We already know the FAA didn't actually review the software implementation. We know Boeing and the FAA didn't initiate any meaningful investigation after the first Lion Air crash. WTF should we believe a single word they say? They're liars.

I think you're in conspiracy mode on this one. Boeing has shown incompetence and stupidity, and there's a good chance that someone should be held criminally liable. But I see no reason to doubt what Boeing says about why MCAS was added. What would lead me to doubt it is a more plausible alternative hypothesis, which you haven't provided.

> There have been plenty of flights where MCAS has activated

I'm sure there have, but as I said, it would not activate at all on a typical flight.

It has though. There have been several articles about reports filed by pilot's where uncommanded nose down occurred, that the root cause was not ascertained for. This is reported at a minimum here.

https://theatlantic.com/article/584791/

These are by definition "typical flights" until the anomaly occurs.

Also, keep in mind, AoA sensors have long been a luxury for pilots, since they tend to learn the envelope in terms of glide slope, and they are trained to be able to safely operate aircraft without it.

The MAX took a historically non-critical component, and added it to a safety critical process.

I understand that with the sheer number of different points in the timeline of the design, deployment, and operation of these craft, that trying to hold all the context in your head at the same time does make it feel like

"Damnit Jim, we've entered crackpot territory!"

Nevertheless, we have to remain vigilant and open to the fact that malfeasance does happen, and those that would commit it also go out of their way to make it as hard to bring into the open (if intentional), or frequently disincentivize any activity that could potentially make the issue easier to observe since the path of least resistance typically by definition involves minimizing points of inspection or other forms of scrutiny (if the malfeasance is merely the result unintentional misguidedness/incompetence).

That's categorically not true.

https://spectrum.ieee.org/aerospace/aviation/how-the-boeing-...

737 Max experiences dynamic instability, where a high angle of attack is prone to increase the angle of attack at an increasing rate.

That is not dynamic instability.

Yes it is: http://nymag.com/intelligencer/2019/03/is-the-boeing-737-max...

There are many sources that reach the same conclusion.

No, dynamic instability refers to planes that require constant control inputs to avoid departure from unstable flight.

There are no expert sources that describe the 737 MAX as dynamically unstable.

Not many people would be in a position to know this. Random aviation "experts" certainly aren't going to know the detailed flight envelope characteristics of a particular aircraft model.

I literally linked 2 articles with expert opinion describing 737 Max airframe experiencing dynamic instability. You have not provided a single source and appears didn't even bother to look at my sources.

Hacker News is a place of discourse. If you think I'm wrong and my sources are wrong, point out specifically where and provide counter sources to validate and support your position.

No one is going to take your word for it.

The article you link in this thread reports comments from this person:

https://en.wikipedia.org/wiki/Peter_Garrison

How would this person know in detail the aerodynamic characteristics of the 737 MAX?

The other article you link gives no source for its claim, expert or otherwise. You should also check the comment by Jake Brodsky on that one.

Ad hominem. Both articles were posted under the banner of internationally known and recognized news sources. Whereas in stark contrast you posted a quora thread with 3 replies.

100% agree. Anytime you are using software to hide a larger problem, that is a huge red flag.