Why did the Boeing 737 Max crash?
October 21, 2019 1:31 PM   Subscribe

David Perell writes: "The actual story of the 737 Max crash begins with that McDonnell-Douglas purchase in 1997, 21 years before the first accident in late 2018. Unfortunately, media coverage of the crash mostly ignores Boeing’s corporate history. "

This is a deep dive into the corporate transformation from an engineering-first culture to an ironically risk-averse corporate culture that led to the deaths of hundreds of passengers.
posted by jenkinsEar (74 comments total) 31 users marked this as a favorite
 
A somewhat related write-up by Jon Ostrower on the DC-10 (DC is the Douglas in McDonnell-Douglas) and how it's grounding 40 years ago and the erie parallels to the 737 MAX crashes and subsequent grounding.
posted by SirOmega at 1:44 PM on October 21, 2019 [5 favorites]


One of the best analyses that I saw was on Twitter (of all places...)

The threading format of Twitter is super obnoxious for something like this, but the meat of the analysis rings painfully true to anyone in software.

I can see why pretending it's "just a software glitch" is appealing... but really, the software was just the last failure in a long, long trail of institutional, design, engineering, and other failures. It's not the 11th-hour hack with the FIXME comment that's the real problem. It's what got you into that situation in the first place.

In a just world, people would go to jail over this.
posted by -1 at 2:00 PM on October 21, 2019 [15 favorites]


I have seen several version of this "just so" narrative with respect to the 737 max - "Boeing used to innovate, used to do engineering, now it's just a shitty cost cutting corporation." Of course there are any number of cases where exactly this has happened to a company, but without evidence to the contrary I always start from the premise that the author is either afraid of change or really just talking about themselves ("I used to be young once...")

Of course there is all too typical corporate negligence here; but any article on this topic that can't square up deaths per passenger mile with their narrative, which this article does not, is not worth considering.

To me that's what's interesting here - how does this industry, which indeed has been ever more penny pinching on the production and operation side for decades, still manage to keep improving passenger safety? Since the era of corporate deregulation that started in the 80s or 90s I've been wondering if / when that chart would start trending back up, but it still hasn't. Why has the industry been unable to roll back regulation? Is it as simple as international compliance requirements, which means planes need to be built to whomever has the highest standard? Or just understanding that sales will drop if crashes start creeping up again? Or something else, something more complex?
posted by MillMan at 2:19 PM on October 21, 2019 [11 favorites]


A good and clear item of evidence that Capitalism is incompatible with life, either on a large scale or small. We often fail to see these things being inside the storm as it were.
posted by Freedomboy at 2:33 PM on October 21, 2019 [4 favorites]


The threading format of Twitter is super obnoxious for something like this, but the meat of the analysis rings painfully true to anyone in software.

ThreadReader version, perhaps?
posted by hanov3r at 2:35 PM on October 21, 2019 [6 favorites]


To me that's what's interesting here - how does this industry, which indeed has been ever more penny pinching on the production and operation side for decades, still manage to keep improving passenger safety?

But isn't that mostly on the airlines and less on the aircraft manufacturers? Once you've got the major bugs worked out of a class of airframe, isn't the deaths per passenger mile just going to trend downward (as is observed) as passenger miles trends upwards? Seems like the major way that the manufacturers could make that statistic go up is by introducing a new airframe—which seems fully consistent with the lack-of-innovation angle.
posted by BlueDuke at 2:35 PM on October 21, 2019 [2 favorites]


It kinda starts okay, but kind of piddles off into supersonic flight and cityjet boosterism. By ignoring the noise and pollution problems of both — serious human and social issues, things that can't be ignored — his views must be taken through a fairly thick engineer's disease filter.
posted by scruss at 2:37 PM on October 21, 2019 [15 favorites]


One of the best analyses that I saw was on Twitter (of all places...)

Yeah all that stuff lead into each other but the essential problem is that a single sensor is being used to override pilots. That's not to say the behavior of multiple sensors fixes things (AF 447) but if you lower the risk of a sensor being wrong to near zero, the chance of two of three being wrong is astronomical.

If Boeing had three AoA vanes mandatory on its aircraft (pilot, co-pilot, standby) and a visual/aural indicator that MCAS was disabling because it couldn't get two AoA sensors to agree, this would not have happened and MCAS would be a non-event instead of a giant scandal.
posted by Your Childhood Pet Rock at 2:39 PM on October 21, 2019 [6 favorites]


A good and clear item of evidence that Capitalism

If there's one thing that's remarkable about the aviation industry and capitalism, it's the fact that enough capital is made available to develop planes at all, and the fact that it almost certainly involves much greater development investment which is amply paid back in reduced operational costs, as well as greater safety.

It is remarkable, and counterintuitive, that it ends up working that way, but I'd say that anyone who understands the industry would agree with me.
posted by ambrosen at 3:05 PM on October 21, 2019 [2 favorites]


Of course there is all too typical corporate negligence here; but any article on this topic that can't square up deaths per passenger mile with their narrative, which this article does not, is not worth considering.

So wait, you are discrediting the article because it did not mention deaths per passenger mile? In this case, given the graph you linked, wouldn't these 2 crashes be so anomalous as to warrant a bigger picture view?

That graph shows all revenue passenger fatalities from all aircraft, right? So A) the number is expected to go down significantly, as miles flown is surely much greater today than in the 1970's compared to fatalities and B) If you extrapolated all fatalities per aircraft from the total miles/fatalities ratio, what does it tell you?

I am not following your point...
posted by Chuffy at 3:19 PM on October 21, 2019


> Of course there is all too typical corporate negligence here; but any article on this topic that can't square up deaths per passenger mile with their narrative, which this article does not, is not worth considering.

Single source counter point, accident rate is probably more sensible against flights since landing tends to predominate in accidents. When you look at that graph things flattened out in the 70s and have stayed relatively unchanged since then.

https://www.skybrary.aero/bookshelf/books/4239.pdf

Trend is the same, but a lot less striking. By passenger mile would include differences in capacity of planes as well as large scale changes in routes.
posted by scamander at 3:26 PM on October 21, 2019 [5 favorites]


>That graph shows all revenue passenger fatalities from all aircraft, right? So A) the number is expected to go down significantly, as miles flown is surely much greater today than in the 1970's compared to fatalities

The number is fatalities per mile, so if miles increased and nothing else changed, fatalities would increase correspondingly. If originally it was one death per thousand miles, and we increased the miles flown to a hundred thousand, you'd expect a hundred deaths.

But miles have increased hugely and deaths have dropped hugely, suggesting that flying has gotten much, much safer.

The article posits that Boeing has neglected passenger safety and innovation and opted for recycling old designs. But it's hard to square a neglect for passenger safety with a huge drop in the ratio of deaths to miles flown. That evidence would suggest that in fact Boeing's approach has been part of a trend towards passenger safety, not away from it.

It's only one part of a much larger story, but it seems like an important one to include when talking about decades-long trends.
posted by BlackLeotardFront at 3:29 PM on October 21, 2019 [3 favorites]


Twenty five years ago, I worked at a company that supplied airplane parts to Boeing. Only to Boeing, despite 2/3 of the company's plants were literally surrounded by McDonnell-Douglas facilities. This was explained to me on my first day of the job - MD had (already) become MBA'd, while Boeing was still an engineering company.

I guess the prey ate the snake from the inside?
posted by notsnot at 3:32 PM on October 21, 2019 [3 favorites]


The article posits that Boeing has neglected passenger safety and innovation and opted for recycling old designs. But it's hard to square a neglect for passenger safety with a huge drop in the ratio of deaths to miles flown. That evidence would suggest that in fact Boeing's approach has been part of a trend towards passenger safety, not away from it.

We're at zero crash deaths in US commercial aviation for the last eight(?) years. What more can we really want from aircraft design? It's basically making sure we can get better ancillary factors like comfort, fuel efficiency, and cost while not impeding the safety record.
posted by Your Childhood Pet Rock at 3:47 PM on October 21, 2019 [3 favorites]


Double, specifically the second link, to a New Republic article subtitled "How Boeing's managerial revolution created the 737 MAX disaster"
posted by Huffy Puffy at 3:49 PM on October 21, 2019 [5 favorites]


We're at zero crash deaths in US commercial aviation for the last eight(?) years. What more can we really want from aircraft design?

That new designs be at least as safe as old ones, that regulators do their jobs, and that aerospace companies put safety above profits. Seems like a reasonable ask. The 737-MAX 8 crashes happened on foreign soil, but the condition that caused them could have easily happened on a domestic flight, no matter how much Boeing tries to deflect blame onto poorly trained foreign pilots.
posted by qxntpqbbbqxl at 4:07 PM on October 21, 2019 [12 favorites]


300+ counts of criminally negligent homicide, IMHO.
posted by j_curiouser at 4:19 PM on October 21, 2019 [3 favorites]


I'm no aerospace engineer but I'm not sure what there is left to innovate until we solve the energy density battery issue or have a materials breakthrough in strength / weight ratios. Then maybe we get e-planes and maybe a bunch of interesting supersonic and LEO options become economically (and maybe somewhat environmentally) viable.
posted by MillMan at 4:21 PM on October 21, 2019 [1 favorite]


Speaking as a Boeing Baby (my parents met while working there) I’ve spent my whole life immersed in Boeing as an economic driver. A lot of us in the Seattle area, even those who never themselves worked for Boeing, saw the McDonnell Douglas merger as a Bad Thing. I remember the days of “If it’s not Boeing, I’m not going,” which is still true only because there aren’t as many Airbuses in the US as there are Boeings.
posted by lhauser at 5:12 PM on October 21, 2019 [1 favorite]


The number is fatalities per mile, so if miles increased and nothing else changed, fatalities would increase correspondingly. If originally it was one death per thousand miles, and we increased the miles flown to a hundred thousand, you'd expect a hundred deaths.

But miles have increased hugely and deaths have dropped hugely, suggesting that flying has gotten much, much safer.


For measuring aviation safety on a year-over-year basis, fatalities per passenger mile are not the best measure, since as planes get bigger (and they cram more seats into the same-sized planes) the numbers of passengers per flight increase.

Tracing the Wikipedia link back, there's this blog post which has fatal accidents per million flights, which is a much better measure. And this has still gone down. Although a lot of the improvement happened in the 1970s; it doesn't seem to make sense to say the aviation industry is doing well because they improved a lot a long while ago. These numbers also appear to be global; it's unclear to me to what degree the improvements in the recent years may be the result of improving conditions in the developing world to the same standard as in affluent nations, and to what degree the improvements are across the board. But to be fair, the rate has continued to drop, albeit slowly, over the past few years.

The counterfactual is that once problems were exposed with the MAX, they were grounded globally after two incidents. One one hand, I guess that means regulation is working. But on the other hand, a new aircraft hit mass production with a fatal flaw in it; that doesn't seem great. If once a mass shooter had shot two people they were stopped automatically, then mass shooters wouldn't be that dangerous, either.
posted by Homeboy Trouble at 5:28 PM on October 21, 2019 [4 favorites]


FWIW, my f-i-l, who has run one of the contract-out aviation subs for his entire career, was of the exact same opinion as the article back when the 737s were grounded. He's of the opinion that a number of people in the Boeing executive suite and senior engineering need to go to jail for negligence causing death. He's also very unhappy with the regulatory capture that's happened with the FAA in the Bush years that enabled this.
posted by bonehead at 5:46 PM on October 21, 2019 [6 favorites]


A good and clear item of evidence that Capitalism is incompatible with life, either on a large scale or small.
Because nothing was safer than Aeroflot? I think even with the 737 blot on things, flying commercial is way more “compatible with life” than whatever the non-capitalist version is, though I’m happy to be educated otherwise.
posted by Gilgamesh's Chauffeur at 6:36 PM on October 21, 2019 [8 favorites]


We're at zero crash deaths in US commercial aviation for the last eight(?) years.

Ten, unless you count the heart attack suffered by the woman when the Southwest engine exploded, and then we are up to a whopping single death since the Buffalo crash in February 2009.

posted by sideshow at 6:45 PM on October 21, 2019 [1 favorite]


Because nothing was safer than Aeroflot? I think even with the 737 blot on things, flying commercial is way more “compatible with life” than whatever the non-capitalist version is, though I’m happy to be educated otherwise.

Happy to provide topics showing the wonderful things this version of Capitalism gave us...Fracking, opioids, plastics ocean pollution, oil spills, glacier melting, over fishing, and in a tie for everyone's favorite rain forest destruction and/or death of coral reefs worldwide.
posted by Freedomboy at 7:07 PM on October 21, 2019 [3 favorites]


Dumb luck is not a great way to ensure that people don't get hurt. Are people in here really arguing the 737 max is a great product, no need to change anything, business as usual, hooray? I guess the folks in the third world can just fuck right off.
posted by maxwelton at 7:18 PM on October 21, 2019 [2 favorites]


I don't consider it particularly reasonable to use the improvements in safety culture among pilot and maintenance groups at part 121 carriers as cover for increasing deficiencies of the safety culture at manufacturers and, to a lesser degree, management at part 121 carriers, who largely seem to think that training can be less of a priority now that things are "good enough."

The increase in safety can be almost entirely attributed to improvement in procedures in the air to better handle abnormal events, better situational awareness allowing aircraft to avoid risky situations in the first place, and improvements in engine reliability, in that general order. That's why I emphasize the training failures that led to two crews crashing two mechanically sound airliners, not because I believe Boeing's process doesn't need work or deserve criticism.
posted by wierdo at 7:20 PM on October 21, 2019 [4 favorites]


A few other pieces from Matt Stoller, who is writing about from a generally anti-monopoly/regulatory standpoint: The Coming Boeing Bailout? from July 3rd, as well as Boeing's travails show what's wrong with modern capitalism in the Guardian.

A WSJ article he linked out to, which reflects on the problems outside of the planes themselves: Boeing’s 737 MAX Grounding Spills Over Into Economy, Weighs on GDP.

Another pieces, which perhaps provides a response the people who are casting this as speculation: Airline surveys point to ongoing production problems at Boeing’s SC plant. Honestly, I think it's much more likely that what wierdo says above is the case - Boeing has been caught out on a cost-cutting measure can't be compensated for by the advances in other aspects of aviation.

Talking about it as a problem of regulatory capture, like bonehead's father-in-law, seems like the best way to approach it - this is the sort of thing that should never make it to production, basically, and we need more oversight.
posted by sagc at 7:46 PM on October 21, 2019 [3 favorites]


To me that's what's interesting here - how does this industry, which indeed has been ever more penny pinching on the production and operation side for decades, still manage to keep improving passenger safety? Since the era of corporate deregulation that started in the 80s or 90s I've been wondering if / when that chart would start trending back up, but it still hasn't. Why has the industry been unable to roll back regulation? Is it as simple as international compliance requirements, which means planes need to be built to whomever has the highest standard? Or just understanding that sales will drop if crashes start creeping up again? Or something else, something more complex?

FAA regulations have gotten more and more demanding over the years, because no one wants a crash. Airliner crashes used to happen a few times a year, with the big headlines on the newspapers to go with it. They are extremely rare nowadays, to the point where many people have forgotten what those days were like. The industry may not like the costs of safety regulation but they also know that those same regulations keep people willing to fly. Here is a current, and very stark reminder of this.

Just 19 percent of business travelers and 14 percent of leisure travelers would willingly take the 737 MAX within six months of returning to the sky, according to an Atmosphere survey.

Nearly half of the 2,000 respondents said they would pay more to avoid the MAX.


If you’re an airline like Southwest, who has ordered 280 Maxes and is starting to retire some of their older 737-700s, this is a massive problem. It’s been said that Southwest has bet the company on the MAX. The grounding of the MAX has been a major disruption to airlines, and there’s whispers that Southwest is looking into buying Airbus planes now.

Safety is everything for the aviation industry. Boeing decided to go forward with the MAX despite everything they had to work around, and they were hell bent on keeping the type rating. Now they’ve got hundreds of planes idled, and nothing on the drawing board to replace them in the foreseeable future.
posted by azpenguin at 11:21 PM on October 21, 2019 [3 favorites]


A good and clear item of evidence that Capitalism is incompatible with life, either on a large scale or small. We often fail to see these things being inside the storm as it were.

They had plenty of risk aversion and arse-covering in socialist societies like the USSR and China; the difference being that the risk was not being fired/having one's project scrapped but being fatally made an example of as a “class enemy” by rivals/superiors needing to detract attention from their own potential risks.

Perhaps they didn't have risk aversion of this sort in anarchist communes/Thoreauvian back-to-nature utopias whose size never exceeded Dunbar's Number, but they didn't have jumbo jets either.
posted by acb at 2:36 AM on October 22, 2019 [2 favorites]


Although a lot of the improvement happened in the 1970s; it doesn't seem to make sense to say the aviation industry is doing well because they improved a lot a long while ago. These numbers also appear to be global; it's unclear to me to what degree the improvements in the recent years may be the result of improving conditions in the developing world to the same standard as in affluent nations, and to what degree the improvements are across the board. But to be fair, the rate has continued to drop, albeit slowly, over the past few years.

spreading NTSB practices?
The National Transportation Safety Board: A Model for Systemic Risk Management (pdf)
We propose the National Transportation Safety Board (NTSB) as a model organization for addressing systemic risk in industries and contexts other than transportation. When adopted by regulatory agencies and the transportation industry, the safety recommendations of the NTSB have been remarkably effective in reducing the number of fatalities in various modes of transportation since the NTSB’s inception in 1967 as an independent agency. Formerly part of the Civil Aeronautics Board (now the Federal Aviation Administration), the NTSB has no regulatory authority and is solely focused on conducting forensic investigations of transportation accidents and proposing safety recommendations. With only 400 full-time employees, the NTSB has a much larger network of experts drawn from other government agencies and the private sector who are on call to assist in accident investigations on an as-needed basis. By allowing and encouraging the participation of all interested parties in its investigations, the NTSB is able to produce definitive analyses of even the most complex accidents and provide genuinely actionable measures for reducing the chances of future accidents. We believe it is possible to create more efficient and effective systemic-risk management processes in many other industries, including the financial services industry, by studying the organizational structure and functions of the NTSB.
posted by kliuless at 4:26 AM on October 22, 2019 [2 favorites]


Perhaps "blame capitalism"/"no socialism also sucks at planes" is a little simplistic here? For one thing Boeing is massively reliant on government contracts.
posted by aspersioncast at 5:09 AM on October 22, 2019 [5 favorites]


Yes, the only alternative to unfettered capitalism is the USSR or Maoism.
posted by Dysk at 5:10 AM on October 22, 2019


Socialist versus capitalist planes are a red herring here. Automated Gay Luxury Space Communism doesn't actually solve any of these problems as long as corporate liability stays as it is. Someone at McDonnell-Douglas made an explicit decision to prioritize shareholder profits over passenger safety; that person and everyone above him/her on the org chart needs to go to jail.
posted by Mayor West at 5:32 AM on October 22, 2019 [4 favorites]


Not great. Not terrible.
posted by Huffy Puffy at 5:36 AM on October 22, 2019


Make no mistake that the big "capital" behind airframe engineering for 50 years has come from military spending, essentially socialist in all the worst ways, the fruits of which are also applied to civilian aviation. There's nothing particularly capitalist about Boeing's monopoly power, which has been sternly enforced by the US government both in military procurement and through trade policy.

I'm always struck by Boeing's TV ads, as if civilians were buying airplanes or missile systems at Walmart. What they're really on about, as are the drug companies, is the market-based public relations version of legislative capture: media capture. You don't run stories critical of advertisers who spend millions sponsoring your shows.
posted by spitbull at 5:36 AM on October 22, 2019 [7 favorites]


Yes, the only alternative to unfettered capitalism is the USSR or Maoism.

The last biggest fatalities in US aviation, Colgan Air Flight 3407 was a Dash-8 and Comair Flight 5191 was a CRJ-100. Sooooo, Canada is a bust too?
posted by Your Childhood Pet Rock at 5:36 AM on October 22, 2019


For one thing Boeing is massively reliant on government contracts.

Socialism With American Characteristics.
posted by acb at 5:43 AM on October 22, 2019


What they're really on about, as are the drug companies, is the market-based public relations version of legislative capture: media capture. You don't run stories critical of advertisers who spend millions sponsoring your shows.

That flash you saw was the lightbulb going on over my head. Thanks, spitbull!
posted by whuppy at 5:48 AM on October 22, 2019 [4 favorites]


The most recent passenger fatality with a part 121 operator was in a Saab as well, by that measure.
posted by ambrosen at 5:49 AM on October 22, 2019


Yes, the only alternative to unfettered capitalism is the USSR or Maoism.

The alternative to capitalism that can create jumbo-jet-scale infrastructure (i.e., not loosely knit patchworks of anarchosyndicalist communes; the problems assembling a 787 from parts made by a few thousand of those would make Boeing's woes look like a walk in the park) and is reachable from the present day (i.e., not Plato's Republic or Iain M. Banks' Culture or something) is some sort of planned economy with its own hierarchies, internal politics and failure modes.
posted by acb at 6:09 AM on October 22, 2019 [2 favorites]


Meanwhile, another technology bites Boeing back: Stunning messages from 2016 deepen Boeing’s 737 MAX crisis, Seattle Times, Dominic Gates and Steve Miletich, October 18, 2019:
Boeing’s MAX crisis deepened Friday with new controversy around an exchange of bantering texts between senior pilots that suggested Boeing knew as early as 2016 about the perils of a new flight-control system later implicated in two crashes in Indonesia and Ethiopia that killed 346 people.

The exchange of messages in 2016 between the two lead technical pilots on the Boeing 737 MAX program was released Friday after regulators blew up at the company for belatedly disclosing the matter. The messages reveal that the flight-control system, which two years later went haywire on the crashed flights, was behaving aggressively and strangely in the pilots’ simulator sessions.

In the exchange, one of the pilots states that given the behavior of the system, known as a Maneuvering Characteristics Augmentation System (MCAS), he had unknowingly lied to the FAA about its capabilities.
...
Anything you ever say can (and will) be used against you.
posted by cenoxo at 6:49 AM on October 22, 2019 [2 favorites]


Dumb luck is not a great way to ensure that people don't get hurt. Are people in here really arguing the 737 max is a great product, no need to change anything, business as usual, hooray? I guess the folks in the third world can just fuck right off.

No. We are arguing that the regulatory culture of fixing what was wrong with the 737 Max, regardless of profits, is working 100% properly and that this minor accident illustrates that very well. Again, more people in the US have been killed in mass shootings this year than died in those crashes and more people in the US alone have been run over since the planes were grounded in March.

So that regulatory culture is able to cut through relentless cost cutting, technological shortcuts, and capitalistic profiteering and it should be copied for every industry.

If engineering culture actually wasn't a veneer of nonsense then yeah that means your Honda Civic should be shutdown when angry Jimmy runs over someone at a crosswalk until Honda and civil engineers can figure out how to make it not happen again.
posted by The_Vegetables at 7:24 AM on October 22, 2019 [1 favorite]


If Boeing had three AoA vanes mandatory on its aircraft (pilot, co-pilot, standby) and a visual/aural indicator that MCAS was disabling because it couldn't get two AoA sensors to agree, this would not have happened

A less cowboy approach to the design of MCAS would certainly help and frankly I was shocked to learn that any manufacturer, capitalist or otherwise, would ever ever design an active stability system capable of overriding the pilot into an aircraft if it's fed from anything less than triple-redundant sensors and that any safety agency would approve such a design. Aero IT is supposed to be the good kind, not the customary commercial-grade shit that's expected to fail randomly and need turning off and on again.

But if MCAS did shut itself off, the altered aerodynamics it exists to compensate for would surely be quite disorienting to any pilot who doesn't already have simulator experience with them. One would hope that pilot training and qualification programs for the Max are no longer built around a base assumption that it handles just like a 737.

I'm not sure what there is left to innovate until we solve the energy density battery issue or have a materials breakthrough in strength / weight ratios

The mass energy density for hydrogen is pretty spectacular. Pretty easy to imagine a future where the vapour trails are all made of the steam from burning distilled sunshine.
posted by flabdablet at 7:39 AM on October 22, 2019 [1 favorite]


When I was an engineer, I was taught that any change to any part, no matter how minor, required the thorough retesting of the whole assembly before it could be signed off.
That included software products as well as hardware and meatware (users)
posted by Burn_IT at 8:10 AM on October 22, 2019 [2 favorites]


But if MCAS did shut itself off, the altered aerodynamics it exists to compensate for would surely be quite disorienting to any pilot who doesn't already have simulator experience with them. One would hope that pilot training and qualification programs for the Max are no longer built around a base assumption that it handles just like a 737.

Technically, yes but MCAS is supposed to kick in when at the limit of the aircraft's AoA which is where the handling difference mostly lies not in general flight. If the pilot isn't flying at the limit it will feel almost exactly like a 737. MCAS is only to stop the pilot from exceeding a different flight envelope.
posted by Your Childhood Pet Rock at 8:27 AM on October 22, 2019


I keep bumping into this assertion (present in this article) that one of the major motivations for MCAS was to avoid needing simulator time for pilots to qualify in the "new" type. Clearly there's a cost associated there, but I thought commercial pilots had to do simulator time on the order of every 6 months anyway just to "stay current". Why was it viewed as so cumbersome to need new simulator time? Presumably crew scheduling could deal with getting enough pilots qualified on almost their normal schedule as the MAX rolled out through a fleet?
posted by scolbath at 8:35 AM on October 22, 2019


The New Republic article linked above touches on the issue of avoiding simulator time:
Southwest always had a lot to say about projected modifications to the 737, and Kelleher’s team mostly wanted as few technical modifications as possible. With the MAX, they upped the ante: According to Rick Ludtke, a former Boeing employee, Boeing agreed to rebate Southwest $1 million for every MAX it bought, if the FAA required level-D simulator training for the carrier’s pilots.


To whoever agreed to this, the rebate probably seemed like a predictably quixotic demand of the airline that had quixotically chosen to fly just one plane model, exclusively and eternally, where every other airline flew ten. Simulator training for Southwest’s 9,000 pilots would have been a pain, but hardly ruinous; aviation industry analyst Kit Darby said it would cost about $2,000 a head. It was also unlikely: The FAA had three levels of “differences” training that wouldn’t have necessarily required simulators. But the No Sim Edict would haunt the program; it basically required any change significant enough for designers to worry about to be concealed, suppressed, or relegated to a footnote that would then be redacted from the final version of the MAX. And that was a predicament, because for every other airline buying the MAX, the selling point was a major difference from the last generation of 737: unprecedented fuel efficiency in line with the new Airbus A320neo.
posted by Tiny Bungalow at 9:13 AM on October 22, 2019 [4 favorites]


Ah yes, the old "the CEO made a BS frathouse bet with his CEO buddy" school of design. There's yer problem...!
posted by scolbath at 10:01 AM on October 22, 2019 [1 favorite]


Technically, yes but MCAS is supposed to kick in when at the limit of the aircraft's AoA which is where the handling difference mostly lies not in general flight.

Right, but if I understand the mechanics of this correctly, the new engine position on the Max meant that this exact limit was more likely to be hit under full power during takeoff, and this is exactly when the MCAS system tweaks the trim to make the machine fly like a 737 would.

Both the fatal crashes happened because MCAS over-corrected due to trusting faulty AoA sensor readings, but it seems to me that if the new tendency to pitch up at takeoff was not enough to cause a genuine risk of stalling when the Max was being flown by pilots expecting it to behave like the 737, then Boeing would not have needed to put MCAS on the plane in the first place.

MCAS is there to make the new plane behave enough like the old one that Boeing marketing can match Airbus's rather better justified claim that their new plane requires almost no new training.

Vox has more.
posted by flabdablet at 10:03 AM on October 22, 2019 [2 favorites]


Tiny Bungalow: "The New Republic article linked above"

Yeah the New Republic piece has essentially the same thesis as the main link of this post but is about 100x better.
posted by crazy with stars at 10:22 AM on October 22, 2019 [2 favorites]


The mass energy density for hydrogen is pretty spectacular.
It's also pretty irrelevant.
posted by groda at 11:13 AM on October 22, 2019 [1 favorite]


To amplify on that, hydrogen leaks through and can even poison/embrittle metals. Hydrogen is a real problem child when it comes to storage.

Bio/synkerosene is a much better idea. The US military specs it for new equipment now. It's been at least tested in their planes, if not in wide use.
posted by bonehead at 12:19 PM on October 22, 2019 [1 favorite]


hydrogen leaks through and can even poison/embrittle metals

Correct, which is why all of the many existing high pressure hydrogen fuel storage tanks designed for mobile applications are implemented in composites.
posted by flabdablet at 12:28 PM on October 22, 2019


You just build a big flexible fuel tank above the plane and use it for free lift. There’s no downside! The new air transport revolution of the 20’s is coming!
posted by Huffy Puffy at 12:31 PM on October 22, 2019 [4 favorites]


Dropping down in purple bales!
posted by clew at 12:45 PM on October 22, 2019


Hydrogen tanks have historically (experimentally) been installed in the fuselage. I don't know how the logistics of sorting out that level of volume for fuel would work (vs a lower weight of fuel carried), but I suspect that development costs would be considerably lower than developing a new airframe.
posted by ambrosen at 1:19 PM on October 22, 2019


Papers I've seen suggest that a LH2 fueled plane would need to have a longer fuselage than one with kerosene stored in the wings, on the order of 10 to 15% more (in 2010 era numbers). I'm not sure if LH2 is what's being referred to above.
posted by bonehead at 1:40 PM on October 22, 2019


Papers I've seen suggest that a LH2 fueled plane would need to have a longer fuselage than one with kerosene stored in the wings, on the order of 10 to 15% more (in 2010 era numbers). I'm not sure if LH2 is what's being referred to above.

For the same carrying capacity, yes. Both LH2 and 700 bar hydrogen have far less volumetric efficiency compared to kerosene. They weighs much less for the same energy but they take up more space.
posted by Your Childhood Pet Rock at 2:51 PM on October 22, 2019


That's always been the concern that I've seen. Hydrogen itself is really efficient, but it needs containment that masses more than a kerosene tank. IDK if carbon composites are enough to overcome that problem now, but its been a major factor in the past, with the older generation fibreglass/metal composite tanks.

I'd still be worried about leakage and aluminum embrittlement of the fuselage though.
posted by bonehead at 3:42 PM on October 22, 2019


Yes, the (lack of) density of hydrogen makes it a not great choice when aircraft are already about as long as they can reasonably get without causing all kinds of issues and the pesky physics of fluid dynamics that make frontal area the enemy of fuel effciency.

Happily, making Jet A from non-fossil carbon isn't terribly expensive (equivalent to ~$80/bbl crude) or even enormously energy intensive, and it can be made from things that would otherwise release methane during decomposition, further reducing overall emissions. It's not sexy like electric power, but it's something we could do tomorrow if we really felt like it, and probably would be already if we had managed to impose even a minimal cost on fossil carbon emissions.
posted by wierdo at 6:48 PM on October 22, 2019 [1 favorite]


My son just got his commercial pilot's license. Not sure how that relates to the post, but I am certainly reading the articles and the comments closely.

I used to be a non-nervous (good?) flier, but as I age, I am not sure any more. With all the glitches in software, the beta versions as actual releases, I am deeply afraid of autopilot. I just envision my 25 yo daughter, a freshly minted programmer, tapping away at her keyboard writing the software.

I sincerely hope that whatever method is used to test the software is good enough to detect whatever flaw might show up at 30,000 feet.

I know that I am way over simplifying this, but maybe not. I don't like when my life is in someone's hands that has no skin in the game other than money.
posted by AugustWest at 12:00 AM on October 23, 2019


making Jet A from non-fossil carbon isn't terribly expensive (equivalent to ~$80/bbl crude) or even enormously energy intensive, and it can be made from things that would otherwise release methane during decomposition, further reducing overall emissions

The aeronautical equivalent of “power your electric car with a solar panel on the roof” may be “power your airliner with biofuel from the onboard lavatories”.
posted by acb at 1:27 AM on October 23, 2019 [1 favorite]


i say we power the cars with the biofuel from the airliner bogs and power the airliners with solar panels on the roof
posted by flabdablet at 1:55 AM on October 23, 2019 [1 favorite]


I'm aware of at least one company trying to use solar/wind to drive renewable jet a synthesis from a carbon capture process. One of my former students is working for them. I've seen the fuel and have been trying to get a sample to test, but no joy so far.

They're still working on input costs, but the economics, given an expected 20-year pay out, don't look insane. I think it's a very realistic possibility in a few years. Solar farms for airplane fuel could convert entire fleets to zero net emissions with no technological changes needed on the user side.
posted by bonehead at 7:15 AM on October 23, 2019 [1 favorite]


The more renewable generators get built to power fuel manufacturing plant, the better off the whole grid will be as well; fuel manufacture can be designed in ways that just don't need to care at all about hour-on-hour energy availability as long as week-on-week energy availability is enough to hit production targets.

Paying fuel manufacturers to time-shift production is a completely viable peak-shaving strategy and potentially far less costly than either storage or on-demand peaking generators.
posted by flabdablet at 9:58 AM on October 23, 2019


I sincerely hope that whatever method is used to test the software is good enough to detect whatever flaw might show up at 30,000 feet.

Second-hand, anecdotal, sample size 1: this department of the airplane manufacturer was pretty good at testing whether the software met the specification, and no good at noticing when the specification was not fit for the software's purpose.

Which to be fair is always, always the hard problem in validating software. But it can be made much harder when you wall "testing" into a separate team, throw a specification over the wall, and stonewall any concerns they raise.
posted by away for regrooving at 10:43 PM on October 23, 2019 [1 favorite]


That said, testing does have its own culture and its own skillset, and specialist testing teams do tend to grow some pretty sensitive bullshit detectors. It kind of comes with the role. Far more common for the C suite to ignore a testing team's recommendations than for the team itself to miss something important. When the internal MCAS memos eventually leak there's going to be a hell of a stink.
posted by flabdablet at 6:08 AM on October 24, 2019


Watch live: Boeing's CEO testifies before House on 737 MAX crashes, Washington Post, 10/30/2019. Boeing Chief Executive Dennis Muilenburg testifies before the House Transportation and Infrastructure Committee about the safety of the 737 MAX that was involved in two fatal crashes that killed 346 people.
posted by cenoxo at 10:17 AM on October 30, 2019


GE to lose $1.4 billion this year from Boeing 737 Max grounding but expects jet’s return, CNBC, 10/30/2019:
KEY POINTS
  • GE manufactures the LEAP engines used for Boeing’s 737 Max airplane, which has been grounded worldwide following two deadly crashes in the past year.
  • “We still expect this year to be impacted to the tune of about negative $1.4 billion” from delays in GE engine production for the 737 Max, Chief Financial Officer Jamie Miller told shareholders.
  • However, GE expects to continue to ramp production of its LEAP engines, as Boeing assumes the 737 Max will begin flying again before the end of the year.
Details in the article.

WRT Boeing CEO Dennis Muilenburg's House testimony earlier today, his previous testimony from Tuesday, 10/29/2019 can be seen at Boeing CEO Dennis Muilenburg testifies before Congress on 737 Max, KING 5 News (NBC Seattle).
posted by cenoxo at 10:45 AM on October 30, 2019


Related Boeing articles at KING 5 News (NBC Seattle).
posted by cenoxo at 7:02 AM on October 31, 2019


The emotional power of photography: Families confront Boeing CEO Dennis Muilenburg on Capitol Hill with photos of 737 MAX crash victims, New York Post, Lia Eustachewich and Aaron Feis, October 29, 2019. (Lead photo by Michael Reynolds, EPA-EFE.)
posted by cenoxo at 7:54 AM on October 31, 2019


The New Yorker weighs in.
posted by mwhybark at 2:09 PM on November 11, 2019


Is there anything new in that New Yorker (/ProPublica) piece? I skimmed it and it looked like it was same basic story (merger with McDonnell Douglas, change in culture, MCAS) already covered by others. The main addition was some reporting on one of the handful of Americans who died in the crash -- a young, attractive, well-connected and rich white blonde woman.
posted by crazy with stars at 9:41 AM on November 12, 2019


She was Ralph Nader’s great niece; that is the primary point of the story. The writer also kinda takes a shot at Langewiesche, and takes him and Boeing to task on the poorly-trained pilots deflection.
posted by mwhybark at 10:03 AM on November 12, 2019 [1 favorite]


« Older Cake village   |   How To Radicalize A Normie Newer »


This thread has been archived and is closed to new comments