Southwest Airlines Boeing 737-8 Max Experienced Dutch Roll | Hacker News Reader

Southwest Airlines Boeing 737-8 Max Experienced Dutch Roll

avherald.com

258 pointsbelter228 comments

Years ago, my father in law ( a heavy jet mechanic) got sent to China for two months as part of a team to diagnose and fix a recurring dutch roll complaint on a particular airframe. This was for a A300-series flown by a large freight carrier.

The local mechanics couldn’t figure it out, so this team was assembled from the mechanics at headquarters to fix the plane.

They inspected the plane and found no damage. They had the pilots make multiple test flights - they could not reproduce the problem. The pilots were adamant that the plane was broken. Engineers from Airbus ended up flying out to assist. They couldn’t identify any problems.

After a few months of effort and a lot of very expensive dart throwing and parts swapping, they gave up. As far as anyone could tell, there was never anything wrong with the plane, and based on comments overheard from the pilots, they simply did not like this particular plane. It had been pulled from a boneyard and recertified, and they considered flying it beneath them.

FIL did come home with some cuban cigars and really high quality counterfeit Fluke multimeters he got from the local market, though.

Eek… I’d never want to fly a plane that’s beneath me. I’d prefer it to be at the same altitude.

Well unless you're an air launched cruise missile operator.

#UnexpectedDrStrangelove ?

Technically any plane should be beneath you. Also above you, but beneath you too.

Mitch Hedburg? Is that you?

> and really high quality counterfeit Fluke multimeters he got from the local market, though.

I got one of those years ago. Wouldn't trust it on any industrial shit but I've used it for residential electrical tinkering (plus smaller projects) for years, works great and has never had a single issue.

These were used for mechanic-y stuff like basic voltage and circuit draw tests. They were fine for that, and the main benefit was that they were made with oil/solvent resistant coatings on the meter body and leads.

> they simply did not like this particular plane.

…and employed a face-saving (though probably very expensive) way of communicating it.

I suspect the face saving culture is holding China back significantly since it adds a ton of noise to interpersonal signal propagation. I wonder if the government is doing anything social engineering-y to try to reduce it?

The government is among the worst of the culprits. “We’re being accused of doing a bad thing? I know! Arrest the speaker and erase it from our version of the Internet!”

seriously. Look at the Kardashians. They turned every embarrassing event into a money making opportunity.

You think the Kardashians are a good model for government?

Of course not. I meant it as an extreme counter-example.

> I wonder if the government is doing anything social engineering-y to try to reduce it?

I know Xi Jinping launching an anti-corruption program was considered a pretty big deal, because it meant admitting that corruption exists in the CPC.

That isn’t true though. All presidents launch anti corruption campaigns to solidify their power when they start. Xi did it, Hu did it, Jiang did it, I bet Deng did it (although the office was merged yet, he was at least supreme leader). When they are all corrupt, it is easy to (a) clean house of people you don’t like (b) protect and promote your supporters and (c) scare lots of monkeys by killing lots of chickens.

The CPC always admitted there was corruption, to do otherwise would simply lose credibility with people. Given that Xi is now president for life, the corruption should simply be accumulating now since the 10 year house cleaning cycle has been broken.

I think this was way more calculated than that. They made up an issue they knew would be taken very seriously and that would be difficult - in the best of circumstances - to disprove.

  "Dutch Roll is a coupled out of phase movement of the aircraft as result of weakened directional stability (provided by the vertical tail and rudder), in which the aircraft oscillates around its vertical as well as longitudinal axis (coupled yaw and roll)."

That sounds pretty scary. I wonder why it wasn't bigger news when it happened last month?

I wonder if the name is related to the Dutch Tilt camera shot - https://en.wikipedia.org/wiki/Dutch_angle

Tha name is believed to be borrowed from an ice skating term:

https://en.m.wikipedia.org/wiki/Dutch_roll

https://www.iceskatesmuseum.com/e-disc-schoon.htm

Here's a video demonstrating it:

https://youtu.be/Zmjam1evDD4

I guess it’s worse than it looks? The video was kind of a dramatic let down to be honest. I preferred the mystery.

Yeah, could this be caused by turbulence? Winds?

Looking at the video, now i'm not sure I haven't been through some dutch rolls also.

The FAA reported: "AIRCRAFT EXPERIENCED A DUTCH ROLL, REGAINED CONTROL AND POST FLIGHT INSPECTION REVEALED DAMAGE TO THE STANDBY PCU, OAKLAND, CA." and stated the aircraft sustained substantial damage, the occurrence was rated an accident.

Standby PCU sounds like a control system. Was that a cause or effect of the event?

The PCU in this context is the part that moves the rudder. PCU issues were common on the 737 in the 90s[0] although that's... probably?... not relevant here.

[0] https://en.wikipedia.org/wiki/Boeing_737_rudder_issues

Exactly, rudder control issues could trigger or exacerbate Dutch roll. Loss of yaw damper control, in other words (yaw damper actuates the rudder).

Dutch Roll is a coupling of yawing and rolling dynamic modes, and is a product of the aircraft's aerodynamics. If the aircraft is disturbed off a steady-state path either by control input, changing winds, or turbulence, then it should return back to it's steady-state path with oscillations that quickly dampen. Dutch Roll is a phenomenon where these oscillations grow rather than dampen as a result of out-of-phase yaw and roll modes.

So Dutch Roll can be triggered by turbulence/wind, but the Dutch Roll itself is the result of something going wrong in reaction to that stimulus. This is different than the aircraft just being batted around by turbulence.

> This is different than the aircraft just being batted around by turbulence.

Does it actually look different or is it just a different cause for similar movement patterns?

Just being batted around by turbulence looks different if you know what you're looking for (although what to look for might be only obvious when looking at accelerometer data). Again, Dutch Roll is a very specific phenomenon as a result of coupling between the roll and yaw dynamic modes. The risk of Dutch Roll is that these oscillations can grow even without further stimulus rather than just dampen out.

For what its worth, that's very unlikely in the last decade or so. More likely to be moderate turbulence, which would be nearly indistinguishable in a non-severe event.

A real-world video looks a lot scarier.

Linked downthread: https://www.youtube.com/watch?v=2tgfkGiHhxs

Especially if close to the ground.

It looks a bit more intense in real footage.

https://www.youtube.com/watch?v=2tgfkGiHhxs

Agreed.

This should be the top comment.

Yeah, this has some elevation reference, and much more dramatic shifting.

this video posted elsewhere here seems dramatic enough

https://www.youtube.com/watch?v=2tgfkGiHhxs

Imagine you are in the cockpit at the top of the plane and your view/perspective - for a layman like me this would seem and feel like the flight is completely out of control. Even the passengers are bound to experience dramatic movements similar to severe turbulence I guess.

Idk. The idea of trying to control the aircraft which is doing this while being responsible for >100 lives seems pretty terrifying to me.

Dutch roll was particularly pernicious with trijets. It has to do with thrust vs lift vectors having different offsets (origins) relative to the cg. Sometimes they would add stabilizers to the tail to correct for this, but eventually manufacturers gave up on the design.

I feel like “roll” is a bad word for what’s happening here. In my brain, when I hear “roll”, I’m thinking a 360 degree rotation in either the vertical or horizontal axis (or both).

This is far from that (even in the real life footage linked in the sibling comments). This is more like a “Dutch wobble” or “tilt”.

But I’m not an aviator, so what do I know.

What you describe might be called a loop. In aviation, “roll” describes any rotation (not necessarily full 360) around the longitudinal axis, as “yaw” around the vertical and “pitch” around the lateral axis.

ETA: 360 degrees pitch is a looping, 360 degrees roll is a aileron roll or slow roll, a 360 degrees roll and pitch is a barrel roll, and 360 degrees yaw is “a 360”, ie a full circle turn.

"Roll" is the name for one of the three axes in aviation. "Vertical" and "horizontal" aren't very descriptive in three dimensions.

You could reasonably call yaw the "horizontal axis", but then assigning "vertical" to either of roll or pitch – and what do you call the other one then? They're arguably both vertical, depending on which side you look at the plane from! Additionally, at least "horizontal" implies an Earth-centered focus, which doesn't help while in, say, a barrel roll :)

Best to avoid the ambiguity entirely and use specific terms, just like how port and starboard avoid the "my left/right or yours" ambiguity nicely by always referring to the ship's frame of reference.

In any case, it sure beats a Dutch crunch.

It can torn off the engines from the main body structure: https://asn.flightsafety.org/asndb/333926

While true, this was a pretty different situation in that it was 1959 (engineering standards have changed a lot in the last 65 years), and the dutch rolls were being performed intentionally for demonstration purposes.

> dutch rolls were being performed intentionally for demonstration purposes.

…with the yaw dampers (that are there to prevent Dutch roll) turned off.

1985 the dedliest single aircraft accident was due to dutch roll. Japan Airlines Boeing 747SR.

2013 KC-135 Stratotanker broke up in flight due to dutch roll.

2005 Airbus A310 survived with serious structural damage.

> 1985 the dedliest single aircraft accident was due to dutch roll. Japan Airlines Boeing 747SR.

This kind of buries the lede, right? A bad repair caused an explosion that blew off most of the vertical stabilizer (ie, the tail). The dutch roll was part of the series of unfortunate events that followed.

Of those three examples, only the KC-135 was caused by the dutch roll. The other two were caused by structural failures, the dutch roll was incidental.

This is extremely disingenuous. A dutch roll if identified and corrected is not structurally dangerous to an aircraft but it typically signifies something wrong with the control surfaces which is a larger issue. In the majority of the cases, it's the yaw damper that's a problem (my suspicion in the 737 case).

JAL 123 crashed because hydraulic pressure on all 4 hydraulic lines and 90% of the vertical stabilizer and 100% of the rudder were lost due to an explosive decompression of the aft pressure bulkhead. Dutch rolls ensued because of the loss of lateral direction control.

The KC-135 crashed because the rudder power control unit on the rudder was faulty and the pilots failed to identify the problem. They then used alternating rudder inputs to recover which caused the structural limits of the vertical stabilizer to be exceeded its structural limits and separate (along with the rest of the tail).

The Air Transat flight (961) had the entire rudder separate from the aircraft most likely due to stress fractures. This caused the aircraft to have extremely limited lateral directional control which caused the dutch rolls.

> engineering standards have changed a lot in the last 65 years

Especially for Boeing where standards are determined by the stock price

Maybe it's a feature, not a bug? Eventually, enough parts fall off that a whole new plane appears on the ground. It's just how the 737s reproduce.

It may have gotten lost in a sea of Boeing bad news.

> why [Dutch Roll] wasn't bigger news

I think the name is a serious problem. Something like “uncontrolled aircraft orientation” is more descriptive. “Dutch Roll” brings to mind thoughts of “is it better than Swiss Cake Roll?” and “is this news-spam for Hostess brand emerging from bankruptcy?”

There are terms of art, jargon, in many fields that mean something different than the “obvious” reading of the term. I had a discussion with someone on Hacker News recently that just could not understand that “dynamic programming” was a very specific solution technique for a broad range of problems with a special structure, rather than any sort of programming with a dynamic aspect to it.

Agreed. To the question at hand: “why 737-max’s ‘Dutch Roll’ wasn’t bigger news,” my conjecture is the term of art obscures the severity of the problem.

As a point of jargon, “Dutch Roll” is poor because the word roll within aeronautics could refer to an orientation movement or an acrobatic maneuver (and probably other things I don’t know about). The use of an ethnic/national identity as an adjective doesn’t help. Is “Dutch” associated at all with oscillations in two dimensions that are not intended by the pilot?

In plain English: unwanted tail waggle

"Tail lefty-righty no-no" is the industry term.

That's plane English.

You're on a roll

Not as scary as flying upside down which is what I originally thought based on title alone. Most people are familiar with "barrel roll" but by comparison this just looks like wiggling.

Counterintuitively though my understanding is that barrel rolls are generally safe even for large planes since it's a stable 1G maneuver (some jock irresponsible air force pilot used to regularly barrel roll B-52s until he did one too low and killed himself and the crew)...

AFAIK only one Boeing passenger jet ever performed a barrel roll (without passengers on board): https://simpleflying.com/boeing-707-barrel-roll-seattle/

> barrel rolls are generally safe even for large planes since it's a stable 1G maneuver

Maybe "safe" here means that merely the stress of the maneuver itself wouldn't destroy the craft? But surely the loss of lift is likely to put it at significant risk, as in your example.

You dont lose lift in a barrel roll, you merely change the direction that lift is being generated.

When the threat is the ground, maybe that's what's worrying anyway.

Ah, good point ;)

That depends on what you mean by “generally safe”. Airline jets are not made to withstand the forces of an aerobatic maneuver and while they can do it in test flight conditions, each time requires pulling the jet out of service afterwards for a full structural inspection.

A barrel roll is a 1-G maneuver and will not overstress an aircraft unless done improperly.

Up to 2 or 3G, but most aircraft should withstand that, even gliders.

I'm not aware of a B-52 being crashed in a barrel roll, but this did make me think of the 1994 Fairchild crash[0] that was caused by a pilot who was known as dangerous due to excessive self confidence. In this case he banked too tightly at low level causing a stall into the ground. The whole crash was caught on camera[1], which is indelibly burnt into my memory.

[0] - https://en.wikipedia.org/wiki/1994_Fairchild_Air_Force_Base_...

[1] - https://theaviationgeekclub.com/the-story-of-bud-holland-the...

Man, that is just one big ass airplane to even consider wanting to do any kind of maneuvering without threat indicators blinking/buzzing/beeping/toning that your death is imminent without doing some of that "pilot shit".

The story from your [0] about one of the crew's last flight with his family on the ground waiting for the post-flight rituals where his wife and 2 kids watching is just horrific. What a fucking asshat of a person.

I'm pretty sure that's the crash GP was referring to. It sort of looks like he could have been going into a barrel roll.

The last 3 seconds of your life as the copilot must have been some revolting emotion knowing that the overconfident asshole in the seat next to you just killed you.

Dutch roll, Bud Holland... Netherlands trifecta is in play.

Thanks, my mistake, this is what I was referring to. Apparently "Bud" would brag about "rolling" B-52s (https://www.historylink.org/File/8716), but the fatal crash was actually just a low-altitude stall as you pointed out. It's now a textbook case study for systemic leadership failure.

The Fairchild AFB crash was not a barrel roll.

> it's a stable 1G maneuver

It’s positive G, but 0.5 up to 2G (otherwise, given that lift is “towards the ceiling of the plane”, so not always straight up during the manoeuvre, you’d accelerate downward, and arresting the fall requires >1G).

Probably wasn't a B52. Tex Johnston was reknowned for barrel rolling a 707 during a demo back in the day - there's grainy video of it on youtube. Apparently the Boing execs at the demo didn't know it was going to happen and nearly had heart attacks when he did.

Video on youtube: https://www.youtube.com/watch?v=iz1Eze3QrDU

Tex Johnson

Thanks, corrected.

Bob Hoover rolled a plane while pouring a tea.

https://www.youtube.com/watch?v=V9pvG_ZSnCc

Tex flew in cowboy boots and, well, was named Tex. I don’t know what they thought would happen.

A barrel roll or an aileron roll? Most people call an aileron roll a barrel roll (no thanks to Star Fox)

Just flew a Max from Denver to Mexico City a couple days back.

It was mostly uneventful, but as we were preparing to land the aircraft banked hard to the left and began to experience something akin to this...like it hit a sudden pocket of turbulence at the apex of the turn and began to pogo around.

It was minor and far from inducing damage to the vehicle, but in that moment remembered where I was while clenching my body for a few nervous seconds until we eased out of the turn.

Do you fly to Mexico city often? The altitude makes landings unusually eventful.

This is my third time and I'm familiar with altitude weirdness having lived in Denver for several years, that rollercoaster effect.

The other parts of the descent were smooth until specifically hitting its maximum bank, then it suddenly got squirrelly and then stopped as the plane began to flatten. It's as if the degree of bank was beyond the plane's limits.

I've flown hundreds of times and can't recall ever experiencing something quite like this.

EDIT: I just realized my original comment was written incorrectly. The plane didn't expectedly bank, it's that this started happening at the top of a hard bank.

I had a similar experience as a child flying into Phoenix... there were heavy storms and about 100ft from the ground, the plane just shifted to the side several hundred feet... the pilot pulled back up and had to circle around the airport for another landing approach. Kinda wild.

> like it hit a sudden pocket of turbulence at the apex of the turn

I wonder if this could have been wind shear? Wind shear was deemed as the cause of the crash for Delta 191[0] on approach to DFW in the 1985. Lots of new equipment was installed after to detect wind shear. I wonder of MEX has that kind of equipment.

[0] https://en.wikipedia.org/wiki/Delta_Air_Lines_Flight_191

Possibly, I don't think so. It happened right at the top of the bank and stopped as the plane began to ease out.

There were no weather issues no other incidents of turbulence or jostling around. It seemed as if the plane was having problems banking at that angle.

Sounds quite a bit like wake turbulence. As planes move through the air, they leave air vortices called “wake” (much like a boat does in water). This effect is worse when airplanes are slow, like when they’re coming into land. Pilots try to avoid each other’s wake, but it’s invisible and sometimes you do hit it if the controller vectors you close behind another aircraft.

It produces a strong rolling motion and can even flip small airplanes.

I hope nobody got hurt (assume so, you may have told you otherwise). Is this something pilots get trained for how to cope with this type of incident?

What happened to passengers (are they usually injured from things falling/flying around)?

Boeing is getting some bad news these days, but multiple whistle-blowers in a row is the worst.

It wasn't that extreme, it was what I would consider a medium level of turbulence. I just realized my comment wasn't written correctly - the bank itself is what caused the issue, not that it just suddenly started banking unexpectedly. What was interesting was that it specifically started and stopped as it eased into and out of the turn.

The question is how bad the oscillations were, and how long they lasted. Maybe for the passengers it was indistinguishable from "regular" turbulence?

E.g. wibble-wobble or the more pronounced wibbly-wobbly.

"Dutch Roll" in context: it's one of the dynamic modes of a plane -- a bit like the "phugoid" (e.g. when your paper airplane repeatedly speeds downward, pitches up, stalls, and speeds downward again on its way to the ground).

In broad strokes, the dynamic modes can have natural frequencies -- in the same sense often used when speaking of resonance, transfer functions -- and there's a trade-off between (1) having low natural frequencies and (2) having a responsive control system.

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

Not the incident from the article, but this footage of some other aircraft doing a Dutch Roll is illustrative:

https://www.youtube.com/watch?v=2tgfkGiHhxs

You gotta love the first comment under that video.

The 737, at least in classic guise, was traditionally very stable in Yaw. Unusually so because the swept wing jets usually require additional yaw damping because the geometry naturally means on yawing one wing gets longer and the other gets shorter as seen from the point of view of the airflow. So much so that the Yaw damper on the 737 was not on the minimum equipment list for dispatch. On other aircraft like the 757 having one of the two yaw dampers out of service limited dispatch to return to a maintenance base (from memory so it might not be exactly that).

I'll keep my eye out for the report on this one because I'm not convinced the initial reporting tells the whole story. It's not unheard of for pilots to damage otherwise perfectly serviceable aircraft through in-proper rudder application - American Airlines Flight 587 being a horrible example.

Four minute video illustrating a dutch roll:

* https://www.youtube.com/watch?v=9Gt-IcCBiQ4

Longer 15m video from Mentour Pilot with some more details:

* https://www.youtube.com/watch?v=z2J_LUDWioA&t=1m

First video is very good for a clear explanation, second one to get a bit more comments from someone who knows.

I was about to link the same video. It was fascinating and took some of the scare out, but if I'm ever in one, now I'm going to be sweating it that the pilot knows what he/she is doing and doesn't overcorrect.

Sounds like a Yaw Damper failure? https://en.wikipedia.org/wiki/Yaw_damper

Not dangerous per se but very annoying for passengers

For those unfamiliar with Dutch Roll -

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

Noteworth there: The 2013 fatal crash of a Boeing KC-135R, after its rudder PCU failed. The article places preliminary blame (for the new incident) on the 737-8's rudder PCU.

One might wonder about the design, components, and processes commonalities between rudder PCU's for the KC-135R, and 737-8.

The 737 rudder PCU also had a defect in the 90s that caused a series of major incidents, although obviously that's been fixed since:

https://en.wikipedia.org/wiki/Boeing_737_rudder_issues

What the news about Boeing has done is make me pay more attention to commercial aviation in general. The number of aviation incidents is actually pretty crazy, and no, Boeing is not at the center of them despite what the media would have you believe. What is amazing though is the actual injury and fatality rate is really low due to well trained crews and measured responses by air controllers.

> the actual injury and fatality rate is really low due to well trained crews and measured responses by air controllers.

It’s the crew and ATC staff, definitely, but also

- the amazingly well engineered airplanes (despite recent Boeing hiccups - just goes to demonstrate the high standards) with many redundancies;

- the archaic, but robust ATC system; and

- last but not least the excellent safety culture, which investigates problems thoroughly to the root, but without blaming individuals.

Boeing could still argue "you see, it's safe, redundant systems worked, we haven't killed anyone this time"

The word that you are looking for is resilience I believe. Commercial and GA aircraft are highly resilient, and the procedures built around that industry are highly resilient as well.

Assuming one follows the procedures... Boeing...

What really blows my mind in the number of near-misses on active runways (at least within the US). I don't know if the trend is up, as I don't have historic data, but certainly looking at Youtube (VASAviation, etc), there are a scary number of them (2 within the last 45 days at one of my local airports).

I watched this video recently and found it really informative w.r.t. commercial aviation near misses:

https://www.youtube.com/watch?v=mIiPt1YVkP8

It gives just enough detail around the classification system, rate of incidents, root causes, etc, that I could search for and learn more on my own after.

Potentially dumb comment, but how long until AI replaces - or at least complements - ATC personnel?

As a Dutchman the endless list of apparent “Dutch” things never ceases to surprise me. Uncles, courage, ovens, double, going, etc., and a new one to my list: rolls. And none of them means any good.

As a new resident, I also find it surprising! It's a shame that most of the substitutions seem to be negative however. Lots of really great stuff and people in NL <3

Don’t forget Angles.

dutch babies are a pretty decent desert!

Half of them refer to "German" things, where the word "deutsch" underwent some pronunciational variation.

Actually, a lot of these pejoratives are from the 19th century. By the 17th century, the word "Dutch" had begun to solidify its meaning around the Netherlands.

This mainly stems from a time when tensions between the British and Dutch were at their peak, due to competition.

> And none of them means any good.

Dude, Dutch ovens are awesome. One of the most useful cooking tools imo.

Dutch oven also means trapping your partner or kid under the blanket and farting. lol

A Dutch book can be good depending on which side of it you are on!

This just isn’t Boeing’s decade.

Feels like IBM or Oracle.

Until the McDonnell Douglas management retires, Boeing will not recover. It's not a matter of an unlucky decade, it's a matter of replacing engineering with MBAing. It's just that the latency between cause and effect are long on the engineering side of this industry.

The merger was 27 years ago. While it may have been the cause at the time, there’s no more McD to blame. Anyone who was a senior leader in 1997 has long since retired. This is Boeing and only Boeing’s problem.

Those guys are top tier you have to admit— tank their company, get acquired, keep their jobs, eventually wrangle their way into managing their acquirer, tank the acquiring company… this is god tier MBAing.

I didn't know what it was called, but I've experienced mild Dutch Roll enough on flights, usually on the approach or landing, that I didn't realize that it was all that rare or dangerous. I had come to think of it as a normal result of wind shear. I imagine it's a matter of degree, and I wish there was some mention of the magnitude of the Dutch Roll in this case vs what is considered normal or acceptable.

(Edited to qualify my experience as "mild".)

How did you tell the difference between dutch roll and the plane catching a glideslope?

I'm going to to laugh at the bad takes on this thread, and await a more accurate report.

Dutch roll is undesirable, but its a basic consequence of some converging aeronautical effects.

Yes but the alarming thing here is the damaged PCU...

Sure. but that's just like, OK, that by itself means very little.

It could mean the standby PCU somehow activated by itself, maybe damage causing a un-commanded activation, maybe software bug, maybe God said fuck this plane in particular.

The point is we have no idea since there are not enough details to know. But of course, on HN we have a army of people who think they know stuff because Wikipedia.

https://trends.google.com/trends/explore?q=dutch%20roll&date...

Yep, exactly what I suspected.

Thing is this could just keep going. There’s no telling how many bad apples Boeing put out there

If it's a software issue, I'd think owners would be pretty clear on its status on each plane.

But for physical defects, yeah that's scary.

>If it's a software issue, I'd think owners would be pretty clear on its status on each plane.

Ha. I have some bad news for you:

https://www.bloomberg.com/news/articles/2019-06-28/boeing-s-...

I can't read that article, but I'm assuming (hoping?) that even if the software is flawed, there's still clarity about which source code / build configuration is installed on a given plane.

Contrast that to the allegedly faked test results, and known defects, of the planes' physical characteristics.

The source article was talking about how their software practices were so bad they’d hire $9/hr engineers to work on core avionics stuff.

As for your point: There’s not much contrast. MCAS was software, after all.

Only upside to software is because it can be updated (maybe), it’s at least more likely to be fixed instead of covered up.

> If it's a software issue, I'd think owners would be pretty clear on its status on each plane.

That assume that it is a known software issue. If it is an issue with some unknown triggering factors how would they be clear on its status?

Consider that it’s also now true that “normal” incidents among the 100k+ daily commercial flights are now newsworthy if they involve a Boeing plane, when they wouldn’t have been a couple of years ago.

You’re still far safer in any commercial flight than walking the dog in your neighborhood.

Looking at flightradar24, the only real upset I see is the turn over the southern Sierra Nevada range which is also ~55 minutes before arrival into OAK.

Wonder if mountain waves play any part in this incident? It's the only thing I can think of that could possibly generate a dutch roll without continuing flight stability issues. (They could have also turned on/off something like the yaw damper in the cockpit but that's not specifically called out int he report)

I can't imagine the standby PCU causing this incident unless there was some kind of electrical or pneumatic issue causing it to engage.

But 737 + PCU issues (yes, I know it's been redesigned) is never a good day.

Hopefully we'll get a NTSB report on this one.

For those of you that are more visual, here is a terrifying YouTube video of what Dutch Roll looks like:

https://www.youtube.com/watch?v=2tgfkGiHhxs

Not bigger news due to severity of issue, most likely. Safety report out today?

Edit: yep, faa confirming today

https://m.youtube.com/watch?v=Zmjam1evDD4

This is what a Dutch Roll looks like: https://www.youtube.com/watch?v=Zmjam1evDD4

This seems to really under-sell the severity; the real-world example someone else linked is much more dramatic. https://www.youtube.com/watch?v=2tgfkGiHhxs

Wow that's an intense example.

That's not real world, though, that is a demo flight where they induced a strong dutch roll on purpose.

It kinda bothers me that airlines are still placing orders for new Boeing airplanes.

Taking delivery of old orders is one thing, since they can't really back out of those, but placing new orders implies that some level of trust remains for this company, and I just don't see how that makes any sense at all?

Boeings are still flying, and Airbus has no availability. On the neos, they have more than a decade worth of orders, and the order book A350 is not much shorter.

Boeing has many years of backlog, too, this is really normal with airplane manufacturers. The airlines commit ahead, they wouldn't actually want to take delivery today anyway, so the manufacturing system is designed to work this way.

Boeing does not have a decade of backlog.

Seems like you haven't seen the latest news.

https://www.washingtonpost.com/business/2024/06/11/boeing-sa...

New orders are significantly down.

Old orders will rarely be cancelled even if possible simply because there's no alternative. Airbus's backlog is longer than Boeing's. And while currently Airbus's manufacturing rate is much higher currently, it won't make a difference if everybody switches to Airbus.

> New orders are significantly down.

For Airbus, too.

I'm not sure what you mean. I'm not aware of any significant drop in new orders for Airbus. Do you have a source?

"Boeing sales tumble as company gets no orders for 737 MAX for 2nd straight month" - https://news.ycombinator.com/item?id=40649943

Institutional momentum at airlines is huge. Southwest flies only boeing to minimize training requirements and keep pilots interchangeable. Now they are appreciating the riskiness of that strategy in hindsight.

1) Even if the max had hiccups, plenty of them, it is an absolutely safe airplane by any measure.

2) Your fleet may mostly be formed by Boeing airplanes, having all of those move to Airbus airplanes is also a safety issue. Pilots have to re-learn and even forget things they accrued and evolved to react instantly for decades. At some point you need to weigh the odds of increasing human error vs relatively safer Airbus alternative

3) Maintenance and logistics conversion would be a huge pain

4) If you order an A320 today, you're gonna get it no sooner than in a decade. Not taking your ordered Max is really not a smart idea.

5) Boeing hasn't received a single new order for max airplanes in two months

> 1) Even if the max had hiccups, plenty of them, it is an absolutely safe airplane by any measure.

Except fatalities.

"Some of you might die but that's a sacrifice I am willing to take." - some Boeing suit

Even by fatalities.

737 Maxes have flown more than a couple of million of flights, at least, since they entered service, that puts their fatal incident rate at similar levels of the A320.

The only two incidents, also, as you know, were related and both happened 5+ years ago.

That’s comparing apples to two sacks of apples, one of which is 30 years older.

The A320neo entered service in 2016 and has over 3,000 of the type in service and has zero fatal accidents involving passengers or flight crew (there have been five fatalities from two incidents involving runway incursions by ground crew). There have been no serious in-flight incidents.

The 737max was introduced in 2017 and has over 1500 in service, about half of the airbus A320 neo. It has had two fatal accidents resulting in 346 fatalities and one serious in-flight incident. The record is not great.

If you want to compare all 737 models back to the first flight of the A320 I’m pretty sure it’ll be closer, but flying is much safer than it was even with the 737max.

However I do believe that Boeing will resolve the issues. It will take time to fix the culture that resulted caused them though.

> If you want to compare all 737 models back to the first flight of the A320

That's easy, the 737 has a better record. The 737NG in particular has arguably the best safety record of any airliner ever produced.

Everything is relative, and compared to other airplane models, the 737max has absurdly high rate of fatality flights. The only commercial model in history with a higher rate of fatal crashes, as far as I can tell, was the Concorde. And we don't use those anymore.

> Concorde

It seems like this would be a pretty big hint that you're looking at statistics the wrong way. When a single fatal crash can make your airliner the 'most dangerous statistically' then perhaps that's not a good airliner to use in any kind of comparison because the number of total flights is way too low to be meaningful.

No fan of Boeing. But where are you going to go to buy something else? There is only Airbus. There isn't a huge competitive market, it is just 2 companies left standing. Perhaps a good argument for fighting monopolies?

And, while Boeing looks bad, when sales go down, corporations adapt.

The pesky customers want "Safety", and we aren't selling "Safety", got to get some of this "Safety". Guess we better do some re-orgs, some focus groups, some training to make sure everyone knows "Safety" is important. Maybe throw in some training on "Quality".

They'll adapt.

>The pesky customers want "Safety", and we aren't selling "Safety", got to get some of this "Safety". Guess we better do some re-orgs, some focus groups, some training to make sure everyone knows "Safety" is important. Maybe throw in some training on "Quality".

Gah hahahaha.

Oh God, don't make me laugh that hard. One does not merely add some more Quality to a product as if it were flour in a cooking recipe. Quality is literally the single most difficult to add back to something once you've decided to skimp on it; as quality itself is a compound output. It's a result of having your entire goddamn company devoted to the goal of creating a quantitatively superior product.

Boeing decided it was time to not sweat the package, and to turn itself into a financial product. As it turns out; it's pretty much impossible to do one while not doing the other...

It all starts when psychopaths on wall st install psychopaths in the C suite.

Not even hiding it.

https://www.forbes.com/sites/jackmccullough/2019/12/09/the-p...

> It kinda bothers me that airlines are still placing orders for new Boeing airplanes.

Are they? Boeing had almost nine hundred 737 MAX orders in 2023; in 2024 they have 66:

* https://en.wikipedia.org/wiki/List_of_Boeing_737_MAX_orders_...

Three hundred 787 orders in 2023; 10 in 2024:

* https://en.wikipedia.org/wiki/List_of_Boeing_787_orders_and_...

One hundred 777s in 2023, twenty-seven in 2024:

* https://en.wikipedia.org/wiki/List_of_Boeing_777_orders_and_...

Their numbers are down bigly.

And who else are you going to order from? Airbus? How much capacity do they have to build more?

I'd note Airbus is seeing the same pattern. The 737-MAX's A320neo competitor, for example:

https://en.wikipedia.org/wiki/List_of_Airbus_A320neo_family_...

1,689 orders in 2023, 154 in 2024.

https://en.wikipedia.org/wiki/List_of_Airbus_A220_orders_and...

141 orders in 2023, negative 12 in 2024.

As a result, I presume the explanation is less "no Boeing" and more something like "orders have a reporting lag" or "2023 was a big year for airlines post-COVID".

These stats alone seem like way more interesting news, suprised I haven't seen that before. I wonder why plane orders from the two biggest manufactures have dropped so dramatically and what it means economically.

My mind immediately goes to whether airlines are quietly preparing for an expected downturn, or if airlines aren't selling to corporations because they're selling to governments and militaries. That could definitely be my own preconceptions sneaking in though.

I wouldn't read too much into it yet; 2023's numbers look like they're abberantly high, which is probably the airlines catching up after the little economic hiccup we all had in 2020.

Could be related to high interest rates and investments being more expensive. Also, I believe that Airbus has a 10 year backlog on their newest planes, so no point in shelling out a lot of cash right now for 10 years down the line.

> And who else are you going to order from? Airbus? How much capacity do they have to build more?

the obvious choice is not to order rather than ordering faulty places because your greeds tells you that risking people's safety is worth the money you will get

Do you think your car (or bike) is safer or more dangerous than the least safe Boeing plane in service?

Per mile or per hour?

Either.

These numbers are always pretty BSy because the big dangers in flight are take-off and landing. Comparing distances just completely obfuscates everything for no real gain.

You can illustrate this clearly with some sort of reductio ad absurdum. Imagine we have an interstellar ship traveling many trillions of miles, but that blows up 90% of the time. If you look at it in terms of this same deaths/mile metric it'd still be way safer than driving, but obviously it isn't - which emphasizes that the metric is misleading.

The most reasonable way would be safety per average trip, with some sort of multiplier for particularly long or particularly short trips. After all that's exactly what people think these numbers mean, even though they most certainly don't.

Okay. So let's compare accidents per flights to accidents per trip with bike/car.

IATA reports 1.19 accidents per million flights in the 2019-2023 years. [1] Your turn to find a comparable number for road vehicles.

1: https://www.iata.org/en/pressroom/2024-releases/2024-02-28-0...

You pretty quickly shifted the goal posts there didn't you? That's for all flight vessels, not just Boeing's 737 class. You're also misreading your stats (that's per sector, not per million), but in either case I'm more interested in fatality rates. Your site gives a hull loss rate of 1 per 4.94 million flights for jets. For vehicles, the average person makes two trips a day over a total of 29.2 miles. [1] So we can say the average trip is 14.6 miles. And the average fatality rate is 1.33 per 100 million miles traveled [2]. So that's 1.33 fatalities per (100 / 14.6) = 6.8 million trips, or 1 fatality per 5.17 million trips.

So on a first level analysis, vehicles are slightly safer than jets on a trip for trip basis. But the math I'm doing dramatically understates the actual difference, because the mortality numbers I'm using for vehicles are per person, not per fatal incident. In other words 1 PERSON dies per 5.17 million vehicle trips. And I'm comparing that against one entire JET 'dying' per 4.94 million flights, which is generally going to have tens to hundreds of people on it. If you do an apples to apples comparison, vehicles would be some orders of magnitude safer than jets, trip for trip.

Of course ideal would be to ignore population and just look at car incidents with at least 1 fatality vs plane incidents with at least 1 fatality, but I can't find those stats unfortunately.

[1] - https://newsroom.aaa.com/2015/04/new-study-reveals-much-moto...

[2] - https://www.iihs.org/topics/fatality-statistics/detail/state...

> You pretty quickly shifted the goal posts there didn't you?

No my dude. You said "These numbers are always pretty BSy because the big dangers in flight are take-off and landing." That is a statement about these kind of statistics in general, not in specific to Boeings. But sure. How many accidents had Boeing during take-off or landing?

> You're also misreading your stats (that's per sector, not per million)

Negative. To quote "All accident rate (accidents per one million flights)" What is even per sector?

> Your site

Not my site? Are you arguing from good faith here?

> If you do an apples to apples comparison, vehicles would be some orders of magnitude safer than jets, trip for trip.

I will let everyone know who is considering commuting by jet twice a day for 14.6 miles.

The post I was responding to was speaking precisely about "the least safe Boeing plane in service." The fact that airplanes are, trip for trip, much safer than planes is something I did not appreciate at the time when making my post. I expected most planes to be safer than vehicles, but Boeings falling well below average. The fact that vehicles are substantially safer than any plane, trip for trip, is far beyond what I expected. And it's going to make Boeing's look like death traps by comparison.

Imagine we're comparing a long flight thousands of miles around the world to a short little domestic flight between a couple of relatively nearby cities. Which flight do you expect to get the latest, most stable, and secure plane? Most of everybody would expect it to be the international flight. In reality, it's the short little domestic flight. The reason is that with planes a trip that goes on for thousands of miles is, in general, almost exactly as dangerous as the briefest of trips imaginable. Take off, landing, and pressurization are where basically all the risk (and wear and tear) come into play. So your little plane going on a hundred mile voyage is going to be exposed to far greater risks than the one traveling thousands of miles. It's why the stats, as typically offered - deaths/mile, are extremely disingenuous. And they serve very little purpose other than to mislead those who don't know better, which is the overwhelming majority of people.

As for the exact numbers, reread your own source, and include the entire quote. You're having a failure where you keep misreading something the same way. Keep in mind you quoted 1.19 for 2019-2023.

> You're having a failure where you keep misreading something the same way.

And you are having a failure where you are not quoting what you think I'm misreading. Look at the table. First column labeled "ACCIDENT TYPE". Second row first column "All accident rate (accidents per one million flights)". Last column labeled "5-YEAR AVERAGE (2019-2023)" The data in second row last column "1.19 (1 accident every 0.88 million flights)".

Go ahead and tell me where do you see "per sector" in any of that?

Ahhh!!! I see how you're reading it, and yeah - I think the mistake is much more on them than you. The parenthesis are not describing the value before them, but indicating what is IN the parenthesis! I'm the sort that ignores infographics and goes straight to the text, and it's much clearer there. Check out the report highlights:

---

The all accident rate was 0.80 per million sectors in 2023 (one accident for every 1.26 million flights), an improvement from 1.30 in 2022 and the lowest rate in over a decade. This rate outperformed the five-year (2019-2023) rolling average of 1.19 (an average one accident for every 880,293 flights).

---

So in the table their "All accident rate (accidents per one million flights)" row is giving you two different pieces of the data. The first is accident rate (which is per million sectors) and then in the parenthesis is the accidents per million flights. And actually even that's wrong, since they actually give you flights per accident -- whoever put that table together clearly was not big on the whole accuracy thing.

So when you see "1.19 (1 accident every 0.88 million flights)" that means the rate per million sectors is 1.19, and the rate per million flights is 1/0.88 = 1.14.

That explains it! Dang! Thank you for the help figuring that out.

> That's for all flight vessels, not just Boeing's 737 class

That works in your favor, though. The 737NG has a terrific record, probably the best in the industry. Including all airplanes drives the crash rate higher, which will help your comparison with other modes of travel.

>The most reasonable way would be safety per average trip

Let's reductio ad absurdum this. I take my 1000CC motorcycle around the block every 10 minutes for a year. Safest transportation mode on the planet! Way safer than walking (which I did once down the middle of a highway, blindfolded)

It's interesting to see the dynamic of the downward spiral (no pun intended) "pile on" media cascade mechanic in action, through your comment.

What I mean is, you are assuming now (you probably wouldn't have a year ago) that this Dutch Roll incident was caused by the plane being Boeing. So, why would anyone even buy a Boeing at this point?

But what indication is there of that this Dutch Roll incident was caused by the plane being Boeing? As far as I can tell, Dutch Roll is a phenomena whose causes aren't fully understood, that rarely, but occasionally, happens to planes, not just to Boeing planes.

But now that we are collectively bought-in on a Boeing hate-fest (and they certainly seem to have brought it on themselves, no doubt!), we just knee-jerk apply the cause to any issue seen as being explained by the aircraft being Boeing.

Such is the state of the airline business/capitalism. There are basically only two manufacturers: Boeing and Airbus. The airlines place orders years in advance and if you cancel a Boeing order now you’re a) out of a lot of money and b) joining the back of the line with Airbus.

The simple reality is that the airlines (and their shareholders) consider a) and b) to be greater risks than the safety record of Boeing planes. Given the parameters they have they’re not wrong. Air travel is still incredibly safe, even if these Boeing incidents make it ever so slightly less safe. To be blunt, the market can absorb it and individual airlines have little motivation to change. If there is going to be a big change it’ll need to come from the FAA.

Yeah, Airbus is extremely backed up with orders

So is Boeing.

The Max 8 are great planes. I rode one recently, in fact it was with Southwest, it was probably the best plane I flown in. Lots of space, comfortable and quiet.

They just need to fix the bad part issues they seem to have.

Personally, I would fly on a Max 8 again.

Looking at Southwest is probably a good way to put the Boeing issues in perspective. It's easy to want to freak out over the negative headlines over the past few years, and the MCAS fiasco is a well-deserved black mark on Boeing's reputation.

But again, looking at bigger perspective, Southwest Airlines operates an all-Boeing fleet of 800+ airlines doing 4000+ flights per day. About a million and a half flights per year. And in their 50+ years of operation, there have been 4 deaths associated with Southwest flights, 1 of which was a suicide where a guy walked onto a runway and another was an unruly passenger who tried to storm a cockpit.

I don't want to let Boeing off the hook - their recent cock-ups absolutely seem to indicate cultural problems that need to be fixed so that quality can improve rather than continue to slide. But honestly, getting on a 737 Max to go on a trip still seems way safer than hopping on the highway.

> Dutch Roll is a coupled out of phase movement [...], in which the aircraft oscillates around its vertical as well as longitudinal axis (coupled yaw and roll).

That sounds ... adventurous. What might cause something like that?

For anyone else (besides me, being Dutch) wondering about the name: https://en.wikipedia.org/wiki/Dutch_roll?useskin=vector#Name

TL;DR: it's from ice skating where there's a similar move. But no idea why it's called that in ice skating.

Ice skating... Interesting... My guess was it having to do with ships rocking side to side.

Maar waarom een schaatsterm?

Maybe the ice skaters took it from ships, also an old Dutch tradition I suppose.

I'm not sure if this is relevant, but: I have read that back in the day (1500s-1700s), the Dutch built flat-bottomed boats that could carry a lot, and the English built round-bottomed boats that handled better in the ocean. Maybe the way a Dutch boat handled in heavier seas is the origin of the term?

Sounds plausible, your guess is as good as mine

That was indeed my guess.

At first I read that as `schatsterm` and thought it was cute, but then I looked it up to double check and it's still cool!

But now I know that the word I thought it was is actually a `bijnaam`. Your language is fun and I have been enjoying learning it over the last couple of years. :)

Doesn't it have yaw dampeners to prevent this? Ps didn't have time to read the article yet, sorry. But I thought this issue was mitigated on the 737

So what you are saying is that we shouldn't fly the 737-8 MAX?

I will walk, thanks!

I wonder what the odds of injury or death are if you walk those 768 miles vs fly. I think I'd fly.

This is an aircraft worth taking great lengths to avoid. Cancelling important trips, meetings, etc.

Statistically it is not. Most of the time travellers don't know if they're on an Airbus or Boeing aircraft and hardly ever figure out that it's a MAX.

Infinite loss multiplied by small chance is still infinite risk

What daily activities don't have a small chance of the same risk?

Eating has a small chance of choking to death. Walking has a small chance of getting hit by a falling tree branch. Pooping has a small chance of causing a brain aneurysm to pop.

Yes, we will all die. Every last one of us.

How does this impact the share holders?

#NothingWorks

Horses are very stable

I hever never once heard of a dutch roll occuring to a train.

Am on a Dutch train right now and there is a distinct lack of roll, Dutch or otherwise.

That said, I wonder where the term Dutch Roll comes from. Perhaps the oscillation is reminiscent of the motion of a ship rocking side to side. The Dutch were a major sea power.

When it happens, it's catastrophic, and then it is not called a dutch roll.

The rail equivalent is called hunting oscillation. https://en.m.wikipedia.org/wiki/Hunting_oscillation

One of my first horse rides as a child made me pretty much swear off ever getting on one again. Stable it was not...

You can buy dutch rolls in a lot of spaces in the Netherlands called coffeshop.

Reply on news.ycombinator.com