Boeing's 737 -- Prone To Problems?
March 21, 2012 3:32 PM   Subscribe

 
Eponysterical.
posted by acb at 3:43 PM on March 21, 2012 [12 favorites]


This is really not surprising for anyone that has seen the 737-200s flown by third world airlines. Maintenance is lax, at best. There's a huge used market for 25 year old 737s to fly cheap flights in developing nations, and pricing pressures are very high. Go visit the cross-runway "Terminal 2" of Dubai airport for an example.

This is essentially the aviation version of buses plunging off cliffs in Peru.
posted by thewalrus at 3:45 PM on March 21, 2012


Jesus Christ, I am flying to China on Saturday! Why the hell did I read this?
posted by Wolof at 3:50 PM on March 21, 2012


Jesus Christ, I am flying to China on Saturday! Why the hell did I read this?

Well, if you're flying to China from where your profile says you live I'd say it's a safe bet ya ain't flying in a 737.
posted by matty at 3:59 PM on March 21, 2012 [5 favorites]


A few weeks ago my mechanic uncle, who works for a US legacy carrier, said without irony "Scarebus".
posted by wcfields at 4:04 PM on March 21, 2012 [2 favorites]


Sure, but I'm flying in China when I get there. Have you seen how old some of those aircraft are? Reading phrases like "explosive decompression" is probably not what I need to be doing.
posted by Wolof at 4:04 PM on March 21, 2012


Holy crap, that picture on the third page with the cabin peeled back like a sardine tin made my toes curl! Imagine riding that thing down to safety!
posted by Ron Thanagar at 4:09 PM on March 21, 2012


Some perspective from this article.


Statistically, the 737 has had a better than average safety record over its nearly three decades of service — 1.21 crashes per million flights for older models and 0.51 crashes per million flights for newer models. The figure for passenger jets of all types is 1.83 crashes per million flights.

posted by sien at 4:09 PM on March 21, 2012 [5 favorites]


Hidden at the end of the article: "Paul Hayes, who oversees one of the world’s most extensive and respected data banks on air safety at the British company Ascend" ... "said that there was no significant difference in the safety record of the 737 and the A320."
posted by markr at 4:14 PM on March 21, 2012 [4 favorites]


Sure, but I'm flying in China when I get there. Have you seen how old some of those aircraft are? Reading phrases like "explosive decompression" is probably not what I need to be doing.

I would be more scared of China's roads than China's airlines.
posted by Talez at 4:19 PM on March 21, 2012 [1 favorite]


Hey mr_crash_davis I got out of one these this morning and tomorrow morning I get into a 737-800. Thanks for thought. (and yes it was AA and yes it sucked; the most dangerous thing about it though to my unprofessional eye was the "food". What was that shit they wanted me to eat?
posted by adamvasco at 4:23 PM on March 21, 2012


Have a nice day :) [SLYT]
posted by steamynachos at 4:33 PM on March 21, 2012 [1 favorite]


I cannot overemphasize how complicated metal fatigue is to analyze in all but the most trivial of cases. And keep in mind that fatigue failures are statistically distributed, so you can't easily say "this part will failure after N cycles." So yeah, airlines (or their outsourced maintainers) really need to be on top of inspections.

I'll also point out that if you see a lot of 737 issues, part of the reason is that they've built an awful lot of them -- about one a day for decade after decade.
posted by LastOfHisKind at 4:40 PM on March 21, 2012 [3 favorites]


This is really not surprising for anyone that has seen the 737-200s flown by third world airlines.

I fondly recall sitting in Cathay Pacific's lounge at HKG and watching an ancient 737 owned by some Chinese domestic airline taxi by. It looked as though someone had taken after it with a ball-peen hammer and then decided to rub black grease on the leading edge of the wings for good measure.
posted by nathan_teske at 4:45 PM on March 21, 2012


Oaky. There have been over 7000 737s built by Boeing, with another 2100 on order. The first one was built in 1967. There is something on the order of 100 737s in the air around the planet *at all times.*

So, yeah, I expect a few of them to crash. However, you'd be surprised at how little they do. The new series, the -600 through -900, has had only 8 hull loss accidents with almost 4000 in the air.

You couldn't pay me to get on a 737-200 or a -300 on a third world airline. That's an airframe that was already over 30K hours when it got to the third world, and has probably spent a lot of time in the air. Worse, the chances of any real maintenance inspections are low to none. Then again, there are legions of airframes of similar vintage that I also wouldn't get onto -- and I wouldn't be comfortable getting on them on a first world flag carrier unless I know that a D check was in the recent past, and was performed by an airline without a history of lying about inspections.

And, indeed, you can't stop the 737. The new generation, the 737-MAX, using the new CFM LEAP1-B engines, is now on offer -- and has almost 1000 orders.

I'm perfectly happy flying on one on anything less than 5 hours. Longer, and I want more room to walk around to stretch my legs. They're good airplanes.
posted by eriko at 4:48 PM on March 21, 2012 [9 favorites]


I wonder if Southwest's unusually short cycle time has anything to do with it. Not only do they tend to fly shorter hops (with 737's anyway) than other US airlines but they also tend to turn the planes around faster.
posted by wierdo at 5:02 PM on March 21, 2012


Official Boeing 737 MAX Video
posted by gen at 5:02 PM on March 21, 2012


Holy crap, that picture on the third page with the cabin peeled back like a sardine tin made my toes curl! Imagine riding that thing down to safety!

One person didn't.
posted by narcoleptic at 5:04 PM on March 21, 2012 [1 favorite]


I was a bit concerned that of the three accidents involving NG 737's listed at the end of the article, two of them were "running off the runway" incidents, where pressurization is not an issue, and the third one sounds like a hard landing, also not a pressurization issue. Why include this? This has nothing to do with any design flaws that may exist.
posted by Mcable at 5:06 PM on March 21, 2012 [1 favorite]


Part of the article touches on how the aft bulkhead weakness could/can result in the air frame breaking apart during a hard landing. Especially in the latest variant which hangs more mass off the bulkhead and farther away to boot. And if the airframe breaks up it'll be hard to get people off the plane.
posted by Mitheral at 5:12 PM on March 21, 2012


The List of airlines banned from entering EU airspace is a good reference for which third-world airlines are reliable, and which are not. Certain PIA (Pakistani) airplanes are maintained in the UK or in Germany by Lufthansta Teknik for B, C and D checks just so that they can fly into the EU.
posted by thewalrus at 5:12 PM on March 21, 2012 [8 favorites]



These types of articles amaze me.

Because really, aviation is one of the best-regulated industries in the developed world. It has everything in its favor: failure is disastrous and extremely visible to the general public, regulators have tremendous popular and political backing, the number of regulated parties is relatively small, the regulated parties also have a vested and serious interest in safety, the problems are usually technical and not social, and the technical problems have a long history of study. The FAA and NTSB can ground an entire aircraft type or an entire airline if it thinks it's necessary. There's really no other industry where the regulatory body has so many factors in its favor.

But with some frequency, journalists seem to get the idea that their help is needed, that there are secrets and misinformation and hidden dangers lurking that the regulators are failing to catch or are for some reason hiding from the public. I mean, this is not some back-country mining operation that the public doesn't know about and where the regulator and industry collude to maximize production. Aviation is wildly public industry, and the technical challenges faced are well known and well studied. The types of issues the article describes are the bread and butter of the field of fracture mechanics. Clearly those failures are a problem, but it's not like the industry is in a state of ignorance over this. Every single design choice on an aircraft, and particularly something as significant as the fuselage structure, is rigorously scrutinized over and over again, and backed up with actual calculations of the crack growth rate, the detectability of cracks, and how the structure will fare if a part does fail and loads have to be transmitted along alternate paths.

I mean, I think it's fascinating stuff, and I can appreciate that the article conveys some of the details of aircraft structural mechanics, but the whole "prone to problems" angle is just weird to me. I guess I would have titled it something like "737: Undergoing Continuous Refinement, Just Like Every Other Airliner".
posted by kiltedtaco at 5:27 PM on March 21, 2012 [23 favorites]


This is essentially the aviation version of buses plunging off cliffs in Peru.

Well, that's aviation too. Briefly.
posted by yoink at 6:06 PM on March 21, 2012 [16 favorites]


I wonder if Southwest's unusually short cycle time has anything to do with it.

I think it's unlikely - two bigger causes would be the aforementioned third-world airlines and corrosion.

For the uninitiated, let's discuss fatigue and how it applies to airplanes. The article did a pretty good job, but some of it reeked of fear mongering to me.

Metal fatigue is something that occurs when metal structures are subjected to repeated, cyclical loads. Take a paperclip, unbend it, and then rebend it. Keep doing this until the end of it snaps off. The number of times you had to bend it is the number of cycles to failure. What's happening at a micro level is that small cracks form, even if your loads are under the yield strength of the material, and these cracks provide very sharp points where stresses are concentrated. As you apply loads again and again, the cracks grow at these points and eventually the structure fails. The cyclical loading can be high or low frequency; high frequency would be something like a vibration profile (the engines vibrate as they operate, for example, putting loads into the airframe) and low frequency is something more cyclical like landing gear extensions and pressurization.

Engineers talk about "cycles to failure" because they're independent of time. Take your paperclip and do one bending cycle. Now put it in your desk drawer and do it again tomorrow. Once per day, bend and unbend the paper clip. It will fail in the same number of cycles as the one you do every thirty seconds, but doing it once a day means that the paperclip lasts many times longer. Same thing with an airplane - fly it once a day and it may last thirty years, but fly it five times a day and you're looking at six years. Number of flights is the same in both cases.

Determining fatigue, as alluded to above, is a bit of a dark art. The analyses available already assume that a crack has begun and simply try to predict how it grows and when failure occurs - there's no way to determine where cracks start or when. In fact, complex structures are already riddled with microscopic cracks that form when the metal is forged; it's just a matter of hitting that sweet spot of loads that causes that crack to develop into something that will eventually cause failure.

So, why do I say third-world airlines and corrosion are the culprits for crashes? The article alludes to both, but let me expound a bit. Corrosion is another structures topic that is again very, very difficult to predict. Aircraft fly all over the world, over the oceans, and are subject to a wide variety of environments. How corrosion forms and its effect on the strength of the airframe are not even statistically quantifiable. There are tests you can perform - salt and humidity tests where the airplane is put in a large chamber and broiled for days and days - but these tests cannot be compared to what the airplane will see in its lifetime. Too often I see test reports from contractors claiming that their equipment has gone through a "lifetime" humidity or salt fog test, and it simply isn't true - the tests were never designed to be indicative of lifetime exposures. What they DO tell you is where you might have missed some kind of galvanic interaction - steel mating with aluminum is easy to overlook in the design but will absolutely start corroding the second it sees salt air.

Poor maintenance practices are also a cause of airplane crashes. One thing that Boeing does well, in my opinion, is collecting huge reams of data from its customers when they do periodic inspections. Airlines will go in every now and then and, directed by Boeing, will look for cracks in the airframe. They send this data back to Boeing, and Boeing uses it to predict crack growth and other areas for inspection. The airlines get Service Bulletins in return from Boeing warning them of potential problems. Airlines that don't prioritize maintenance don't do these inspections and will not find potential problems until it's too late.

The last point I'd like to make is that there are very few tests that consider an airplane as a whole system, and that's something that's always bugged me a little bit. Consider what the article mentioned - pressure tests without the wing box or landing gear cycles. This is common practice, but what you're missing with this test is the condition of crack growth (the pressure cycle) followed by a large shock. That extra shock may be enough to do substantial damage, but that's not how it's tested. On the other hand, testing the airplane as a system would basically mean flying a prototype for its entire expected lifecycle, which would be impractical.

I wouldn't be concerned by the article. I would be concerned if I were flying an airline in a country that would allow them to patch it together with duct tape. If you're in the USA or Europe, flying has never been safer than it is today. I don't deal with EASA much, but I think that the FAA is one of the best-run government agencies we have despite the continual funding crises and they are very conservative about safety. As it should be.
posted by backseatpilot at 6:07 PM on March 21, 2012 [27 favorites]


Hey, it's the 21st century. Where's my plasteel? Or better yet a way of moving large masses through the air that doesn't involve stresses and strains.

> The types of issues the article describes are the bread and butter of the field of fracture mechanics

It would be nice if the field of fracture mechanics didn't have or need a bread and butter. OK, they can analyze ice cubes. I have one they can borrow, just don't test it to destruction.
posted by jfuller at 6:10 PM on March 21, 2012


After reading that I am struggling to understand what the author was trying to achieve with this. Scary sounding opinions are trotted out with no context (safety versus equivalent aircraft and the industry as a whole) and little analysis on why the design decisions being criticized (sticking with the basic 737 design) made sense at the time.

The issue with the aft pressure bulkhead seemed typical of the article. It is a problem and needed correction. It also was found during airline safety checks in only 4 instances after planes had been landed hard enough to damage the landing gear (or similar). That seems a pretty low rate with two 737s landing every 5 seconds, many rough landings must have occurred with no identifiable damage.

I would also say the system worked - the required inspection found the damage and a solution was put in place. The inspection is not trivial (I'm no maintenance engineer but in a random Aircraft Maintenance Manual for a 737-300 I looked up the Hard Landing inspection is a comprehensive 16-page checklist) and if it is not followed then the airline will lose that plane's airworthiness certificate. As they learn more about the plane they make it safer - go figure.
posted by N-stoff at 6:30 PM on March 21, 2012 [2 favorites]


Holy Crap that video steamynachos posted is crazy scary. Has anyone who has issues with the article watched it?
posted by thekorruptor at 6:53 PM on March 21, 2012


I would also say the system worked - the required inspection found the damage

This is a good point that's worth reiterating. Boeing fully expects small cracks to develop in the 737, growing from the microscopic cracks present in the metal originally as described above. To make this safe, they calculate the time or number of cycles it will take for a microscopic crack to grow enough to cause a failure (typically by joining up with a crack from a neighboring rivet hole), then mandate that the airframe be inspected frequently enough that those cracks will be detected before the failure occurs. Then, even if a crack does grow across multiple rivet holes and cause a failure of that piece of skin, there are periodic reinforcements along the fuselage that prevent a crack from growing more than a few feet. One of those did successfully stop the crack on the southwest flight, though it looks like one or two of them failed, but I can't quite tell from the pictures I can find.
posted by kiltedtaco at 7:12 PM on March 21, 2012


Holy Crap that video steamynachos posted is crazy scary. Has anyone who has issues with the article watched it?

Unquestionable, there are gaps and shortcuts in aircraft maintenance. However, the proof is in the numbers. There are just an unfathomable number of 737's flying. It's the best selling airliner in history.

According to wikipedia (tm), there have been a total of 159 hull-lose accidents. While 159 out of 7000 might seem like a scary number, consider that almost all of these planes have been continuously flying. Despite it being the beat up city-bus of airliners, it's incident statistics are lower than average.

This includes the vast majority of those ill maintained airframes being flown by third world carriers. It's mostly unchanged design is a testament not to obsolescence, but to an awesomely well engineered and crafted object, tweaked and improved since 1967.

Continual vigilance of our safety agencies is always called for. However, this article is pure 100% bunk.
posted by PissOnYourParade at 7:26 PM on March 21, 2012 [1 favorite]


A few weeks ago my mechanic uncle, who works for a US legacy carrier, said without irony "Scarebus".

Related: "If it ain't Boeing, I ain't going." Call me a control freak, but I am a little uncomfortable in any plane unless I am in the left front seat. The one with a full set of flight controls.
posted by exogenous at 7:44 PM on March 21, 2012 [2 favorites]




"I'ma take the rocket-bus to Galveston!"
posted by bardic at 8:04 PM on March 21, 2012 [1 favorite]


Scariest airplane I've personally been on: a 30-year old ex Air India Airbus A300 flown by Ariana Afghan Airlines. Now that was bad.
posted by thewalrus at 8:32 PM on March 21, 2012


I would also say the system worked
In an odd way, I think slightly silly, hysterical articles like the one linked to might serve as part of, or be important to, that system. Part of the reason that air travel is so safe is precisely because the industry has to overcome strong public responses to aircraft incidents and accidents, as well as many people's fear of flying—which leads to strong public support for centralized regulation.

That said, lines like "Certainly, Boeing will not be abandoning its cash cow any time soon" are pretty ridiculous. When you find that a plane has a problem, the best course of action probably isn't to throw one's hands up in the air, ground the entire fleet, and halt current production and future development. Especially in a widely popular line with a better-than-average safety record. Boeing has successfully identified and dealt with other 737 safety issues over the years (rudder hardover, for instance). It's fine to suggest that Boeing should better their fuselage design for future models, or create a retrofitting regime that addresses the problem. It's stupid to suggest that they simply need to "abandon their cash cow."

Where's my plasteel?
Well, there's composite, which is coming into much greater use; I believe the 787 Dreamliner's fusilage is almost entirely composite. But composite comes with its own set of issues, including vulnerability to lightning strikes.
posted by evidenceofabsence at 9:59 PM on March 21, 2012


Sure, but I'm flying in China when I get there. Have you seen how old some of those aircraft are? Reading phrases like "explosive decompression" is probably not what I need to be doing.

On one interior flight in China I got an ancient Tupolev on the outbound leg; the thing looked and felt like it had been milled from a single giant immense hunk of pig iron. Even the tray table was made of steel; I see why they make you put them up for takeoff and landing; in an emergency landing a.k.a. crash the damn thing would cut you in half.

Going back the other way I got the other extreme, an Airbus so new I think I was the one to remove the plastic wrap from my seat. The thing gave every indication of having no metal in its structure whatsoever: looking down (or rather up) the cabin during takeoff you could see the fuselage visibly distort. But the whole plane weighed about what da Bic lighter does and probably used less fuel. All of course in my completely scientific, accurate to 8 decimal places opinion.

Note: despite the astonishingly persuasive impression I am not an aeronautical engineer and I am certainly not *your* aeronautical engineer.
posted by George_Spiggott at 10:47 PM on March 21, 2012 [6 favorites]


But composite comes with its own set of issues, including vulnerability to lightning strikes.

From what I understand the remainder of the flight after the first in-service lightning strike on one of the ANA 787s was an anxious time (though the shielding worked fine). But between this, the sensitivity of the composites to UV and the extra hassles of repairs if you accidentally dent a 787 body compared to an old-school airframe there is something to be said for familiarity.

I'm pretty confident that the 787 (and A380 and Embraer E-198, any new model really) will have a great safety record but the 40+ years of knowledge on how the 737 ages is comforting in many ways. There is little uncertainty on the hours left in a 737 at various stages in its life - there is so much historical data available. It is a well-known problem.
posted by N-stoff at 11:07 PM on March 21, 2012


There are thousands in the air right now. This is a silly article.
posted by mattoxic at 1:09 AM on March 22, 2012


But between this, the sensitivity of the composites to UV and the extra hassles of repairs if you accidentally dent a 787 body compared to an old-school airframe there is something to be said for familiarity.

Not too mention the difficulty of inspecting a material that may be damaged in a less-than-obvious manner (delamination).

BTW, congratulations on your magnificently obscure aerospace geek handle.
posted by Skeptic at 2:16 AM on March 22, 2012


Not too mention the difficulty of inspecting a material that may be damaged in a less-than-obvious manner (delamination).

They may have that licked. Turns out you can sense it by sound - delaminated composites sound different. Tools are being developed that may allow the inspection of the entire skin of a composite aircraft without needed to strip the paint or take weeks to do so.

Aside: the funny thing about AA buying the 787 is that they're going to have to paint them. Unlike many airlines, AA's livery has been remarkably unchanged, and involves not painting the aircraft - its a big silver bird with a red/white/blue stripe.

But the 787 has to be painted. I kind of hope the can use a metallic base paint and keep the gleam, but I suspect it'll be gray.

Contrast this with UA, where changing the livery is what they do best.
posted by eriko at 3:27 AM on March 22, 2012 [1 favorite]


Because really, aviation is one of the best-regulated industries in the developed world.
Obviously, then, the answer is to immediately de-regulate the industry and release the invisible hand of proper maintenance.
posted by Thorzdad at 4:44 AM on March 22, 2012 [2 favorites]


Apparently, if we're really serious about safety, we need to allow smoking on planes again.
posted by Benny Andajetz at 5:39 AM on March 22, 2012 [2 favorites]


Yeah, this article sounds like someone trying to legitimize their anxiety. I flew on a few MD-80s a while back, and was similarly terrified. These were some OLD aircraft- the plastic walls had clearly been repainted with a roller and what appeared to be just latex semi-gloss. The cockpit was that gross military mint green, with layers and layers of it rubbed off and repainted. When it took off, it creaked and rattled and buffeted like crazy. So, I went home and looked at the stats and found that while the aircraft were old, they were at the time the absolute safest plane in the sky, in an age+passenger miles/crashes sort of sense. In other words: my common sense ("I wouldn't ride in a BUS that acted like this plane!") was wrong.

(I will, however, complain loudly that sitting between the engines in the rear of the plane scrambled my brains. The sound of the engines being not-quite-in-tune with each other and creating a deep sonic interference pattern was unpleasant.)
posted by gjc at 7:01 AM on March 22, 2012 [1 favorite]


They may have that licked. Turns out you can sense it by sound - delaminated composites sound different. Tools are being developed that may allow the inspection of the entire skin of a composite aircraft without needed to strip the paint or take weeks to do so.

Uhmm. It's admittedly a long time ago since I actually worked with composites (15 years), but it has never been as simple as just tapping on them. Some forms of delamination are more insidious than others, and can be quite difficult to detect, even using ultrasound systems.

Also, stripping the paint has never been the issue: delamination can be quite invisible from the outside, even when there isn't any coating.

Apart from the familiarity issue, composites are simply trickier than metals, because, being more complex, they just can fail in many more, and more complicated, manners. With metals you have metal fatigue and its interaction with corrosion. In composites, you have to consider the potential degradation of both the fibres and the matrix, as well as that of the fibre-matrix interface, and the interfaces between adjacent layers, not to mention the anisotropic behaviour of the material, the effect of some environmental factors such as UV radiation, etc.
posted by Skeptic at 7:06 AM on March 22, 2012


> Turns out you can sense it by sound - delaminated composites sound different. Tools are being developed...

> It's admittedly a long time ago since I actually worked with composites (15 years), but it has never been as simple as just tapping on them.

By "sound" and "tools", I think we're talking less a ball peen hammer and a good ear and more something akin to a sonogram aren't we? The composite equivalent of using x-rays to exam welds and such. Not that'd I'd know anything more about it than you'd read in Aviation Week.
posted by adamt at 8:06 AM on March 22, 2012


I believe there are a couple of other possible issues with the composites:
Some of the assemblies are made of materials that are export restricted, all fixes will have to be done by Boeing MRO's and/or under the goldcare program. This could be a problem in remote locations.
Secondly there may be a serious problem with long term maintenance costs. We don't really know how composites behave over time - so even if repair costs are just slightly higher than aluminium it might turn these things into flying white elephants.
That said the 787 is a thing of beauty.
posted by fingerbang at 9:48 AM on March 22, 2012


Holy crap, that picture on the third page with the cabin peeled back like a sardine tin made my toes curl!

See! I told you to turn off your Kindle!
posted by banshee at 10:32 AM on March 22, 2012


One of my fellow students at a composite aircraft building class I was in last month worked at some multinational whose name I forget (GE I think), on devices to analyze structures of composite aircraft. He brought along a fancy ultrasound analyzer along with a square of carbon fiber laminate prepared to have a number of hidden defects.

It wasn't possible for anyone in the class to find the any of defects by tapping with the edge of a coin (a typical technique with amateur-built aircraft). But by rolling the sensor of this device along the laminate, the defects would appear on the display at various depths.
posted by exogenous at 2:38 PM on March 22, 2012


We were fools to think we could improve upon the safety record of nitrocellulose-doped fabric. Fools!
posted by George_Spiggott at 2:57 PM on March 22, 2012 [1 favorite]


gjc: "I flew on a few MD-80s a while back"

Ah, the good old MD-80s in their latter years. I gotta say that, as scary as they were, I always enjoyed the suspense of the flight more than the fear of catastrophe. Same with the 727s, I guess. Both of those old types were some serious workhorses, and I admired both of those two whenever I was on one.
posted by InsertNiftyNameHere at 12:00 AM on March 26, 2012


« Older Goal, Squid Attack for USA.   |   No, not the warehouse in Seattle, but the real one... Newer »


This thread has been archived and is closed to new comments