I am not an expert, however. Can metal fatigue be detected with such infrequent inspection?
[1]: https://en.wikipedia.org/wiki/Aircraft_maintenance_checks#AB...
On things like D check, the aircraft is essentially completely taken apart and inspected at that level typically taking 50,000 man hours and 6 month-1 year of time.
The article mentions the cost and that Boeing underestimates it. When you divide the cost by the number of hours, it seems very reasonable. Parts and materials being included. I’m surprised any job that extensive isn’t even more expensive.
I believe that it's the opinion of experts that Boeing either misjudged the safety risk of the bearing assembly when they should not have or that they incorrectly downplayed that risk when disclosing the flaw to their customers and I'm inclined to believe those experts because Boeing has already demonstrated themselves to be outright dishonest and negligent when it comes to the safety of their products.
That said, while I would not be surprised if things truly are as they currently appear to be, my assessment is always subject to change if additional information comes to light which makes that less likely to be the case.
Even now there is a lot of uncertainty around this crash, maintenance - or lack thereof - or even wrong maintenance could still be a factor. But given the location of the part asking for a 'visual inspection' is a pretty strange move, the part is all but inaccessible when it is in its normal position and even with an endoscope it would be pretty hard to determine whether or not the part had weakened. That's just not going to show up visually until it is way too late unless the part has been especially prepared to announce the presence of hairline cracks.
You'd have to disassemble a good chunk of the wing to gain access to the part based on the pictures I've seen of how it all holds together when assembled.
The same happened with MCAS, the pro-Boeing argument was that if those were American pilots it would have been fine.
There is no way pilots form all over the world could "regress to the mean". They could not have been all, or most, "above the mean". The mean would be higher then.
I mean, those are pretty standard maneuvers, up to 4gs or so, in small aircraft, and I used to fly aerobatic frequently... but it just hits different somehow on an aircraft that weighs 70 tons and flexes visibly.
Between US pilots having to lie about their mental health and then having average pay far lower than people who use Claude code daily, they really aren’t sending their best.
Unironically whatever you think pilot training involves, we should probably double it. This would be extra good as raising the prices of flights will make fewer people fly. Far too many people right at this moment who shouldn’t be flying are flying.
If we could finish by forcing all airline seats to be at business class quality (thus average flight ticket costs are about 2X now as economics of scale kick in), than flying would become a humane practice rather than war crimes in the sky.
I’m pretty sure no American airline had the same situation that the airlines with the crash had because they paid extra for the redundant AOA sensor.
The MCAS issue was a major issue, but the ultimate fundamental flaw was Boeing not including a redundant sensor (which is the one that was malfunctioning in the crashes) in the base package as they should have.
The inexplicably considered redundancy in this part an optional extra, and as far as I’m aware there were no US airlines that hadn’t taken the optional extra package.
There was no redundancy AOA sensor option for MCAS.
All the planes were built with two AOA sensors, with the original MCAS implementation only using data from 1 sensor.
Edit: I was misremembering. Both sensors were enabled on all planes and MCAS only used one at a time on all planes.
What was disabled, unless paid for, was software which displayed to the pilots that the 2 sensors were disagreeing, which would immediately have alerted them to what may have been wrong.
> According to Bjorn Fehrm, Aeronautical and Economic Analyst at Leeham News and Analysis, "A major contributor to the ultimate loss of JT610 is the missing AoA DISAGREE display on the pilots' displays."[109] > The software depended on the presence of the visual indicator software, a paid option that was not selected by most airlines.[110] For example, Air Canada, American Airlines and Westjet had purchased the disagree alert, while Air Canada and American Airlines also purchased, in addition, the AoA value indicator, and Lion Air had neither.[111][112] Boeing had determined that the defect was not critical to aircraft safety or operation, and an internal safety review board (SRB) corroborated Boeing's prior assessment and its initial plan to update the aircraft in 2020. Boeing did not disclose the defect to the FAA until November 2018, in the wake of the Lion Air crash.[113][114][115][116] Consequently, Southwest had informed pilots that its entire fleet of MAX 8 aircraft will receive the optional upgrades.[117][118] In March 2019, after the second accident of Ethiopian Airlines Flight 302, a Boeing representative told Inc. magazine, "Customers have been informed that AoA Disagree alert will become a standard feature on the 737 MAX. It can be retrofitted on previously delivered airplanes."[119]
https://en.wikipedia.org/wiki/Maneuvering_Characteristics_Au...
Boeing: Do you want a two line code which triggers a potentially life-saving warning when your flying sausage with wings has an important sensor malfunction?
Customer: Of course!
Boeing: That'll be $25K, thanks.
Is that what I'm reading?
If you want to see the way this looks on the flight displays that a pilot sees, this video shows some examples (generated from a flight simulator): https://youtu.be/L5KQ0g_-qJs?si=AtYkellEROnHZ89e&t=349
Unknown to the customer was that all machines were identical. The technician's "installation procedure" was to enter the Service Mode password, select the feature enable option, and exit Service Mode then run a test to make sure it worked.
This is pretty common in commercial/industrial manufacturing. The exception cost to omit certain hardware subsystems when building a product is often higher than the cost of the hardware itself, so it makes more sense to build everything identically and enable/disable features in software.
What a load of bullcrap. Full stop.
The crews of the two crashed 737Max were also well trained, skilled professionals.
That the US-based crews decided to re-engage the auto-pilot, and with that action, by sheer luck, managed to bypass the fatal MCAS issues, shows you exactly what it was: sheer luck.
These pilots reacted to a system malfunction of a system they hardly knew existed (thanks to Boeing's lies), that changed the aircraft subsystems behaviour in fundamental, undocumented ways compared to the previous generation of 737s, and that they were therefore not trained to handle. So skill differences did not enter the equation, luck did.
The choice was between doing the manual procedures they were trained to do to try to regain control, and the hail mary approach of re-engaging the autopilot wtith the hope the problem went away. With no time to do both. The crashed crews chose option 1, the US crews option 2.
This is like Tesla claiming that all crashes due to autopilot failures are driver faults because they are not properly trained... it is supposed to be a car driveable with a regular car license! If you need extra train to drive it properly, be explicit.
there's a 1992 wrap up book on that: https://www.cambridge.org/core/journals/robotica/article/abs...
all a bit before Boston Dynamics.
There was an optional 'AOA disagree' system that an airline could buy that could help pilots know when the MCAS was going crazy. US airlines, perhaps having more money, may have bought those (helping pilots with situational awareness), but airlines in developing countries (with presumably less money) may not have gotten them.
See perhaps §6.4 about Boeing giving that functionality to everyone:
* https://www.faa.gov/sites/faa.gov/files/2022-08/737_RTS_Summ...
Very different from how a pilot has to handle strange situations. Being ready for anything in an airborne plane without a pause button is so much harder, impossibly hard, and not every air authority tries as hard to reach the impossible.
This is not just filling out reports and looking at stuff, they're in no way comparable to your local garage mechanic (and not to dump on them either: they too have to deal with out of the ordinary situations).
The responsibility issues are the same as with the pilots as well, they fuck up people die.
Also what fraction of engine test cell use is for engine maintenance? Is it a big amount?
But if that kind of test goes wrong the main outcomes are "hit stop" and "oh no it's too late". An emergency like that is not where much of their expertise is needed, their expertise is in other parts of their job.
After every overhaul. Typically every 2500 to 15000 hours depending on the type of engine and the workload. It depends on many economic factors whether or not an overhaul is economical, in some cases it is cheaper just to buy a new engine.
Boeing knew of the flaw, and sent a letter to airlines about it. In 2011.
There's a lot of gray going on here.
Doesn't seem like gray to me. It seems a company who has a history of cutting corners and ignoring or downplaying safety problems did exactly that in this case too which resulted in the deaths of many people. UPS made an error here as well in trusting Boeing when they said it wasn't a safety issue and they should have installed the revised bearing assembly out of an abundance of caution, but I don't know much they would have known back in 2011 about the changes at Boeing that prioritized profit over safety following the merger with McDonnell Douglas
Actual question: would an airline have the engineering competence to second-guess an airplane manufacturer's engineering guidance? They operate airplanes but don't build them, and I'd assume they'd out of necessity need to trust the manufacturer's judgement.
Airlines have every reason to be skeptical of their supplier even if they do not have the engineering competence to second guess them. They could for instance look through their past communications with the manufacturer and see for themselves which advisories they agree with because for instance they are obviously not safety critical, this would then allow them hire specialists to evaluate the remainder for a second opinion.
For almost any act, we rely on other people. That doesn't absolve us of our personal responsibility.
But I'm just some guy with no incentive to endanger human life if I think it will save money so what do I know
Frankly I put it squarely on Boeing.
No, UPS bought the plane from Thai Airways International.
> Boeing built them
No, McDonnell Douglas built the plane in question; Boeing hadn't merged with MD at the time this aircraft was manufactured.
The other elements are probably true, but this was not a Boeing aircraft.
Why did you need to tell me about the wire then?
The answer is an attempt to transfer the liability to me. The liability for a thing they think could happen, but didn't tell me about.
And the worst thing is I don't think even after mcas things have substantially improved there. I've seen more spin and damage control than actual safety focus. They could have launched a huge company program and management reorganization to really turn this mindset around.
I think the biggest issue that Boeing is too big to fail. They'll never fall because the government needs them for all their warplanes.
This is how you get mentally and morally weak bean counters running companies instead of engineers with a conscience. It’s an engineering company and yet it’s run like a bank that just so happens to have an engineering branch.
$5mln? $100mln? Old school, $50?
In the US we've done a pretty poor job of doing that and it's resulted in countless lives lost and every living person and animal on earth being poisoned. It's long past time our government and its legal system took their responsibility to public safety more seriously.
I’m sorry, but this phrase has worn out its welcome.
In aviation, there is little room for error. It’s also the case that resources and time are limited. So there are multiple constraints.
We both agree that Boeing is the big problem. I’d also say its a problem of the FAA and the aviation industry.
But UPS? Why would they be taking action “out of an abundance of caution”?
The worst you can say for UPS is they could have sought a second opinion out of “an abundance of caution”, and recommendations of next actions and how.
Keep in mind UPS core competency isn’t aerospace and aeronautical engineering.
Would they even be able to assess the risk of changing said bearings en masse?
The actual lesson here is that most of the advisories and self-certifying from Boeing over the past 30 years need to be reconsidered; most likely redone, by independent third parties and also an FAA with a mandate to be fully independent.
Seems like a perfectly fine phrase to me.
Because my naive conclusion after looking at the part in question is exactly the same "would not result in a safety of flight condition." if the bearing cracked at the point in question it is going nowhere, the bearing is still captive in its housing. hell it looks like it could have been designed as two pieces and it would work the same. the large bolt is what is holding the engine on.
The best I can come up with is that a split bearing causes increased wear on the mounting bracket and nobody noticed for a long time.
Anyhow, here is the ntsb update in question https://www.ntsb.gov/investigations/Documents/DCA26MA024%20I...
So if that bearing went that's not quite a smoking gun yet but it would definitely be a step closer to a root cause.
With a proper tolerance bearing in place, the force is constrained so that other parts are only stressed in directions they're well suited to handle (because the bearing takes the load).
Once the bearing develops excess tolerance, you've got a bucking engine that (to your point) is directly loading other parts in unexpected ways/directions, eventually causing failure.
The fact that Boeing supposedly modeled this and came up with non-safety critical in the event of bearing breakage... curious how that will turn out.
They'd have to show at least one plane with a bearing gone that still flies as intended. I suggest we break one on purpose, put the full complement of Boeing execs on that plane to prove its safety given the alternative of retracting that statement.
That depends on the meaning of “safety of flight”. I don’t know what it means in aviation, but do not rule out that there is significant room between “flies as intended” and “result in a safety of flight condition”.
For example, if an engine were to complete drop off the plane, would that necessarily result in a safety of flight condition, or does “the plane will be able to continue take off and land again” mean safety of flight isn’t affected?
But a takeoff does seem like the worst time to catastrophically lose 1/3 power, even without FOD intake by the central engine.
This same scenario combined with the amount of vibration and stresses caused by the engine, should scream "this is a catastrophe waiting to happen" for any engineer.
The bearing would have to sieze up and the bearing axle be locked to the race. There is some limit to rotational torque even with a siezed bearings.
Metaphor: arthritic joints are not smooth, but they will rotate if given enough torque.
From the images, it looks like the bearing had siezed. So presumably rotational vibration was transmitted to airframe and the vibration caused structural failure?
I'm assuming it is not an issue of extreme rotational torque causing the issue (and given it is a bearing the design is for very little torque there!)
IANAME (not a mech eng)
As for your 'limit to rotational torque': seized bearings do not 'rotate if given enough torque' they will break right out of their casings and whatever those casings are surrounded by. The reason is that unlike your cartilage the bearings are orders of magnitude harder than the materials around them. For a bearing to seize indicates that the material has already deformed, you either catch it before the race goes or it will crack and after that all bets are quite literally off. I'm not aware of any design that would spec a bearing in a situation with such forces that would still happily work with that bearing replaced by a bushing welded to the shaft and the surrounding material even if it is statically in exactly the same position.
What you describe is a worn bearing with an excess of play, not a seized one, which tends to exhibit roughly the same characteristics as a welded joint with dissimilar materials.
Bearings are wear items, bearings that are worn or seized are something that should never ever happen in an aircraft, there is no way that this particular design would continue to function with sufficient margin if that bearing would fail. If not caught before it breaks the next flight is going to be a disaster. Take off in a fully loaded aircraft of this size puts extreme stress on the engine mounts. They are designed with all of their parts in working order, this is not a case of 'oh, we'll fix that the next time this craft is in for maintenance'. All parts of a plane that is certified as airworthy are supposed to be operating as originally specified.
The default assumption is that it all looked good during the last inspection and that the time between the failure occurring and the plane going down was short. If it was not that would be highly unexpected. But again, until the final report is in that's speculative, and if anything the people at the NTSB are scary good at getting to root causes.
Yeah. Worn or seized bearings are relevant to rotation, but on second thoughts, rotation isn't the issue here.
Rereading the PDF, I can see that I entirely misunderstood the function of the bearing and how it failed, and I suspect I've mislead you. The two lugs mislead me! I would guess they make the lug as two parts for redundancy (if the lug was a single part then it's failure would be bad). My previous comment was wildly incorrect about rotation, but now I think rotation is not the issue.
The casing split in half all the way around the circumference at the weakest point (where the recess is), splitting into two pieces, a forward half and a rearward half. The half forward of the split moves forward and the half rearward of the split moves rearward. That is what they inspect for every sixty months to see if the bearing casing has broken.
An unbroken casing is normally prevented from moving forward or backwards by the ball (how the hell do they make the bearing like that?!).
It appears that the unbroken casing itself is designed for the outside to be able to slide forwards and backwards within the lugs (very little movement?).
The primary force this bearing is preventing is pitching of the engine relative to the wing (vertical force). And secondarily to prevent yawing of the engine relative to the wing (horizontal force). Rotation (roll of the engine relative to the wing) has to be prevented by the main mount and the engine surely can't twist much therefore I suspect rotational forces at that bearing are rather irrelevant.
As the engine thrusts and stops thrusting, the thrust changes create pitching forces on the engine, and there would be vertical movement at the broken bearing - a clunk!?
The main mount would flex a little more due to the extra pitch movement; and I guess we'll have to wait and see whether the bearing failure is relevant to the crash. It appears to be a smoking gun, but could be a red herring?
The main mount is obviously not supposed to fail even if that bearing has broken.
The NTSB investigation found that for this crash, not only did the bearing race crack, but also that the bearing lugs, which hold the bearing in place, were fractured. I don't have access to the original text of the letter Boeing sent out, but based on the NTSB report, it sounds like only the issue with the bearing race was previously identified. The two may very well be related, but that doesn't mean that the lug fractures are an expected result of the race failure - perhaps some contributing factor made the lugs more susceptible than predicted. It also remains possible that the bearing damage is a red herring; the aircraft was nearing the end of its service life and had known structural issues in other parts of the pylon. The fact is that for more than a decade after the bearing race issue was reported, it didn't result in a safety of flight condition.
The insinuation that Boeing was deliberately trying to hide or downplay a known issue is simply unwarranted. It would be irresponsible for the NSTB not to mention a known issue that could have potentially been relevant, it's not evidence the issue was improperly handled.
[1] https://www.ntsb.gov/investigations/AccidentReports/Reports/... [2] https://admiralcloudberg.medium.com/rain-of-fire-falling-the...
I also expect that they were much less complex than an aircraft that provides a comfortable, pressurized cabin; the high level of safety mentioned above; freight capacity; etc.
Also, despite Boeing's recent problems, I would guess that commerical passenger planes are far more safe than they were decades ago when the SR-71 was developed. Accidents were much more common despite many fewer flights, iirc.
Obviously they could have designed something that could expand and contract if they thought it was worth it.
Also, this was done because airframe skin temps exceeded 400F during flight due to the high speeds.
Even commercial passenger flights are not pressurized to sea level; I think it's something like 8,000 ft. IIRC, Boeing's 787 was designed to be pressurized a bit more which, from on-the-ground experience acclimatizing to altitude, I think could make a noticeable difference.
They have many other unprecedented expectations, such a fuel usage and safety.
It was also insanely expensive to operate: $300k/hour in 1990 dollars, and there aren’t reliable numbers on development costs with all of the black budgets.
I know satellites and drones have replaced the sr71 but it would be cool if someone would build a plane as capable again.
Spy satellites are as of yet off limits.
U2 is still in operation.
UPS is as old as the plane in question and has only had three fatal accidents in that time with millions of flight hours, most of them on retired airliner frames.
Yes, Boeing had a monumental fuck up with the MAX redesign. However, their last blank page design was the 787 and is seen as completely revolutionary in terms of materials and efficiency. Let’s talk about that plane. It burns 20% less fuel than the planes it was designed to replace, and has a number of incredibly impressive engineering feats purely for passenger comfort- pressurization altitude and window size being the most impressive. It doesn’t sound impressive, but the design ask is: make a lighter plane, with bigger holes in the structure, that can withstand more pressure, and use a material and process that has never been used before. The only fatal incident on the 787 is still under investigation, but is almost certainly pilot error or suicide. Other plane and engine safety technology have allowed ETOPS making it possible to use efficient twin engine jets operate overwater flights that would have been unthinkable 40 years ago.
Jets today are quieter (by such a huge margin that it isn’t legal to operate the original 707 engines at most western airports), more efficient and safer than ever.
In the era that the SR-71 existed in, it was actually pretty common for planes to crash due to design defects (DC-10, Comet, 707, and more). The 737 MAX defect was so shocking because it has been 50+ years since that was common.
The SR71 is a simpler plane in many ways than a modern airliner. The composite technology to build a 787 didn't even exist at that time, and the engine alone on the 787 is far more impressive engineering and material science than the SR71. And there are two companies that figured out how to make them without a blank check from the CIA. The 787 produces more than double the thrust of the SR-71, and most passengers barely are aware of the miracle they are participating.
The SR-71 is an undeniably cool project. I have seen several up close, sat in the cockpit and they are literally awe inspiring. What we build today are airliners that are seemingly boring but built and designed with technology and materials that Skunkworks couldn’t have even attempted.
We aren’t building things like the SR-71 anymore because we are building things that are far better and more complex. We have Lockheed producing the F-22 and F-35, multiple companies reusing space launch rockets, etc. the real problem is that we have lost our sense of wonder at just how impressive modern aerospace engineering is.
To reduce negative outcomes, we use risk management: assessing the likely lifetime cost of the flaw, and taking cost-effective measures to reduce the risk to an acceptable level. As a familiar example, redundant mass storage drives are much more cost-effective than high-reliability mass storage drives.
And the DC-10 was not decommissioned. It is, in fact, still in service.
https://www.flightglobal.com/safety/us-faa-broadens-md-11-gr...
Aviation rules are written by blood, you either follow them or you add a few more lines with your own blood.
Please, what fool subjects their own blood to the absence of regulation? If you've got blood on your hands, much better for it to be a customer that has already paid you.
The fundamental reality is, we can always spend more to prevent another death; and we must draw the line somewhere.
People don't like it, but your latter example is the risk management I'm talking about, and it's unavoidable. Nobody can make airplanes that have no risk of killing people - the only answer would be no airplanes at all (which would result in more automobile deaths, more deaths because life-saving resources are unavailable, etc.). The calculation of cost per death prevented is a real one, and is done by manufacturers of planes, cars, etc.
The problem with your wording is the criteria of paying damages, rather than the industry-standard value for human life. Again, that is awful to think about but there's no way around it.
Edit: And I'm not saying, at all, that Boeing made the right decision here. I don't know enough to say, and Boeing's safety reputation is poor.
And "shit happens; suck it up, buttercup" is not an approved PHA determination.
Usually this is because the design constraints are complex and in satisfying one you wind up having orders of magnitude more overkill than you need on others.
For example, in situations involving hollow shafts with through shafts or perhaps fluid passages often times you wind up with insanely huge for the load bearing supporting the outer most part because it simply needs to be that big in order to fit around the shaft and have space for reasonable sized roller elements for the speed and realistic race thicknesses, etc. Sure you could go custom, but $$, sure you could use needles or balls, but maybe the stuff on either side has reasons it shouldn't be hard like a race and that might add assembly/construction cost. Now say this overkill bearing is held up by a big web in a big honkin cast housing, because the housing needs to be like that for structural reasons (say it's a specialty pump or maybe this housing is load bearing in the overall assembly, like a tractor's gearbox). Now, say this bearing is in some more complex gearbox that has lubrication windage problems. A valid fix might be to go and cut out a chunk of the web that holds this bearing. Sure it's only supported by 300deg now instead of 360, but it was so overkill to begin with that doesn't matter.
Edit: better example: You can roach dozens of automotive cartridge style wheel bearings without hurting the knuckle it presses into because the knuckle has to be so strong to withstand suspension forces you basically can't apply enough force via the wheel failure to break it and the assembly becomes unserviceable faster than you can get to the point of damaging it by wearing through it.
Edit2: You also need to consider the cost of QA and testing. Sometimes it's cheaper to do a simple overkill waste of material design than something than to do speed holes and engineered webs, etc, etc, because all those features add testing cost as well as manufacturing cost and (especially in ye olden days of the slide rule) make it harder to predict stuff like resonance, exact failure mode, etc, etc and every feature has to be QA'd to some extent. And this all needs to be balanced against expected production volume.
[1] https://www.ntsb.gov/investigations/Documents/DCA26MA024%20I...
We determined that the probable cause of this accident was the in-flight separation of the left MED plug due to Boeing’s failure to provide adequate training, guidance, and oversight necessary to ensure that manufacturing personnel could consistently and correctly comply with its parts removal process, which was intended to document and ensure that the securing bolts and hardware that were removed to facilitate rework during the manufacturing process were properly reinstalled.
The part was redesigned without the groove but wasn't mandated because Boeing said even the old part could be used, which is insane. Clearly, the new part wasn't installed and likely 1-2 inspections failed to notice it was broken.