[cyberferret]

It's a stress test. You would expect things to break during a stress test.

> "They loaded it up well beyond capacity and bent its wings in an extreme manner, in a way almost certain to never happen in the real world."

Here is another video of a 777 wing test where it broke at 154% of the [s/normal/maximum theoretical/] load - a bit beyond what was expected by the engineering team. https://www.youtube.com/watch?v=Ai2HmvAXcU0

[cyberferret]

Also - need to ensure that all testing methodologies were up to scratch. Here is an article about a KC-135 pressure test that ended up making the plane pop like a balloon [0] - all because the pressure gauge used by the tester didn't have a 'stop' peg and he missed the fact that the needle had cycled around.

A door hatch from the aircraft was found 70 yards away.

[0] - https://theaviationgeekclub.com/time-kc-135-stratotanker-air...

[kbenson]

> Here is another video of a 777 wing test where it broke at 154% of the normal load

I really hope you mean 154% of the max load, and I hope that max load is engineered to be far above what should be seen in actual service. 50% over the normal service load is not what I would consider extreme, and not something I would expect to be acceptable in an aircraft.

[cameldrv]

Design limit load is for an aircraft at max allowable weight, with the max authorized g loading (often -1 to +2.5g) or the worst anticipated gusts (about 50 ft/sec depending on altitude). You then need to tack on 50% to get the required ultimate load.

50% is indeed not a huge safety margin like you'd get with a bridge or something, which is why you need trained professionals flying it and regular inspections. If you fly into a big enough thunderstorm, the wings can definitely come off. Of course a bigger safety margin would be nice, but every pound of wing spar is a pound less payload. If the FAA required 200% of design limit, flying would be significantly more expensive.

[cmurf]

This is why there's maneuvering speed (VA), or turbulent air penetration speed (VB). It isn't necessary to over design it, you just slow down and, voila, the airfoil will sooner stall than break - and the stall is brief, feels like sluggishness, like a squishy slow stall, not abrupt, and recovery is similar perhaps a bit more abrupt as the airfoil regains lift.

[cameldrv]

Sure, as long as the gust is mostly vertical (usually true), and you're able to maintain below Va. All of the up and down motion in a serious thunderstorm can easily cause you to exceed Va or even Vne. Many aircraft have broken up in flight because of this. Now it could be that in many accidents, it wouldn't have mattered how strong the wing was -- eventually the speed build build up to the point where a big gust would break it, but I'm sure there are many where say an extra 50% strength would have made the difference.

Now that's also not to say that if you're going to add weight to an airplane in the name of safety that the best place to do it is the wing spar, but there is certainly a tradeoff between safety and weight.

[fluffything]

> 50% is indeed not a huge safety margin like you'd get with a bridge or something

Safety margins for bridges are like 25-50%.

[cameldrv]

That's interesting. I know that they used to be a lot more. Perhaps we should be glad that most of the bridges we drive over were built before FEM, and they just used a lot more steel. I wonder how well a bridge designed with a 25% safety margin is going to perform when it's poorly maintained for 50 years and rusting like a lot of bridges in the U.S.

[fluffything]

The reason this works is that the safety margins accumulate. You have a safety margin on:

* the loading on the bridge, which can be higher than the calculated amount.

* the resistance of the materials used, that can be lower than the requested amount.

* the quality of the soil,

* the lifespan of the bridge,

* ...

So the loading can be 25% percent higher, the materials can be 25% worse, the soil can be 25% worse, the lifespan can be 25% higher, ... and at the end you get a bridge that would still work if everything fails with a margin of 25%.

If the loads are 30% higher, the bridge might or might not fail. Who knows. Even though there are many other safety factors, and the bridge probably won't fail, it was not designed for that. There might be a critical part somewhere that has a resistance that's 25% worse than what it should be, and it therefore can only support a 25% overload, and with 30% that part might fail.

Bridges are designed to fail very slowly, loudly, and visibly, to avoid loss of functionality (e.g. when a bridge start to fail, this becomes obvious, and you still have months or years to sanitize the bridge).

I expect airplanes to have much tighter safety margins (<15% or <10%). The loads are more accurately known, the materials are higher quality, everything passes more extensive quality assurance tests, weight is much more important and companies are willing to pay prime prices on materials, manufacturing, etc. to reduce it, etc.

Safety margins aren't chosen arbitrarily. The job of a safety margin is to quantify an uncertainty, and with enough money thrown at each one, most of the uncertainties can be quite precisely quantified. Then it's the job of the designer to say "This plane should be in service for 50 years", and from the uncertainties and statistical analysis you can compute the highest load that the plane will receive in those 50 years with a certain quantified margin of error, and continue the design using that.

[cyberferret]

Sorry, I mistyped. It is indeed the maximum theoretical load as calculated by the engineering team.

[dmix]

Specifically from the video description:

> This Boeing 777 wing was tested to destruction, finally breaking at one fifty four percent of the designed limit load.

[rtomanek]

I would still expect the `maximum theoretical load` to be well above the `designed limit load`.

[ChuckMcM]

And it is, it is 154% or slightly more than 2.5x the 'design limit load'.

[kortilla]

90% of $100 is not $190

[tempestn]

Slightly more than 1.5x you mean. And yeah, since the design limit load is already going to have a safety margin over the max anticipated load, as long as the test performs some amount beyond that design level, it's good.

Edit: This is assuming the 'design level' is the level it was designed to withstand, which is how I would normally understand the term. Having watched the full video now though, it sounds like it actually exceeded the expected max force that could be expected in the real world by 1.54 times, and that the design anticipated about 1.5x. So indeed, it hit dead-on the design expectation, rather than vastly exceeding it—but that's fine, because the design included a 50% margin of error (and one would hope, a conservative worst-case estimate).

[macoovacany]

Ok, I'm hijacking this thread because there's some confusion about 'limit' load.

http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgFAR.nsf...

[mkl]

You're not hijacking it, just adding some contextual information, namely: "Strength requirements are specified in terms of limit loads (the maximum loads to be expected in service) and ultimate loads (limit loads multiplied by prescribed factors of safety). Unless otherwise provided, prescribed loads are limit loads."

[rntrg]

Was that an airplane joke?

[RcouF1uZ4gsC]

Good point about it being normal for stuff to break during a stress test. However, it seems the door blowing off is not something that normally happens and that to me is more worrisome than if it was a common failure when the stresses greatly exceed the theoretical limits.

>Former Boeing Engineer Dr. Todd Curtis runs Airsafe.com and said this doesn't happen often. "I've never heard of a case where a door popped off like this during a stress test before," he said. "Doesn't mean it hasn't happened before, I'm just not aware of it."

[ajross]

You expect things stressed beyond limits to break where you expect them to, and the point of the test is to show that the failure modes are the expected ones. Having a door pressure seal fail explosively during an airframe stress test sounds to me very much like an UNexpected failure mode. That's bad.

[inferiorhuman]

It's doubly unexpected as, in theory, the 777X is just a variant of the 777, itself a plane that didn't blow its doors out under testing.

[dhimes]

in theory, the 777X is just a variant of the 777

Uh oh. Haven't we seen this story before with a different model (cough 737 cough)?

[tpmx]

"A door blew off a Boeing 777X as the new plane was undergoing what was supposed to be its final structural inspection by federal regulators."

It's literally the first sentence of the article.

I doubt Boeing were pushing the envelopes further than the FAA required. And a door fell out.

[Groxx]

The second, third, and fourth sentences seem to imply otherwise?

>The test is meant to push the plane beyond its limits. Engineers had the plane pressurized and on the ground. They loaded it up well beyond capacity and bent its wings in an extreme manner, in a way almost certain to never happen in the real world.

and also further down, more "not necessarily a normal test" evidence:

>... The ultimate load test is the latest in a series of tests that Boeing has been conducting on this full-scale test airplane over the past several months."

[inferiorhuman]

The second, third, and fourth sentences seem to imply otherwise?

Semantics.

https://www.seattletimes.com/business/boeing-aerospace/door-...

The Seattle Times explains it thusly:

During the ultimate load test, the wings are then pulled upward. To pass the test and be certified, the wings must bend without breaking until the load on them reaches at least 150 percent of the normally expected load.

…

Sometimes this final test is continued beyond the 150 percent load target until a wing actually breaks. But not always.

…

This time, however, though the wings did not give way; it was one of the doors that failed — an outcome that is definitely not supposed to happen.

[ImprovedSilence]

I mean, when it comes down to it, I’d rather lose a door than a wing though, right...?

[mannykannot]

Losing a door can have very serious consequences:

https://en.wikipedia.org/wiki/Turkish_Airlines_Flight_981

The cargo hatch blew open and the cabin floor collapsed on account of differential pressure, disabling the rudder, elevator and rear engine controls.

This may well not happen on a 777X -- various other airplanes have lost part of their cabin structure without crashing -- but an unexpected event calls into doubt your design.

https://en.wikipedia.org/wiki/Aloha_Airlines_Flight_243

[raverbashing]

It depends on the stress test

If you're testing it up to the limit then no, it is not expected.

But if you're pushing it beyond the design limit then yes, it is expected to break.

Some wing tests are done beyond the limit, some are done below the limit (non-destructive tests cost less, that's why they're preferred, but for a new design they might just push it beyond the limit to see when does it effectively fail)

[Havoc]

>You would expect things to break during a stress test.

Given the stunned silence and halting of further testing I'm guessing this wasn't a stress it till breaking type stress test

[97b683f8]

If I ever write a library to perform stress tests, I'll call it one-fifty-four.

[anigbrowl]

That's...not something any client wants to hear.

[Groxx]

if they're interested in knowing non-theoretical failure limits, and how it behaves under failure: yes, of course they want to hear that.

[jussij]

> You would expect things to break during a stress test.

Except, the whole point of the test is to stress the plane and to have nothing break.

The test is designed to make sure the aircraft has built-in safety (i.e. is over engineered).

When the A380 was attempting to earn it's certification it failed a similar static wing test:

http://www.airmech.co.uk/forums/showthread.php/5340-A380-Win...

[kbenson]

> Except, the whole point of the test is stress the plane and to have nothing break.

That depends on the type of stress test. Some stress tests are to confirm it acts as expected. Other stress tests are used to determine modes of failure. Good engineering isn't just building stuff to spec and assuming it works, just as good programming isn't just writing code and assuming it works as expected because the compiler doesn't throw any errors.

It's very useful to try to break something and see what happens. That's why engineers provide more stresses than systems are designed for to see what breaks, and that's why programmers sometimes run programs in environments constrained in ways it wasn't meant to run in (no disk space, out of memory, network drop, etc), and run fuzzers.

[agumonkey]

Also it depends on how far in the design process they are. Early failure is good. Late is bad, but mostly for them, they'll have to loopback.

[i_am_proteus]

Very importantly, this hasn't been described as a "test failure."

No indication of what the test criteria were, if any. Some tests, especially those past design limits, are just for informational data.

[inferiorhuman]

Very importantly, this hasn't been described as a "test failure."

That was heavily implied:

Sources tell KOMO there was a stunned silence after it happened.

…

an event occurred that forced the test team to halt testing. … The team is currently working to understand what happened and ensure the area is safe for work to continue.

Edit: You're not going to silence the crowd with a successful test and you don't temporarily halt testing because of a successful test.

Edit 2: Seattle Times says this explicitly: This ground test failure is another blow.

[i_am_proteus]

Yes, but this isn't the language they'd use if one of the test success criteria was "no door blows out" and then a door blows out.

This is evidenced by their plan to "analyze the door and run the test again." If this were actually a test failure, they'd be "analyzing the door, redesigning a bunch of stuff, and running the test again."

[txcwpalpha]

>Except, the whole point of the test is to stress the plane and to have nothing break.

What? No it's not. The entire point of these tests is specifically to break things. If nothing breaks, they did not perform the test correctly.

>When the A380 was attempting to earn it's certification it failed a similar static wing test:

No, it didn't. It passed the test handily. The wing broke well beyond the max load limit, which is what they were testing for.

edit for the sake of being less of an argumentative asshole: I see where you're coming from when you say that it was a 'failure'. However, what I mean to say is that while the wings failed in the test, the failure was intentional, and the data gathered during the test was exactly what was needed to prove that the A380's wings were strong enough for certification, and the A380 was certified based on the results of this test. In my mind, that means the test passed.

[inferiorhuman]

What? No it's not.

Yes, it is. The point is to check that the wings can take 150% of the rated load. The point of an ultimate load test is NOT to break things.

[jussij]

The first time the A380 tried the test it failed. Read the link.

And yes it did eventually pass that wing test only because if it had not it would not never have been certified.

[txcwpalpha]

>The first time the A380 tried the test it failed. Read the link.

I suggest you read it. The A380 did not fail. It says so in the link you provided.

[jussij]

"We'll be able to demonstrate" that the plane meets requirements, "maybe with some refinements needed for certification in time for first delivery at the end of the year," Kracht said in a telephone interview Thursday.

We'll is we will (i.e some time in the future).

[txcwpalpha]

The A380 was already in production at the time of the Feb 2006 test. No modifications to the wings were required because it did not fail it's structural stress test. The entire point of the test was to gather data on what load the test wings (which are weaker than the production wings) failed at, which is why they intentionally broke the wings. The engineers then used this data to calculate the load limit that the production wings (which are stronger than the test wings) failed at. The calculated load limit for the production wings was above the EASA certification requirement, meaning the wings passed their test.

[jussij]

> The entire point of these tests is specifically to break things

When you do these types of tests you are testing for a expected outcome.

If you do a test and the outcome is un-expected (i.e. the fuselage fails prematurely) it means your engineering and the expectations built around that engineering are wrong.

[txcwpalpha]

This is incorrect. These tests are specifically done because they want to gather data on what the failure loads (which are previously unknown) are.

[inferiorhuman]

No, as the Seattle Times pointed out the intent of the test is to ensure the wings support 150% of their rated load. They may continue until things break, but the typical purpose is simply to ensure there's that 50% wiggle room.

[29_29]

I'd rather it break in a stress test - than when I'm on it.