Door blows off 777X in stress test

...

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.

As the ground test was underway and as engineers and FAA inspectors watched, a door blew off the plane.

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

Dr. Curtis said this is not the time to race to conclusions, and it could be something totally innocuous that caused the door to come off. But it could cause delays. "It's unlikely this will speed up certification," he said. "It's more likely it will make the certification team, whoever's involved with Boeing and the FAA, do extra work to figure out what happened.

Investors were disappointed recently when Boeing announced delays in delivering its first 777X model into next year.

link


Have done some "testing to destruction" and survivability testing. Not entirely uncommon to see some unplanned failures in those, but during final stress test trying to get a design certified by the FAA is not when you want to see that. In the stress test, you should have a good idea of what is going to be the outcome. "Unlikely to speed up certification" is a British-sized understatement.

a reply to: RadioRobert

Even if they got certification today they can't fly until at least January according to the GE timeline. Yes, you want to get everything done, but barring this turning into a major failure requiring a design change, it's probably not going to hurt them badly.

a reply to: RadioRobert

"Unlikely to speed up certification" is a British-sized understatement.

Rawther ...




Cheers

a reply to: Zaphod58

Having a door seal fail is different than the stress transferring enough energy to literally blow a door off the hinge. It means you're getting fuselage twisting and bending. Depending on the conditions under which it failed, you may even have a wingbox problem. It's not necessarily the end of the world (again depending on where this extra unaccounted stress is being transferred from), but instead of being one-and-done, they will be doing this a minimum of twice again, I'd bet, and probably a redesign. Maybe minor; maybe major.
They're supposed to begin delivery in summer of 2020. They already have a production gap. It's going to be a mess. Depending on the damage, you might be cannibalizing one of the prototypes for new stress tests to keep two airframes available for statistic tests. They have to decide what to do with the production line employees through the gap. Do you slow other deliveries of 777F or -300's so you don't have an idle line? Furlough or reassign those workers during the gap and hope they come back when you're ready to transition to the X? No easy decisions for the bean-counters on this one.


MAX deliveries are still suspended. P-8 and KC-46 have been ongoing nightmares. Wouldn't buy stock anytime soon.

a reply to: RadioRobert

The 747 had a door latch issue where the door showed closed and locked, and the latch was only halfway engaged. There could be manufacturing issues involved, and the latch failed. It's far too soon to jump to "there's a serious design flaw". The original 777 had a similar problem, and it was fixed fairly quickly and hasn't been a problem since.

a reply to: Zaphod58

Simple unengaged latch failure doesn't carry enough stress energy to blow off a door and shake the plant.

a reply to: RadioRobert

Probably not, if it failed the way the 747 door did. If the latch broke it might. There's no reason to jump straight to design flaw before they even look at the door.

At least it wasn't some new feature like the Lockheed Constellation astrodome.

Well a latch and the hinges would have to both break simultaneously or in rapid succession to separate the door from the fuselage... They aren't designed to do that...
The 747F incident at pre-delivery was latch failure during the high blow. It didn't separate from the fuselage; it just banged up against it.

Just read this was also part of a high-blow. Means the fuselage failed under less than 15 psi. It wasn't part of the flex tests. That's good because it does not involve transfer of stress from wings to fuselage which would be terrible. That's still really bad because it means your fuselage pressurization failed at a point it really, really should not have. The 747F failure on delivery was under 5 psi and still resulted in damage to the airframe. Side hatch is not as likely to create damage, if the pressure was low. Hard to imagine complete separation under low pressure. If it was higher, then might have been fuselage expansion which led to the door failure. There is no good look for this, unfortunately.

Even if this is a workmanship issue (say they installed the hatch or some fasteners incorrectly) and not a design-flaw, how does this happen on your test aircraft? You have the Dreamliner deliveries with improper fastener installation, both civil and military customers refusing delivery because of FOD problems, the well-publicized MAX ongoing issues, P-8 way, way behind schedule, the engine delays to the 7X (which isn't really Boeing's fault, but remains their problem). It's a mess.

"The test".....you do know they find hundreds of problems on new aircraft designs don't you?? that's why they test them...

a reply to: manuelram16

Yes. I've even done testing... This particular test is not a test that should fail at this point in the game. This should have been caught well before this point if it's a design failure. And if it's a workmanship issue and not design related, it'll be the latest in a string of them.

I've seen tests fail. Even on occasions when there should be no failure such as this one. I've experienced the stomach-drop and stunned silence. It's not a good thing. It means a mess and lost money and time. My personal most memorable "This should not be happening, wtf is happening to make this fail now" moment traced back to workmanship and was a colossal # up that cost millions in waste and a bunch of money on trying to make sure it couldn't happen again, and took more than a month to find where it was originating.

It's not unique to Boeing, but Boeing has a giant pile of high-profile messes right now on projects that should represent their best feet foreward. There is an institutional problem there that has to be sorted, and the sooner, the better for everyone.

a reply to: RadioRobert
I actually saw a photo a couple of days ago of the high blow test on the prototype 744ER taken back around 2000/2001. The guy who had the pic was present as a customer engineering rep. You can see the airframe bulging between every stringer and frame so much it almost looks like a chocolate bar. He said every window including the cockpit #1's could be seen bulging. At the time they were standing only about 15m away too which is scary if not unwise. When he asked the guy running the test how many he had done before he simply replied "all of them". Apparently he had done every test since the beginning of the 707 program and had gone on to do 727, 737, 757. When he was asked if he had ever seen failures he said plenty, he once saw a hatch on a B-50 blow off and hit the roof of the test hangar.

This latest set back in the 777X is certainly going to strengthen the case for putting the -8 variant on ice for a while. It had already been announced mid August that it would be delayed but I'm wondering if the entire thing may be scrapped, at least as a passenger variant. Sales of the short body were nowhere near as much as the larger -9 and pretty much limited to the big Gulf carriers. If anything I can see the idea of a further stretched -10 model having a better business case. The only other likely decent market for a short bodied 777X is as a freighter, however Boeing could probably convince freight operators of a -9 variant that would offer growth room, having only one model to produce would offer production savings. And if they do produce a further stretch in a -10, its generally easier to go up rather than down in size. I would expect that given the timing this will pretty much turn the Qantas "Project Sunrise" into a one horse race, especially if Airbus can do a viable A350-1000ULR. The Sunrise project was the only other likely order seen for now for the 777-8 and even that was probably no more than 8-10 aircraft at best.

a reply to: RadioRobert

I do have a docu where they talk about the testing that takes place during design, they pressure test a fuse and hold & behold the cargo door fails....(maybe it's the same)
There is also an instance where a wiring harness falls short a couple inches and can't attach properly so back to the drawing board and make it again....

a reply to: manuelram16

Yes, and sometimes things are tested "to destruction". If it's the last test to be performed, and you've accomplished your goals, then one might sometimes say, "well, we know it works and takes abuse. Let's see how exactly how much abuse before she fails." What you don't want is a failure before you meet test goals which leaves you with incomplete testing and a damaged fuselage.
We'll see how it all shakes out and where they lie the blame in time.

I watched a pre production pickup go through stress testing many years ago and after about 100 hours it was stopped. But it was PRE Production sometimes you need to spend the extra dime.

a reply to: RadioRobert

I was there for a test to destruction test in 1989. It was a wing flex test. It made one of the loudest noises that I ever heard.

a reply to: JIMC5499
Out of interest what airframe type did you witness the ultimate load test on? That would have been before the 777 so I'm curious.

originally posted by: RadioRobert
a reply to: manuelram16

Yes, and sometimes things are tested "to destruction". If it's the last test to be performed, and you've accomplished your goals, then one might sometimes say, "well, we know it works and takes abuse. Let's see how exactly how much abuse before she fails." What you don't want is a failure before you meet test goals which leaves you with incomplete testing and a damaged fuselage.
We'll see how it all shakes out and where they lie the blame in time.

It also will validate whatever model you are using in that the predicted point failed at the predicted load etc