3 ms·

Disclosure: definitely not an airplane expert. This is a problem discovered during testing. It could mean testing is being done properly; probably it’s a bit l

Disclosure: definitely not an airplane expert. This is a problem discovered during testing. It could mean testing is being done properly; probably it’s a bit late in the dev cycle, but it’s not necessarily a bad thing; now it’s making news because of what happened with Boeing in the past. But not finding failures during testing maybe means you’re not testing hard enough.

Dylan16807 2y ago
Halting testing is a lot more drastic than "found a failure", though.
smaudet 2y ago
> But not finding failures during testing maybe means you’re not testing hard enough. Also not an airplane expert... Agreed that finding issues is (often) good (it means you are human and you also found your errors). But of course such testing, certification, is costly and uneconomical, and if you aren't allowed to fail, then you can't do the correct testing either... I suspect the issue is two-fold: Firstly, Boeing can't afford to fail, and hence can't test. Secondly, it is driven primarily by profit, not safety, which is part of why it can't afford to fail. Profit drives/motivates new features, efficiencies, and ultimately complex systems, which require exponentially more time to thoroughly test than simple, inefficient, easily verifiable systems. Falling apart because they are no longer profitable may not be a good thing either - sure, something could "rise from the ashes", but that thing will be in all likelihood, immature, poorly tested, and mostly incompatible with existing systems. The best bet would be to simplify, as much as possible, all systems, suppliers, etc, and to double, quadruple, perhaps even expand by an order of magnitude the testing effort(s). Perhaps even some sort of new New Deal could come about...
angry_moose 2y ago
I've done a couple rounds through aerospace or aerospace-adjacent over the years. This seems to be a certification test (https://www.seattletimes.com/business/boeing-aerospace/boeing-discovers-engine-part-defect-and-grounds-its-777x-test-airplanes/ https://www.seattletimes.com/business/boeing-aerospace/boein...), aka the "make-or-break, 0-failures allowed test before we get final approval to sell this aircraft, and we are 99.999% sure this test is going to go flawlessly". Failing is a pretty big deal. This should have been found far sooner in the process.
fransje26 2y ago
> But not finding failures during testing maybe means you’re not testing hard enough. This is not quite the same. This is finding a failed part that should not have failed, as it should have been designed to -at least- pass the flight test requirements it was designed to fulfil. And what this article failed to mention was that similar structural issues were found on 2 other test aircrafts.
AdamJacobMuller 2y ago
> The component “didn’t perform as designed,” What they are saying here is that the component was designed to X, but, not manufactured to those specifications. Depending on who made that actual component, this is either a failure of Boeing directly or a failure of Boeing to ensure that their suppliers have adequate processes in place to capture these kinds of problems. This is actually a much more serious failure than a design failure because the problem can't easily be systemically addressed, there is no single component they need to address now. Boeing can not simply say "we checked the other engine mounts in production and they are fine". Even saying that they addressed the issue with the production of that component is insufficient. The real root cause here is why any component which did not meet design specifications was accepted and allowed to be installed on the aircraft. Boeing will need to do a huge review to understand this and may need to inspect and replace huge numbers of other, seemingly unrelated, components.
lnxg33k1 2y ago
Not an airplane expert either, maybe Boeing hired a bunch of people from microsoft and started testing in production?

Dylan16807 2y ago

Halting testing is a lot more drastic than "found a failure", though.

smaudet 2y ago

> But not finding failures during testing maybe means you’re not testing hard enough. Also not an airplane expert... Agreed that finding issues is (often) good (it means you are human and you also found your errors). But of course such testing, certification, is costly and uneconomical, and if you aren't allowed to fail, then you can't do the correct testing either... I suspect the issue is two-fold: Firstly, Boeing can't afford to fail, and hence can't test. Secondly, it is driven primarily by profit, not safety, which is part of why it can't afford to fail. Profit drives/motivates new features, efficiencies, and ultimately complex systems, which require exponentially more time to thoroughly test than simple, inefficient, easily verifiable systems. Falling apart because they are no longer profitable may not be a good thing either - sure, something could "rise from the ashes", but that thing will be in all likelihood, immature, poorly tested, and mostly incompatible with existing systems. The best bet would be to simplify, as much as possible, all systems, suppliers, etc, and to double, quadruple, perhaps even expand by an order of magnitude the testing effort(s). Perhaps even some sort of new New Deal could come about...

angry_moose 2y ago

I've done a couple rounds through aerospace or aerospace-adjacent over the years. This seems to be a certification test (https://www.seattletimes.com/business/boeing-aerospace/boeing-discovers-engine-part-defect-and-grounds-its-777x-test-airplanes/ https://www.seattletimes.com/business/boeing-aerospace/boein...), aka the "make-or-break, 0-failures allowed test before we get final approval to sell this aircraft, and we are 99.999% sure this test is going to go flawlessly". Failing is a pretty big deal. This should have been found far sooner in the process.

fransje26 2y ago

> But not finding failures during testing maybe means you’re not testing hard enough. This is not quite the same. This is finding a failed part that should not have failed, as it should have been designed to -at least- pass the flight test requirements it was designed to fulfil. And what this article failed to mention was that similar structural issues were found on 2 other test aircrafts.

AdamJacobMuller 2y ago

> The component “didn’t perform as designed,” What they are saying here is that the component was designed to X, but, not manufactured to those specifications. Depending on who made that actual component, this is either a failure of Boeing directly or a failure of Boeing to ensure that their suppliers have adequate processes in place to capture these kinds of problems. This is actually a much more serious failure than a design failure because the problem can't easily be systemically addressed, there is no single component they need to address now. Boeing can not simply say "we checked the other engine mounts in production and they are fine". Even saying that they addressed the issue with the production of that component is insufficient. The real root cause here is why any component which did not meet design specifications was accepted and allowed to be installed on the aircraft. Boeing will need to do a huge review to understand this and may need to inspect and replace huge numbers of other, seemingly unrelated, components.

lnxg33k1 2y ago

Not an airplane expert either, maybe Boeing hired a bunch of people from microsoft and started testing in production?