6 ms·

So, if I'm parsing this correctly, this is a requirement to immediately do a one-time inspection that's normally done periodically anyway. I'm all for safety,

by dlgeek 6y ago

So, if I'm parsing this correctly, this is a requirement to immediately do a one-time inspection that's normally done periodically anyway. I'm all for safety, but this has a vibe of "Something must be done! This is something... so it must be done!"

throwawayboise 6y ago
Yes, it's something that's already done at intervals. The implication is that maybe that interval is too long. Also from TFA: Based on the initial results as we receive them, as well as other data gained from the ongoing investigation, the FAA may revise this directive to set a new interval for this inspection or subsequent ones.
- fennecfoxen 6y ago
  There's also the possibility that the usual inspection shop was doing a bad job of applying its standards with the necessary consistency and rigor, and that new inspections after engine explosions might see them paying more attention. See this interesting article on the specifics of the P&W fan blade inspection shop and the process, which the NTSB previously concluded in summer 2020 has hidden behind the label "new and emerging technology" for far too long, avoiding the scrutiny which might make it more reliable: http://aerossurance.com/safety-management/ndi-failures-b777-pw4077-fbo/ http://aerossurance.com/safety-management/ndi-failures-b777-...
  - TylerE 6y ago
    Possible, but less likely since this has now happened 3 times, on different (but otherwise basically identical planes) in an ~18 month time span
    - pdonis 6y ago
      > this has now happened 3 times, on different (but otherwise basically identical planes) in an ~18 month time span Wouldn't that make it more likely that there is some systemic cause, such as a flawed inspection process?
tialaramex 6y ago
The apparent failure is in this specific component and looks like metal fatigue. This inspection would detect metal fatigue in that specific component. It's the obvious place to look. It may be that they don't find anything. Coincidences can happen, if you fly enough jet planes, enough times, sooner or later something like this will happen. Or it may be that this inspection finds one or more further engines with unexpectedly degraded compressor fan blades. That gives them a pattern on which they can base a subsequent investigation, while also keeping passengers safe.
ohazi 6y ago
They're demanding inspections to collect more data, and will revise the inspection interval if deemed necessary. Turbines are designed so that fan/blade separation failures are contained within the engine casing. Uncontained failures are considered a really big deal, because turbine blades carry a shitload of energy, and whether the blade shoots off into the distance or shreds a wing is essentially luck. The FAA's response to this (uncontained) turbine failure is typical and expected. Edit: Apparently this may have been a contained failure. See child comments for details. Leaving the text above alone for context.
- shakezula 6y ago
  Yeah i can’t see how the parent comment is framing this as bureaucracy. This seems like a solid move for relatively little risk to gather much-needed data. I’m glad the FAA is at least trying. This is their purpose.
- cmurf 6y ago
  All of this is true except the assertion this incident is an uncontained failure. There's no evidence it was uncontained. Parts exiting the rear of the engine is a contained failure. http://avherald.com/h?article=4e35503b&opt=0 http://avherald.com/h?article=4e35503b&opt=0
  - ohazi 6y ago
    I only took a quick look, and this source doesn't seem fantastic, so perhaps it's wrong, but anyway... here's my source [1]: > Engine failures are defined as contained or uncontained. When engine debris leaves the engine, it is an uncontained engine failure. There appears to be a tear in the rear of the fuselage possibly from a fan blade strike. If this is the case, UA 328 suffered an uncontained engine failure. Edit: Yeah okay, after looking at the video in your link, I agree that it appears contained. [1] https://www.travelcodex.com/united-flight-328-boeing-777-engine-fan-blade-failure-update/ https://www.travelcodex.com/united-flight-328-boeing-777-eng...
    - mjg59 6y ago
bedhead 6y ago
Story of the last 11 months.
m463 6y ago
But "periodically" means every 6500 takeoffs and landings. That seems like an awful long time between inspections. I don't know how often international flights takeoff and land, but the interval might be decades.
- WalterBright 6y ago
  The jet engines on the ME-262 had a service life of 20 hours. The service life of the airframe was so short they didn't even bother painting them. My dad's B-17 was scrapped after 30 some missions for being flying junk, or around 300 hours. For comparison, the Boeing 757 was designed to fly around 65,000 hours, for comparison. A car is designed to run for 3,000 hours or so. Your car is very poorly made compared to an airplane :-)
  - thelean12 6y ago
    Uh, we also don't overhaul car engines every several thousand hours. That comparison is basically useless.
    - ReptileMan 6y ago
      After several thousand hours of work a car engine will have to have a lot of stuff replaced or serviced. Probably only the block, head, pistons and shafts will remain.
  - m463 6y ago
    I would think an inspection interval should be shorter than an overhaul interval. But I don't know how they do it.
mjg59 6y ago
There's an existing inspection period of every 6,500 takeoffs and landings[1]. The expectation is that it would take sufficiently long for a detectable flaw to turn into a dangerous fracture that this would be sufficient to notice it. But, given this failure (and presumably also because of the 2018 one), that assumption appears to be untrue. Checking all of them now and comparing any discovered defects to the number of cycles since they were last checked should allow a recalibration of how frequent checks should be (or, if there's an indication that flaw propagation is significantly faster than anyone expected, potentially stronger measures) [1] My understanding is that it's based on cycles rather than flight time because takeoff is the part of flying that puts the most stress on the blades - spending 10 hours in cruise isn't much worse than spending 2.
- rkangel 6y ago
  Or alternatively, the data might show that 10 hours in cruise IS much worse than 2, and just scheduling on cycles isn't sufficient (obviously they would have flight time data available as well to make that determination).
  - water8 6y ago
    One reason why turbines are much more reliable than engines is very few parts are ever accelerating. Once the turbine has spun up everything is rotating in unison unlike a crankshaft for example that is constantly changing direction / acceleration

by dlgeek 6y ago

throwawayboise 6y ago

Yes, it's something that's already done at intervals. The implication is that maybe that interval is too long. Also from TFA: Based on the initial results as we receive them, as well as other data gained from the ongoing investigation, the FAA may revise this directive to set a new interval for this inspection or subsequent ones.

fennecfoxen 6y ago
There's also the possibility that the usual inspection shop was doing a bad job of applying its standards with the necessary consistency and rigor, and that new inspections after engine explosions might see them paying more attention. See this interesting article on the specifics of the P&W fan blade inspection shop and the process, which the NTSB previously concluded in summer 2020 has hidden behind the label "new and emerging technology" for far too long, avoiding the scrutiny which might make it more reliable: http://aerossurance.com/safety-management/ndi-failures-b777-pw4077-fbo/ http://aerossurance.com/safety-management/ndi-failures-b777-...
- TylerE 6y ago
  Possible, but less likely since this has now happened 3 times, on different (but otherwise basically identical planes) in an ~18 month time span
  - pdonis 6y ago
    > this has now happened 3 times, on different (but otherwise basically identical planes) in an ~18 month time span Wouldn't that make it more likely that there is some systemic cause, such as a flawed inspection process?

tialaramex 6y ago

The apparent failure is in this specific component and looks like metal fatigue. This inspection would detect metal fatigue in that specific component. It's the obvious place to look. It may be that they don't find anything. Coincidences can happen, if you fly enough jet planes, enough times, sooner or later something like this will happen. Or it may be that this inspection finds one or more further engines with unexpectedly degraded compressor fan blades. That gives them a pattern on which they can base a subsequent investigation, while also keeping passengers safe.

ohazi 6y ago

They're demanding inspections to collect more data, and will revise the inspection interval if deemed necessary. Turbines are designed so that fan/blade separation failures are contained within the engine casing. Uncontained failures are considered a really big deal, because turbine blades carry a shitload of energy, and whether the blade shoots off into the distance or shreds a wing is essentially luck. The FAA's response to this (uncontained) turbine failure is typical and expected. Edit: Apparently this may have been a contained failure. See child comments for details. Leaving the text above alone for context.

shakezula 6y ago
Yeah i can’t see how the parent comment is framing this as bureaucracy. This seems like a solid move for relatively little risk to gather much-needed data. I’m glad the FAA is at least trying. This is their purpose.
cmurf 6y ago
All of this is true except the assertion this incident is an uncontained failure. There's no evidence it was uncontained. Parts exiting the rear of the engine is a contained failure. http://avherald.com/h?article=4e35503b&opt=0 http://avherald.com/h?article=4e35503b&opt=0
- ohazi 6y ago
  I only took a quick look, and this source doesn't seem fantastic, so perhaps it's wrong, but anyway... here's my source [1]: > Engine failures are defined as contained or uncontained. When engine debris leaves the engine, it is an uncontained engine failure. There appears to be a tear in the rear of the fuselage possibly from a fan blade strike. If this is the case, UA 328 suffered an uncontained engine failure. Edit: Yeah okay, after looking at the video in your link, I agree that it appears contained. [1] https://www.travelcodex.com/united-flight-328-boeing-777-engine-fan-blade-failure-update/ https://www.travelcodex.com/united-flight-328-boeing-777-eng...
  - mjg59 6y ago

bedhead 6y ago

Story of the last 11 months.

m463 6y ago

But "periodically" means every 6500 takeoffs and landings. That seems like an awful long time between inspections. I don't know how often international flights takeoff and land, but the interval might be decades.

WalterBright 6y ago
The jet engines on the ME-262 had a service life of 20 hours. The service life of the airframe was so short they didn't even bother painting them. My dad's B-17 was scrapped after 30 some missions for being flying junk, or around 300 hours. For comparison, the Boeing 757 was designed to fly around 65,000 hours, for comparison. A car is designed to run for 3,000 hours or so. Your car is very poorly made compared to an airplane :-)
- thelean12 6y ago
  Uh, we also don't overhaul car engines every several thousand hours. That comparison is basically useless.
  - ReptileMan 6y ago
    After several thousand hours of work a car engine will have to have a lot of stuff replaced or serviced. Probably only the block, head, pistons and shafts will remain.
- m463 6y ago
  I would think an inspection interval should be shorter than an overhaul interval. But I don't know how they do it.

mjg59 6y ago

There's an existing inspection period of every 6,500 takeoffs and landings[1]. The expectation is that it would take sufficiently long for a detectable flaw to turn into a dangerous fracture that this would be sufficient to notice it. But, given this failure (and presumably also because of the 2018 one), that assumption appears to be untrue. Checking all of them now and comparing any discovered defects to the number of cycles since they were last checked should allow a recalibration of how frequent checks should be (or, if there's an indication that flaw propagation is significantly faster than anyone expected, potentially stronger measures) [1] My understanding is that it's based on cycles rather than flight time because takeoff is the part of flying that puts the most stress on the blades - spending 10 hours in cruise isn't much worse than spending 2.

rkangel 6y ago
Or alternatively, the data might show that 10 hours in cruise IS much worse than 2, and just scheduling on cycles isn't sufficient (obviously they would have flight time data available as well to make that determination).
- water8 6y ago
  One reason why turbines are much more reliable than engines is very few parts are ever accelerating. Once the turbine has spun up everything is rotating in unison unlike a crankshaft for example that is constantly changing direction / acceleration