Alaska Airlines flight forced to make an emergency landing... 82

[3DDave]

It was required for Alaska's peace of mind or they would not have done it.

A serious problem on a new plane requires an extraordinary investigation. Did what happen seem like a reason to ground planes? Because everyone else does.

 

[Murph 9000]

Hindsight is 20/20. Yes, of course it was appropriate to ground them after the loss of the door plug, and I'm a little concerned that the EAD and grounding does not also apply to the 900ER with apparently the same door plugs.

As I understand it, prior to the incident, the plane was not actually losing pressurisation, it was just appearing to be a fault with the automatic pressurisation system. The system has a high level of redundancy, and the ultimate contingency of masks and rapid descent to 10,000. They were following the FAA and manufacturer rules in how they dealt with it.

 

[3DDave]

It's not hindsight. It's knowing from past industry experience how a severe problem can originate from a minor problem, as the Lion Air crash clearly showed.

Alaska owns a piece of this.

Would you rather this had been discovered before the accident, or is it better to have the accident happen?

 

[LittleInch]

The pressurisation thing could indeed be red herring as it was apparently "solved" by switching computers. If there was a continued air leak, this would not have worked.

I'm pretty sure the 737s don't have an ECAM system for monitoring these things and telling pilots what the issue is and what the fix is so if the first action is try the other computer / monitoring system and the alarm goes away then there's not much record of it.

And intermittent faults are always a bitch to fix.

Having a large part of the aircraft depart in flight is indeed a reason to ground them. An alarm going off three times in 100 flights is not.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[3DDave]

The move to the backup gives time for the seal to be energized by the pressurization system, covering the fact that the seal may not be making initial good contact.

The pilots don't need diagnostic info - that info is available to the maintainers.

There are only a few things to check. Are the sensors working properly? Is the wiring complete? Is the reported problem a pressurization problem?

Rather than take it out of service to find the problem they gambled that it was not a real problem.

 

[LittleInch]

Now if there had been passenger complaints about a whistling noise or air movement next to the plug door, I would grant you it's unreasonable not to suspect something is wrong. But unless they stripped the plane back to skin you would never find bolts missing from a plug door.

So I think this is all very secondary as we don't know what they did or how long it happened for or any real details.

I'm sure someone somewhere will stand in front of a lectern holding four bolts and ask people how much they are worth. In Boeings case $13.5 BILLION dollars.

That's some expensive bolt.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Murph 9000]

I really do feel that it's unreasonable to expect Alaska to discover Boeing's failure here (assuming Boeing were the last to close the plug and fit the interior panelling over it). To me, it falls under the saying from the medical profession, "when you hear hoofbeats, think horses not zebras". Boeing/Spirit not properly assembling / securing the plug feels like a zebra to me, as far as dealing with the pressurisation system issues is concerned; I'm not convinced that it was reasonable for them to suspect a problem with the semi-permanently closed plug, based on the information and symptoms presented to them at the time.

 

[TugboatEng]

Just to throw this out there, it seems that the first assumption was that a leaking door may have caused the cabin air pressure faults. But, a faulty cabin air pressure system may also cause a high pressure condition that caused the door to fail. We have not heard any statements on whether the pressure was (or was being indicated as) high or low.

 

[3DDave]

There would not be air movement - it is covered by an insulated panel. Nor would it whistle. Having been seated next to an poorly sealed emergency door the main effect is that the external jet engine noise enters the cabin but, again, this involved a cover panel with insulation.

I guess anyone who ever notices their brakes aren't working should just operate the car as normal and blame the maker when there is a crash?

There should be clear evidence of what caused the pressure system fail indication. It doesn't matter if it's a Zebra. If the problem is that the plane is pressurizing too slowly that means there is a leak. Finding a leak is a common action. If the airline doesn't want to, contact Boeing and have them find it. At worst they do a smoke test to see where smoke exits.

I don't feel like, if I am plummeting through the air to my death, that the last thought would be "I hope the airline can shift blame to the maker for not dealing with a problem they knew about."

 

[LittleInch]

"If..."

If that was happening then switching to a different control package would make no difference... And the fault would persist.

We're all making wild assumptions about something we know very little about. There are many causes behind a potential pressurisation fail - it may be lack of pressurising air from the engines, it might be more air going out the vents than is supposed to be, it might be a faulty transmitter, it might be a faulty control circuit, it might...

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Alistair_Heaton]

737 don't have ECAM or a health monitoring unit.

There are independent pressurization protections for both ends of the envelope that don't involve human or computer input.

And waross Airbus isn't an option for a lot of airlines mainly because the order book is so full and delivery times so long.

 

[stevenal]

But, a faulty cabin air pressure system may also cause a high pressure condition that caused the door to fail.

The pressurized vessels I work with have pressure relief devices intended to prevent failure due to overpressure. Are they not present on aircraft?

 

[3DDave]

There are pressure reliefs in both directions, although for ground testing they can be locked. Forgetting that has had problematic results. Anti-vacuum valves to stop the fuselage becoming a crushed can and over-pressure reliefs to stop it bursting.

I'd rather fly an airline that had sufficient diagnostic ability to return full confidence than one that depends on drop-down masks to keep me safe from a detected problem that just hasn't gotten that bad yet.

Maybe switching to a different system isn't what made the difference. Maybe it just gave time for the pressure to push the seal back into place; maybe the warm interior air softened the seal some tiny amount while the pilots were playing wack-a-mole. The coincidence seems too on the nose.

 

[dik]

Alaska could have been a little more cautious, not knowing why the 'seal' failed or at least an error was registered. Because of the latent problem, Alaska would not likely be able to determine the source of the problem. In addition it is a little bit of a reach to anticipate the consequences of the problem caused by Boeing or Sprint. My $0.03CAN...

-----*****-----
So strange to see the singularity approaching while the entire planet is rapidly turning into a hellscape. -John Coates

-Dik

 

[1503-44]

Wiki only discusses ETOPS in regard to limits for engine failure on a twin. I do not see where it is applicable to any other potential failure. Any reference showing that it is applicable to decompression? Distance limits are rediculous for decompression warnings. It must be an altitude limitation, or it makes absolutely no engineering sense. I find it difficult to believe that flying at 10,000 ft for 3 hours to land with a door blown out was ever contemplated as an approved emergency procedure. Are there any approved time to airport limitations under which an AC can fly in commercial service with a blown door/Hull failure. I'll eat my socks if its something other than "Descend to 10,000 ft and land at the nearest airport". Even if there were, would anyone be dumb enough to do it? This plane was apparently climbing to an altitude of at least 21,000 ft, but aborted that plan when it decompressed at 16,000. I do not know if an altitude higher than 21,000 was planned as its ultimate cruise altitude. But in any case, if an op limitation were to be made, it should have been "Nothing higher than 10,000ft".

If Alaska Air did not know that 3 warnings, 2 in the previous 2 days on a Max is not potentially very serious, well I guess they do now, and this is not going to be the end of that lesson. Anyway you look at it, unless you disagree with my calculation, 99.9994% is a sure thing something is WRONG with PRESSURE CONTROL. Not the engines. My opinion is that its just plain stupid to continue operating under any circumstances, made even worse by not understanding that a distance limitation was incorrect in the face of pressure problems. It's BASIC physics. If it's true that 3 warnings on one sensor group is not enough of a reason to "stop work", then tell me how many are enough for you. Where do you draw the line. Sure, Intermittent problems are difficult to fix, but that's why they get paid more than auto mechanics. ITS THEIR JOB! If they continue flying until they find the problem, (esp. After 3 warnings) what does that mean? It means they were testing with live guinea pigs. We clear the freeking area for a hydrostatic vessel test. We do not put 175 people on top of the thing when we do it. How difficult is it to do a soap bubble test around door and window frames?

IMO, Alaska owns this just as much as Boeing does. They know it too. They just have their spin doctors out trying to clean it up as much as possible and control the naritive. "Abundance of caution", OMG. They may fool the physics flunkies. I'm not buying it.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[3DDave]

A smoke machine would find the problem of a seal out of place; they used to use tobacco stains for smaller leaks. There are also ultrasound systems that can detect leaks if a smoke generator isn't suitable.

Seeing that all openings are actually secured isn't too big a stretch as would seeing if the doors are actually closed, something that is visible from the ground.

 

[1503-44]

Bucket of soapy water.

--Einstein gave the same test to students every year. When asked why he would do something like that, "Because the answers had changed."

 

[waross]

But, a faulty cabin air pressure system may also cause a high pressure condition that caused the door to fail.

Given the hard evidence at hand, That statement sounds like it could come from a lawyer who studied with tRump's lawyers.

Maybe switching to a different system isn't what made the difference. Maybe it just gave time for the pressure to push the seal back into place; maybe the warm interior air softened the seal some tiny amount while the pilots were playing wack-a-mole.

I suspect that that is exactly what happened.
The seal was imperfect and leaked at low differential pressure.
As the differential increased, the seal was force into place.
But the initial leak caused the alarm.
The switchover to the alternate system gave the extra few seconds for the differential pressure ti rise so that the seal stopped leaking.
I have seen similar effects inflating tires so as to seat the bead.
If the air going in is more than the air going out, then the bead will seat.
If the air going in is not enough volume to overcome the leak, then we go to plan B or plan C.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!

 

[jrs87]

https://www.faa.gov/sites/faa.gov/f...ols/FFP%20Buck%20Stops%20with%20Me%20HTML.zip

Interesting read from the FAA for anyone with spare time. Tales of FFP (Failure to Follow Procedure)

 

[btrueblood]

Agree with 3DDave.

"We're all making wild assumptions about something we know very little about."

Yes, and a lot of "us" (you) are putting blame on Boeing before we know who really dropped the ball. It's as likely that Alaskan service crews pulled the plug out during some refit as it was for Boeing to do so.

To coin an idiot's overused phrase, "we'll have to wait and see".