Ethiopian Airlines Crash; Another 737 Max

I think it’s absurd when a Continental blows to crap we ask Continental to be in on inspecting why it went wrong...
In many if not most investigations I think it's good to have the manufacturer(s) involved, as they have the detailed design, test data, historical operational data, and general expertise in their own systems. It's up to the team and board to make sure conflicts of interest are managed but that doesn't necessarily have to mean excluding the people who know the most about the airframe, engine, or whatever part you want to consider.

Nauga,
who hasn't seen the conflict but doesn't doubt it happens
 
In many if not most investigations I think it's good to have the manufacturer(s) involved, as they have the detailed design, test data, historical operational data, and general expertise in their own systems. It's up to the team and board to make sure conflicts of interest are managed but that doesn't necessarily have to mean excluding the people who know the most about the airframe, engine, or whatever part you want to consider.

Nauga,
who hasn't seen the conflict but doesn't doubt it happens

I don’t disagree on “involved” ... I should rephrase: they shouldn’t be the sole expert, nor in charge of determining their own fault or lack there of... maybe their primacy is overblown in my head...
 
I just read an article that said an AoA disagree warning is available on the MAX, but neither Lion Air or Ethiopian had purchased the optional feature.

Now Boeing is saying it will become a standard feature on future builds.

Again, this seems like a crazy omission on Boeing's part. Making it optional? Jeez.
Revenue Enhancement!
 
This thread is TL:DNR all of it. In software development, a bug is known as an undocumented feature. As originally shipped, MCAS was an undocumented feature, literally!

Boeing shipped a broken airplane with the FAA’s blessing, and airlines world wide bought it. Well trained crews could deal with the ‘feature’ and poorly trained crews were the third link in the accident chain.

Disclaimer: I am not and have ever been any sort of a commercial pilot and am past 65 so I’ll never be one for any international carrier.
 
This thread is TL:DNR all of it. In software development, a bug is known as an undocumented feature. As originally shipped, MCAS was an undocumented feature, literally!

Boeing shipped a broken airplane with the FAA’s blessing, and airlines world wide bought it. Well trained crews could deal with the ‘feature’ and poorly trained crews were the third link in the accident chain.

Disclaimer: I am not and have ever been any sort of a commercial pilot and am past 65 so I’ll never be one for any international carrier.

“TLNR”?

In software development that “undocumented feature” is just a joke. We still really know that it is a true bug, and it gets added to the list of bugs to fix. Occasionally (very rarely) a “bug” happens to actually have some positive aspect, but even then, you then document it and usually rewrite the code so it is a stable ex-bug, now feature.
 
Here is a window on EASA's take.

Presentation to European Parliament by the Executive Director of EASA
3 Sept 2019

https://www.europarl.europa.eu/cmsdata/186500/20190903_EASA_Ky-original.pdf

Next major milestones p 16
  • - Safety assessment of the new design changes proposed by Boeing, including operational procedures
  • - Human factor evaluation and functional tests of the new software
  • - Flight tests on a modified B737 max [one full week - at Boeing Flight Test Center]
-- MCAS operations (nominal behavior)
-- Flight without MCAS (including high speed turns and stall)
-- Scenario of stabiliser runaway (uncommanded MCAS activation, manual trim wheel forces)
-- Approach to stall with autopilot engaged​
  • - Crew Training requirements, in particular using Computer Based Training or Simulator
  • - Coordination with EASA Member States on Return to Service actions
 
Taking these statements together:

Boeing has said that the purpose of MCAS is not to prevent a stall but simply to make sure it handles exactly like the earlier model 737 when going through certain stall testing.

and

.
The report states that the FAA raised concerns to Boeing about the cumulative effect of cockpit system changes from the previous 737 model to the MAX and suggested that might create a need for simulator level pilot training.

Boeing’s response to this concern, which the FAA accepted, was that there was “no precedent” for a demand for enhanced training in previous certifications of derivative models.


In essence, when MCAS is disabled the crew is effectively, and instantly, flying an aircraft in which they have never been trained and which has unfamiliar flight characteristics.

This thing is so screwed up in so many ways I hardly know where to start, but I do wonder how many other similar hazards are lurking elsewhere in the MAX and even in other Boeing designs.
 
In essence, when MCAS is disabled the crew is effectively, and instantly, flying an aircraft in which they have never been trained and which has unfamiliar flight characteristics.
Huh? How do you figure? After the Ethiopian bird strike, US operators of the MAX did a review of their technical data and found that MCAS had exactly zero activations on their flights fleet-wide. None. So the entire time they were operating "an aircraft in which they have never been trained and which has unfamiliar flight characteristics", yet somehow nary an incident? Makes no sense.

MCAS was only designed to engage in abnormal flight regimes, which obviously never happened on a US-flagged flight.
 
Huh? How do you figure? After the Ethiopian bird strike, US operators of the MAX did a review of their technical data and found that MCAS had exactly zero activations on their flights fleet-wide. None. So the entire time they were operating "an aircraft in which they have never been trained and which has unfamiliar flight characteristics", yet somehow nary an incident? Makes no sense.

MCAS was only designed to engage in abnormal flight regimes, which obviously never happened on a US-flagged flight.


You miss the point.

Boeing stated that the entire purpose of MCAS is to make the MAX handle like previous generation aircraft during a stall. Switch off the MCAS and now it DOESN’T handle like previous gens during a stall. In an emergency, disabling a failed MCAS effectively puts the pilot into a “new” plane that he hasn’t trained in.
 
Huh? How do you figure? After the Ethiopian bird strike, US operators of the MAX did a review of their technical data and found that MCAS had exactly zero activations on their flights fleet-wide. None. So the entire time they were operating "an aircraft in which they have never been trained and which has unfamiliar flight characteristics", yet somehow nary an incident? Makes no sense.

MCAS was only designed to engage in abnormal flight regimes, which obviously never happened on a US-flagged flight.
Do you see no need for simulator training that covered how to handle the type of stall vane failure that occurred outside the U.S.?
 
Do you see no need for simulator training that covered how to handle the type of stall vane failure that occurred outside the U.S.?
I think the reasonable answer is yes, but on the flight prior to the accident flight in the Lion Air plane, a deadheading British pilot showed the crew how to resolve the MCAS issue they had without knowing that MCAS even existed, either. The other remarkable thing about that flight is that after gaining control of the aircraft again, they flew the remaining two hours to their destination with the stick shaker going off the whole time. That tells me all I need to know about the safety culture of Lion Air.
 
You miss the point.

Boeing stated that the entire purpose of MCAS is to make the MAX handle like previous generation aircraft during a stall. Switch off the MCAS and now it DOESN’T handle like previous gens during a stall. In an emergency, disabling a failed MCAS effectively puts the pilot into a “new” plane that he hasn’t trained in.
Actually it was designed to mimic the handling of the NG nearing a stall, not in it. And the alternative is just not to stall it. As I said, it kicks in in abnormal flight regimes. Like my instructor told me during private, "what the hell would I want to teach you how to stall an airplane for? I want to teach you how not to stall an airplane."
 
Like my instructor told me during private, "what the hell would I want to teach you how to stall an airplane for? I want to teach you how not to stall an airplane."


So I guess you never learned how to recover from a stall? Or unusual attitudes? Must have been an interesting checkride.
 
But the crews wouldn’t know that...,
Except that the jumpseat pilot on the Lion flight the previous day knew what to do and killed the stab power switches...and the PIC of the accident Lion flight kept the aircraft flying through 21 MCAS activations...but if you think it's Boeing fault that Ethiopia Air put a 361 hour SIC in a MAX...well I guess there's nothing more to discuss.
 
Stalling an airplane should be a routine event, if you are a properly trained pilot. I have lost count of the intentional stalls that I have done, but the number of UN INTENTIONAL stalls is zero. I have entered an intentional stall at 7,500 feet, and held wings level to 3,000 feet before recovering, went back up, and repeated with full flaps, to prove to a partner in the plane that it would not automatically spin if stalled.

It is more than curious that none of the Max's flying in the United States had a problem, although they flew hundreds of times as many hours as the total of Lions and Ethiopians pilots. If you land a defective plane and don't fix it, Lion, the next pilot is flying a defective plane, not from the factory, but from local damage. If you do not tell your pilots about the Boeing procedure for failure of TCAS procedure, how do you expect them to react correctly? Ethiopian Air did not have that letter in their manuals, and had no documents presented to their pilots for this problem.

Poor training and failure to maintain aircraft will produce crashes with the best planes available. Lion Air changed from Airbus to Boeing because their pilots were crashing too many Airbuses, they felt that the problem was bad design of the Airbus.
 
Stalling an airplane should be a routine event, if you are a properly trained pilot. I have lost count of the intentional stalls that I have done, but the number of UN INTENTIONAL stalls is zero. I have entered an intentional stall at 7,500 feet, and held wings level to 3,000 feet before recovering, went back up, and repeated with full flaps, to prove to a partner in the plane that it would not automatically spin if stalled.

It is more than curious that none of the Max's flying in the United States had a problem, although they flew hundreds of times as many hours as the total of Lions and Ethiopians pilots. If you land a defective plane and don't fix it, Lion, the next pilot is flying a defective plane, not from the factory, but from local damage. If you do not tell your pilots about the Boeing procedure for failure of TCAS procedure, how do you expect them to react correctly? Ethiopian Air did not have that letter in their manuals, and had no documents presented to their pilots for this problem.

Poor training and failure to maintain aircraft will produce crashes with the best planes available. Lion Air changed from Airbus to Boeing because their pilots were crashing too many Airbuses, they felt that the problem was bad design of the Airbus.

If I recall correctly, wasnt at least one of the two due to a faulty sensor? In which case it is not that strange that it could have flown hundreds of hours in the US without sensor failure?

I work on a complex computer system that has been in production since 1990, running essential 24/7. About three years ago a problem came up, and when I was troubleshooting it, I discovered some very badly written code, a bug, where there was a very obvious oversight in not checking a value that any “idiot” of a programmer would know SHOULD be checked and throw an error if not correct. It amazed me no end that this problem did not show up in all these years, although I suspect it might have hit a few times earlier, but been attributed to something else, or that it “hit” during a less loaded period and didn’t cause as many problems as it did when it cause real problems a few years back.
 
If I recall correctly, wasnt at least one of the two due to a faulty sensor? In which case it is not that strange that it could have flown hundreds of hours in the US without sensor failure?

I work on a complex computer system that has been in production since 1990, running essential 24/7. About three years ago a problem came up, and when I was troubleshooting it, I discovered some very badly written code, a bug, where there was a very obvious oversight in not checking a value that any “idiot” of a programmer would know SHOULD be checked and throw an error if not correct. It amazed me no end that this problem did not show up in all these years, although I suspect it might have hit a few times earlier, but been attributed to something else, or that it “hit” during a less loaded period and didn’t cause as many problems as it did when it cause real problems a few years back.
It was a sensor failure, which Lion Air did not perform the maintenance required by Boeing and then lied about it not only to their own crew, but also lied about and falsified evidence in the accident investigation. So yes, but...
 
Grilling Boeing CEO now. You can listen in that at CNBC.
 
Back
Top