Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

'But the FAA said that other civil aviation authorities had not "provided data to us that would warrant action".'

Pretty sure the data is that two planes crashed in five months in perfect weather conditions, killing everyone on board. Statistically, that's sufficiently improbable that it warrants investigation.



> Pretty sure the data is that two planes crashed in five months in perfect weather conditions

Not only that, but the data we have so far from the second crash (ADS-B data showing irregular vertical speed, see https://www.flightradar24.com/blog/flightradar24-data-regard...) is consistent with both accidents having a similar cause. Grounding all airplanes from the same model until preliminary data from the (already recovered) flight recorders can confirm or discard that hypothesis is prudent.


If it was a "similar cause", it is absolutely pilot error.

The MCAS system, at fault in the first crash, may be disabled via a cockpit control. In other 737s, moving the stick was sufficient to disengage it. For some reason, this was changed in the Max, and you must use the other control. However, any competent pilot would know every control in the cockpit. The captain of the latest flight to crash allegedly had 8,000+ hours of flight time.

If an aircraft begins behaving incorrectly, the first thing a flight crew should do is disable automation and try and fly it manually. That wasn't done if MCAS caused the crash. This was an emergency that developed over minutes, not seconds, and there's really no excuse.


There's a good hint at what's going on in a thread from yesterday about the 737MAX reports [1]:

> 1555013 is pretty damning in telling how the airlines treated the MAX as the same type as the 737-800:

> "I had my first flight on the Max [to] ZZZ1. We found out we were scheduled to fly the aircraft on the way to the airport in the limo. We had a little time [to] review the essentials in the car. Otherwise we would have walked onto the plane cold.My post flight evaluation is that we lacked the knowledge to operate the aircraft in all weather and aircraft states safely."

To provide a parallel to the software world, it kind of sounds like they released a "version 2.0" but called it "version 1.1", giving the sense this was a minor revision and anyone that already knows 1.0 will be fine.

[1] https://news.ycombinator.com/item?id=19374386


> If it was a "similar cause", it is absolutely pilot error.

That's a very blunt argument to make. Do watch the below video - I'm not educated as much on these matters, but the pilot here makes a compelling point as to why it could be from the Lion crash: https://www.youtube.com/watch?v=zfQW0upkVus

He makes reference to the system for disabling not being included in the type training for the MAX, and given there wasn't an update on the fix for the issue (read that this is partly due to the US government shutdown meaning Boeing and the FAA didn't have as direct comms channels), it could still have been the same issue without the blame sitting on the pilots.


Unless I am mistaken, the regular 737s do not have the MCAS installed,just the MAXs (https://theaircurrent.com/aviation-safety/what-is-the-boeing...)

Additionally it has been suggested that documentation for this physical switch is not well laid out or presented in the manual because Boeing was confident that it would simply stay out of the pilot's way during flight.

Assuming pilot error would seem to be premature if these points are in fact true.


Either way "pilot error" is defined with reference to some "reasonable level" of competence with the aircraft.

If multiple pilots are making the same error the issue is with the aircraft changes, not the "reasonable competence".

Ie., its redesign hasn't been thought through properly


As far as I'm aware, the previous versions gave an audible warning when the flight parameters were exceeding limits. The new system attempts to actively take control and sink the nose. If you know the system from previous aircraft, you'd expect the same behavior in the new one. I'm certain they mention this in training, but if English is not your first language, you'd have to pay particular attention to this part or you're screwed.


There was no new training for this is what I've read. Boeing thought the new system was simple and similar enough that no retraining was required for the new model.


>However, any competent pilot would know every control in the cockpit.

Not if they tell you the plane acts the same, and then it doesn't! Did Boeing inform pilots of this change proactively? Was the pilot trained on the change, and perhaps even recertified?


Depending on the situation, disabling autopilot is not the correct thing to do. That was identified as a contributing factor in TransAsia 235 https://www.asc.gov.tw/upload/cont_att/9b051632-ebf6-427f-b0...

Through simulations they think a crash would have actually been avoided had autopilot been left on.


> However, any competent pilot would know every control in the cockpit.

Except it seems like many of them didn't receive training on what changed for the new model.


Most would even say statistically significant enough to order the planes grounded in their own countries...


> Statistically, that's sufficiently improbable

Is it?


Yes, it is. The 737 NG (700 series, 800 series, etc) has a hull loss rate of 0.27 per million departures[0]. I am having difficulty finding the exact number of departures for the MAX but 737s average about 3-5 departures per day[1]. Only 350 MAXs have been delivered and the first one was delivered on May 6, 2017[2] (676 days ago). So even being generous and supposing that all 350 aircraft were delivered exactly on May 6, 2017 and that each plane made 5 flights daily, that would be a total of 1,183,000 flights. Therefore the MAX line has a best-case hull loss rate of 1.69 per million departures. I.e. the MAX generation has a hull loss rate 6.25 times that of the NG line.

[0]: https://en.wikipedia.org/wiki/Boeing_737#Accidents_and_incid...

[1]: https://www.researchgate.net/publication/274782362_Character...

[2]: https://en.wikipedia.org/wiki/Boeing_737_MAX


the question is if this difference is statistically significant


You are taking a commercial flight. You have the choice between 2 different aircraft which are essentially identical but you are more than 50 times more likely to crash in one of them. Which do you take?

EDIT: I have learned that I was far too generous with the flight numbers. The MAX series has only logged ~150K flights. Therefore it's hull loss rate is roughly 13.33 per million making it almost 50 times more dangerous than the NG line.

Yes, the difference is statistically significant.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: