Hacker News new | past | comments | ask | show | jobs | submit login
[dupe] Boeing Pilots Detected 737 Max Flight Control Glitch 2 Years Before Deadly Crash (npr.org)
79 points by elorant 23 days ago | hide | past | web | favorite | 22 comments

Title is incredibly misleading.

The text messages exchanged are about the unexpected (but normal for 737 max) amounts of speed trim applied. I.e. during normal operation, MCAS was applying more trim than a non-max 737 pilot would expect. The individuals exchanging these text messages never encountered the MCAS anomaly that led to the two crashes. The pilots explicitly did not encounter the 'glitch'. The simulator that the pilots were using was not even capable of replicating the glitch, as it didn't factor trim screw forces.

The text messages are evidence of a systemic lack of communication in the testing and certification process, but they are not the 'pilots knew MCAS would cause crashes' smoking gun that the mainstream media is about to start misinterpreting them as.

Here's Juan Browne breaking it down. For those not in the know, he ranks amongst the most experienced pilots in the country.


Is it not indicative, though, of the progression of changes to MCAS over time, different speed envelopes, significantly more elevator control, etc. And no process to decide when those changes should warrant additional documentation or training requirements?

Meaning, unrelated to "the glitch", but could have exposed info or training that would have made "the glitch" more survivable, because the pilots might have known MCAS was in play.

> Is it not indicative, though, of the progression of changes to MCAS over time, different speed envelopes, significantly more elevator control, etc. And no process to decide when those changes should warrant additional documentation or training requirements?

Yes, that's exactly what it is. And that in itself is a big deal.

But it is not the 'omg pilots knew about MCAS causing crashes and didn't/couldn't report it' that the media coverage is spinning it as. The title very misleading.

I've had people call me a 'boeing shill' (what would such a person even have to gain??) for pointing out these sorts of issues in the coverage back when the media was saying aoa-disagree would have prevented this from happening (it wouldn't, most commercial pilots have no native idea what aoa is, only airlines with ex-military pilots would get benefit from that package). The vast majority of people just read the headlines, assume them to be true, and move on.

I've had people call me a 'boeing shill' (what would such a person even have to gain??)

I'm not agreeing with them, but if they're calling you a shill they're accusing you of being paid by Boeing.

There was already a process to decide if MCAS should be in the manual.

That was exactly the job of these two technical pilots who both signed off saying the MCAS was “benign” and didn’t need to be included in the manual.

These guys had direct communication with the FAA who clearly had the final word on the manual. The FAA trusted the technical pilots word (and I doubt it was just these two guys alone that decided).

The question is whether these simulator test pilots were under any pressure from Boeing to downplay it but based on their lawyers statements they seem to be defending themselves and saying the conversation was simply being misinterpreted and they didn’t have concerns. The other question is whether the FAA was too quick to trust the information coming to them.

Things are not so simple as the pilots knowing everything and just hiding the truth. They still identified there was an issue, but couldn’t match it to MCAS as they didn’t know about it, and misattributed the effects to a simulator error.

> As the two pilots banter back and forth in the messages, Forkner says the system is "running rampant in the sim on me," then adding, "I'm leveling off at like 4000 ft, 230 knots and the plane is trimming itself like crazy."

For me this clears the bar of pilots detecting there is an issue.

The pilots knew about MCAS. What the texts indicate is the behaviour of MCAS was changed during testing without the pilots being informed, and the pilots also failing to report that change despite becoming aware of it. This was certainly a contributing factor to the events that occurred later, but not at all the smoking gun that the title implies.


"Oh shocker alerT!

MCAS is now active down to M .2 [mach 0.2]

It's running rampant in the sim on me

at least that's what Vince thinks is happening"

It is not clear from the transcript whether it was a misattribution when they attributed the problems they were discussing to the simulator. It also seems clear that they were not getting flying time on the actual airplane, and so were not in a position to know if the airplane itself had problems.

There are definitely issues to be investigated here, and it is not going to be helped if people are taking the fifth, as Forkner himself has done. He might want to reconsider that.

The issue that resulted in the crashes was angle of attack sensors reporting broken data, which caused MCAS to nose down the plane; this is absolutely not what the pilots are talking about.

Everything else that has been discussed (whether the pilots should have been told how to disable it and tried it in the simulator, whether Boeing should have added the second vane or even a third one on all planes, etc.) fundamentally depends on having incorrect data from the sensors and the plane not being able to deal with that.

The exchange might hint that MCAS had an awful design history, but even a perfect MCAS could have had issues if supplied with incorrect sensor data; garbage in, garbage out. Now if there was somewhere an email about the lack of a voting system on vane readings, that would be another story.

The issue is the undocumented (and obfuscated) behavior of the airplane after the failure of a single AofA sensor.

Right, and it has nothing to do with the issue reported by the test pilots in the simulator.

I see what you mean - there is no evidence (at least in the transcript) that the problems were in response to a simulated system failure, and the actual MCAS seemed to work well enough when nothing was broken.

I truly don't know what runs through the head of a writer or editor who hits publish on a story with this title right now.

Surely they must know that they're actively misleading people with it. That putting it out into a world where ten times as many people read headlines as entire articles means they're lying to ten times as many people as they're informing.

This has been happening from the start of the 737 MAX situation.

It has been incredibly eye-opening to see how the media will amplify and outright misrepresent the facts of a situation to create outrage/clicks/uptake/whatever. I know a little bit about aviation and chat to commercial pilots so watching the media coverage and comparing it to what the people in the industry are saying has been a bit bewildering and eye opening. I can only assume the media is equally manipulative in all other subject matters where I don't know enough to discern.

Right from the start, except, right from the start media reports have consistently understated Boeing’s involvement.

Thats what allowed the FAA to prevent grounding the MAXs until days after the 2nd crash and only after nearly every other regulator in the world had grounded it.

The fact that anyone can complain that this situation has been presented in a way that is unfair to Boeing is remarkable.

Thanks for the clarification. The article also mentions that the same pilot who sent some of the text messages voucher for the safety of the plane to FAA regulators. It looks like whatever concerns he had, he chalked up to issues in the simulator itself.

Indeed if this was the case this story would be a far bigger deal.

This is now marked as a dupe, so not likely to garner much more traffic, but I am curious what the “M.2” refers to in the transcript?

Also, it’s pretty clear to me that they are confused by the parameters of when MCAS should be active and how much trim it should be applying. They are trying to figure out if the sim code is broken or out of sync with the spec. They are asking for spreadsheets identify exactly which flight envelopes it should be activating under.

This transcript perhaps demonstrates a cultural problem at Boeing around a safety critical system. What it doesn’t actually show is that Boeing knew about the specific MCAS issue that contributed to crashing two planes.

A question in my mind is what happened between this exchange and the next time this pilot went to the FAA to tell them to keep MCAS out of the docs.

I really hope they can trace this all the way to the top and not simply hang this test pilot out to dry.

M.2 is mach 0.2 - the speed range in which MCAS was active had been expanded downwards.

Your second paragraph is precisely to the point, and it is not clear to me whether these two pilots were ever given enough correct information to make an informed decision about the safety of MCAS.

Looks like https://en.wikipedia.org/wiki/Whistleblower protection assurance is not efficient; Time to amend it;

There are no whistleblowers in this story... both pilots signed off on the safety and the guys lawyer is arguing the conversation is being misinterpreted.

> “If you read the whole chat, it is obvious that there was no ‘lie,’” David Gerger, a lawyer for Forkner, said in an email. “The simulator was not reading right and had to be fixed to fly like the real plane. Mark’s career -- at Air Force, at FAA, and at Boeing -- was about safety. And based on everything he knew, he absolutely thought this plane was safe.”

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