Due to time constraints, I’m going to deal with the current state of Boeing’s woes in broader strokes than usual, and I hope readers will chime in with details, calibration, and any quibbles or corrections.
The latest stories are not pretty:
Boeing new sales have collapsed. However, no one has cancelled orders but new orders have stalled. From CNN:
Boeing had no new orders for planes in April.
Not only did the troubled 737 Max receive zero new orders since it was grounded March 13. Boeing’s other jets, such as the 787 Dreamliner or the 777, also did not get any new orders last month, according to a company report released Tuesday….
None of Boeing’s other jet models have crashed, and airlines have not reported any safety problems other than the 737 Max. But the 737 Max’s problems could be the reason airlines put orders for the other jets on hold, said Philip Baggaley, the lead credit analyst for the transportation sector for Standard & Poor’s.
Airlines are holding out to determine whether Boeing will lower its prices in light of the Max issues, Baggaley believes.
“If it turns out that Boeing makes some accommodations with customers over the grounding of the 737 Max, it probably won’t be a straight cash compensation,” said Bagggaley. “It might be a lower price on future orders, or some other change in those orders. There could be things going on behind the scenes that could essentially turn into orders.”
Pilots pressed Boeing unsuccessfully for safety fixes right after the Lion Air crash and were rebuffed. The New York Times and CBS News obtained a copy of a recording of a November 22 meeting between representatives of the American Airlines pilots’ union and Boeing. From CBS:
Frustration boiled over during the tense meeting in November 2018, less than a month after the first Max crashed, and four months before the second crash.
“We flat out deserve to know what is on our airplanes,” one pilot is heard saying.
“I don’t disagree,” the official said…
The pilots at the meeting were angry that system was not disclosed to them until after the first crash.
“These guys didn’t even know the damn system was on the airplane — nor did anybody else,” one pilot said.
Boeing vice president Mike Sinnett, who does not appear to know he was being recorded, claimed what happened to Lion Air was once-in-a-lifetime type scenario….
The pilots in the room were not satisfied with that answer. “We’re the last line of defense to being in that smoking hole. And we need the knowledge,” one pilot said.
FAA didn’t question Boeing misrepresentation of not identifying MCAS as capable of producing a catastrophic failure. We’ll say more about this issue later in the post. The Wall Street Journal may be correctly anticipating the spin that Boeing may be able to pass off, that the FAA failed to have senior officials review MCAS, as opposed to (as the article clearly states, but further down) that Boeing didn’t identify MCAS as a critical system. Remember, as we described in an earlier post, how the FAA had changed in 2004 changed its review system . Before, the agency representatives at manufacturers were manufacturer employees but reported to the FAA. After 2004, the FAA designees reported to the aircraft makers, and for the most part, only more senior managers dealt with the regulator. This was a system rife for abuse, yet it appears that the safety culture in the industry is so strong that only Boeing abused it.
The [internal investigation] results, these officials said, also indicate that during the FAA certification process for the 737 MAX, Boeing didn’t flag the automated stall-prevention feature as a system whose malfunction or failure could cause a catastrophic event. Such a designation would have led to more intense scrutiny.
FAA engineers and midlevel managers deferred to Boeing’s early safety classification, the inquiry determined, allowing company experts to conduct subsequent analyses of potential hazards with limited agency oversight. Boeing employees who served as designated agency representatives signed off on the final design…
Was MCAS sure to crash planes when triggered by false readings? A self-proclaimed “737 Pilot” in comments as well as a known commercial pilot, Jeremy Grimm, have both contended that the Lion Air and Ethiopian Air crashes were due to the pilots being insufficiently experienced. “737 Pilot” argued that if the pilots had followed standard procedures, they would have used “trim stab” and been able to regain control of the plane. An Australian 60 Minutes show, prominently features an actual 737 pilot in flight simulator who contends the reverse. Moreover, if I interpret it correctly, a new story in the Seattle Times further calls into question whether the pilots could have regained control.
Australia’s 60 Minutes documentary did a better job of identifying and presenting a key MCAS flaw than the rest of the press. See the segments starting at the very beginning and at 21:35, of a 737 pilot replicating the Lion Air and Ethiopian Air crashes. Please do watch these two short segments. There’s no substitute:
The critical part, which I haven’t seen in any of the 737 Max stories I have read (but Lambert says has been in some of the ones he has seen, but the implications appear not to have been understood by the reporters) is that MCAS would turn on for ten seconds to push the nose of the plane down, which in the Lion Air crashes, put the nose at 40 degrees, and do this repeatedly. This is the key bit:
The Ethiopian pilots would have been trying to follow Boeing’s checklist. But what the pilots didn’t know was that the MCAS system will repeatedly turn on and off. It means they have only five seconds of control before MCAS overrides the plane, forcing it to nosedive.
So work that out. MCAS goes on for ten seconds. The pilots have only five seconds to try to pull the plane back up. Then MCAS kicks in for ten seconds again, shoving the nose into a steep dive again. This is guaranteed to eventually run the plane into the ground (or ocean). It is impossible in five seconds for the pilots to recover the loss of elevation that MCAS creates in ten second of initiating and keeping the plane in a dive.
Moreover, the press reported that the black box of the Ethiopian Air jet showed the pilots consulted the MCA manual, followed its instructions, yet even though they succeed in turning off MCAS, they were also unable to make the trim controls respond. Reader “737 Pilot” argues that all the pilots needed to do was rely on their training or a Quick Reference Card on the glasshield. The 737 pilot who simulated the crashes disagreed, showing where the relevant instructions were (in a paper manual).
The critics of the Ethiopian Air pilots argued that the plane was flying too fast; that’s why they couldn’t make the trim controls behave when they did turn off MCAS. The Seattle Times article is not crisply written but it indicates that Boeing had made yet another change to the 737 Max that it kept to itself: that it changed how a key “auto-diable” switch worked. If the Seattle Times has interpreted the manuals it received correctly, turning off MCAS didn’t just turn off the auto trim systems (for the aircraft tail); it also turned off the normal manual control for the trim, a thumb button. If this reading is correct, no wonder the Ethiopian Air pilots couldn’t control the trim. Their normal way for doing so had also been turned off.
In the middle of Boeing 737 cockpits, sitting between the pilot seats, are two toggle switches that can immediately shut off power to the systems that control the angle of the plane’s horizontal tail.
Those switches are critical in the event a malfunction causes movements that the pilots don’t want. And Boeing sees the toggles as a vital backstop to a new safety system on the 737 MAX – the Maneuvering Characteristics Augmentation System (MCAS) – which is suspected of repeatedly moving the horizontal tails on the Lion Air and Ethiopian Airlines flights that crashed and killed a total of 346 people.
But as Boeing was transitioning from its 737 NG model to the 737 MAX, the company altered the labeling and the purpose of those two switches. The functionality of the switches became more restrictive on the MAX than on previous models, closing out an option that could conceivably have helped the pilots in the Ethiopian Airlines flight regain control.
Boeing declined to detail the specific functionality of the two switches. But after obtaining and reviewing flight manual documents, The Seattle Times found that the left switch on the 737 NG model is capable of deactivating the buttons on the yoke that pilots regularly press with their thumb to control the horizontal stabilizer. The right switch on the 737 NG was labeled “AUTO PILOT” and is capable of deactivating just the automated controls of the stabilizer.
On the newer 737 MAX, according to documents reviewed by The Times, those two switches were changed to perform the same function – flipping either one of them would turn off all electric controls of the stabilizer. That means there is no longer an option to turn off automated functions – such as MCAS – without also turning off the thumb buttons the pilots would normally use to control the stabilizer.
Peter Lemme, a former Boeing flight-controls engineer who has been closely scrutinizing the MAX design and first raised questions about the switches on his blog, said he doesn’t understand why Boeing abandoned the old setup. He said if the company had maintained the switch design from the 737 NG, Boeing could have instructed pilots after the Lion Air crash last year to simply flip the “AUTO PILOT” switch to deactivate MCAS and continue flying with the normal trim buttons on the control wheel. He said that would have saved the Ethiopian Airlines plane and the 157 people on board.
Let’s take what we know already and what we think we know from the 60 Minutes and the Seattle Times story above:
Boeing used only one sensor rather than two to provide input into what was designed as a critical safety feature. That alone is a departure from long-standing industry practice
MCAS, rather than making a minimal or mild change, which would seem more appropriate given that the angle of attack sensor input was at much greater risk than usual of being wrong due to having only one reading, instead made an extremely aggressive correction
MCAS was insistent and allowed pilots only a short window in which to attempt to regain control before it would kick in again. The duration of the “on” versus “off” time would assure a crash unless the pilots could figure out how to disable MCAS and regain control of flight altitude
Boeing changed how the 737 Max switch that turned off automated functions worked. Not only would it kill MCAS, but it would also disable the thumb buttons that pilots used to control the “stabilizer” which they’d need to engage to pull out of an MCAS-induced dive
Boeing hid the above information from airlines, pilots, and regulators, and then tried to downplay its importance when it came out in bits and pieces
If these are all correct, Boeing should burn to the ground. Next best would be for its CEO Dennis Muilenburg to get the Jeff Skilling treatment.
But Boeing is a national champion, part of a duopoly, and otherwise too big to fail. However, if the outline above is more or less accurate, foreign regulators may feel they have no choice but to refuse to certify the 737 Max unless more than just the software is fixed, and that could keep the plane grounded for a very long time. I am not holding my breath that that will happen, but they appear to be the only channel for forcing real change on Boeing. Airlines face too many costs in “divorcing” Boeing, and there are very few passengers who even if they have resolved not to fly on a Max, could afford time and money-wise to refuse to fly if one were swapped in at the last minute. However, if you are really worried, you can fly Delta or other airlines that haven’t bought the Max.
The only upside to this sordid affair is that it shows why deregulation and self regulation are fatally bad ideas. But what a horrific way to do that.
Many purchases are financed (leased) and the aircraft-leasing companies might be concerned about the ability of these assets to generate cashflow to pay for the lease (maybe there are more issues forcing groundings?) and maybe there are some concerns about the value of the underlying security. No financing -> no purchase/lease. It could be a factor, it is only a guess so take it for what it is worth.
It looks like Boeing will soon have plenty of time to sort out their quality control issues.
Not to forget, as Lambert has signalled, the McDonnel Douglas “culture” contaminated Boeing and that should also have to be fixed somehow.
The McDonnell Douglas “reverse takeover” story is well known in the industry; how the MBAs took away the controls from the engineers. There’s a lot of growsing about it, even today.
I think the key airline to look for is Ryanair. They are the one of the biggest single buyer of 737’s (they operate around 400 737-800’s), that’s the only aircraft they’ve used for much of their history. They are also notorious for using circumstances to drive a very hard deal with Boeing (they famously ordered dozens in the weeks after 9/11, allegedly at a loss to Boeing for each one). They’ve been very quiet about this, I suspect they are looking at their options. It would be an enormous blow to Boeing if they switched, as many budget airlines follow their lead.
A particular consideration for Ryanair is resale values – they churn aircraft at a high rate – usually selling them on to Asian budget airlines. So they may be worried at the long term implications if the entire 737 line ends up tainted.
As Jesper suggests, another key market are aircraft leasing companies. They have their own particularly complex set of calculations to make – I doubt they’ll go buying until they are absolutely sure the 737 will be certified fully worldwide.
The one thing that can save Boeing is that Airbus doesn’t have capacity to deliver more A320’s and A220’s quickly. And the Sukhoi Superjet 1000 is now equally tainted after the crash last week. I doubt Tupolev, Comac or the Mitsubishi or Embraer can step up quickly either, although they’ll all hope to benefit.
Unfortunately the public is largely unaware of the criminality of Boeing and the risks of the 737 Max. A Reuters poll today:
Fears of flying Boeing’s 737 MAX won’t get in way of price conscious ticket shoppers: Reuters/IPSOS poll
Few outlets are covering this story in any depth.
I’m about as price conscious as a flyer gets, and I certainly don’t plan to fly Boeing. Any Boeing. Ever again. Funerals are expensive.
The business press has been covering this story heavily, so the business traveler is well aware.
CNBC quoting says the reverse, and they poll business passengers. Headline: Boeing downgraded by Barclays on survey showing flyers will avoid 737 Max
The survey was of over 1700 people, so it’s a decent sample.
And anyone who has done marketing will tell you a survey of the sort Reuters did is crap. Asking people about what they will pay for is notoriously unreliable. Respondents lie about how they behave. The way the Reuters poll questions are phrased is both leading and generally guaranteed to be unreliable.
The reason to worry is that business customers aren’t paying for their flights, so they can afford to exercise safety preferences. And business travelers provide roughly 75% of airline revenues:
The confidence interval for n=1700 and a large population (air travelers) is 2.33. So the 52% figure could represent either 50% or 54% of the air travelers population. That is at a 95% confidence level. Most common.
Not good for Boeing.
I fly a fair amount and usually with upgrades. Anecdotally most people I meet travelling are ignorant of the 737 Max facts reported here or in the Seattle Times. I read the FT every day and I feel it lacks the frequency and depth of NC or the Seattle Times. Of course once you get to cable “news” there is little depth and nothing like the Australian 60 minutes piece.
I was unaware of the CNBC piece. That is a big deal, as is a downgrade.
I agree the Reuters poll using the internet is problematic, but I do not classify it as meaningless.
That “auto-diable” switch sounds fiendishly apt.
RE: However, if you are really worried, you can fly Delta or other airlines that haven’t bought the Max.
Delta anecdote: I have flown Delta four times in the past two years. Each time have been impressed with the airline’s on-time performance, baggage handling, ground staff, and cabin staff. I now recommend Delta to friends.
Delta in the throes of a bitter anti-union campaign:
The sprocketheads have been trying to get voted in since forever. They still have hurt feelings from being voted out at the last merger. Luckily they still represent many defense industry workers.
Delta Air Lines still has three unions on the property. Everyone else acts like they have union representation but they don’t have to pay dues.
if the destination is under 1000 miles, i’m driving. and i’m not getting on one of those deathtraps in any case.
Terra firma, baby, and the firma the terra the betta!
i’m waiting for the story about the software update that crashes a plane in mid update. underpaid pilots (i read some are paid $15 an hr) being overworked and undertrained do not induce me to leave our fair green earth.
It would seem our friend “737 Pilot” is more likely to be “Boeing Damage Control“, I would guess we’ve seen the last of that moniker.
I hope everyone noticed the amount of empathy evidenced by the pilots in the video for the crew and passengers lost to Boeing’s hubris.
That, in contrast to the callous, and now obviously dishonest declaration that all the doomed pilots had to do was rely on “training” that never happened, or follow the “Quick Reference Card“.
Cold-hearted condemnation of men who died after spending endless, horrific minutes fighting for their lives, and the lives of their passengers.
I suspect you are correct.
I suspect he is not.
“He would say that, wouldn’t he?”
You are saying he is not a pilot or you are just making stuff up? And what he did say is that pilots in this time of expanding air travel and pilot shortages are undertrained and too reliant on automation. Do you disagree or in fact know anything about this at all because this has been claimed many other places? Small commuter planes are now flown by very low paid aspiring pilots looking to eventually move up to the parent airlines. The old days when pilots were highly experienced ex military flyers are largely over. This undoubtedly had something to do with Boeing including a poorly designed and probably unneeded piece of automation that itself crashed the plane.
The moral being that we should not attack people who actually know something about the situation but instead attack the design decisions. And those revolve around automation and software far more than the placement of the engines. There are other blogs run by pilots (you know doubt would disagree based on your great expertise) and I’ve yet to see one of them that says the plane itself should be permanently grounded.
You’re right, and why is it that things are the way they are?
Are you saying it isn’t corporate policy to hire ”very low paid aspiring pilots”?
Are you saying that the fact that “The old days when pilots were highly experienced ex military flyers are largely over” is the natural result of market forces?
I say there is not a pilot shortage so much as there is a reluctance to pay for experienced pilots.
Are you saying that “737 Pilot” didn’t blame the crashes on the dead pilots inability to rely on their training or follow the Quick reference card?
And finally, when did I say the plane should be permanently grounded?
What I am saying, what I did say, was that based on evidence now in hand, the 60 minutes Australia video, the audio of Boeing execs claiming the first crash was a “once in a lifetime event”, and the pilots union reaction, that IMO, “737 Pilot” was testifying here in the interest of diluting the blame Boeing rightly deserves.
And I’d say the most convincing evidence is the stark contrast between how the pilots in the video treat the facts, and the obvious empathy they felt for the dead, as compared to the blame “737 Pilot” leveled at the doomed pilots.
He replies to you below so no point in my putting words in his mouth.
But the truth is that you don’t know that pilot error wasn’t partly to blame since no conclusions on either accident have officially been drawn.
And finally I don’t believe in attacking other commenters and imputing motives to them which you don’t know either. We are all here just giving our information and opinions. Moderation is not your job.,
> But the truth is that you don’t know that pilot error wasn’t partly to blame since no conclusions on either accident have officially been drawn.
That’s not strictly true: “Ethiopian Airlines disaster: Read the full preliminary crash report.”
I did read the report, and I do believe that the pilots (lack of training and experience in particular) contributed to the crash.
The key assertion in 60 minutes is that when MCAS kicks in, you cannot do anything for 10 seconds, and then have only 5 seconds to do anything. (“It means they have only five seconds of control before MCAS overrides the plane, forcing it to nosedive.”)
This assertion is false based on the facts in the report, which shows that when MCAS kicked in, 8 seconds later (right when both captain and pilot started trimming) ,MCAS is out and the pilot-electric trim kicks in, and they manage to increase the trim. You can see it also clearly on page 26 which trim/notrim graph. At around 40:28 the lower one (MCAS) stops immediately when the upper (pilot-controlled trim) starts, and does not engage until 5 seconds after the pilots stop trimming (which goes on for about 20 seconds), and the trim moves from 2.1 to 2.3 (page 10)
What we do not know now, and may not never know, is why the pilots did not continue trimming.
What we can see though, is that 20 seconds of trim by pilots got the trim up from 2.1 to 2.4, while 10 seconds of MCAS got it from 2.4 to 0.4
That to me suggests that there was some additional problem with stabiliser (and the captain seems to idicate there was a problem before the first large trim occured) that was preventing it from operating.
Please read this comment. I need to turn in and this covers the waterfront very well:
The repetitive “doom loop” nature of the MCAS system — on, then off, then on again (as the sensors kick in again) — has been part of the coverage from early days (and hence, can trim/can’t trim). I’m not sure I’d set to much store in a simulator number (10 sec) vs. the real number (8 sec).
I don’t have time to dig this out again, but IIRC there is an issue — this is very poorly written because I haven’t mastered the jargon — where it’s possible for the force of the air working against trim to overpower the physical ability of the pilot to turn the wheel, as the jackscrew makes the position of the stabilizers more and more extreme, due to the doom loop.
I always thought one of the ground rules here at NC was to “play the ball and not the person.” It seems that rule depends on which side of the discussion that person is on….
You are arguing from an assertion of authority. First, that is a questionable basis of argumentation. Second, we cannot verify your claims about your background. So you have made this sort of challenge legitimate.
Having read almost all of the comments by 737 Pilot on this subject, I have never seen him use his authority as his primary method of argumentation. What I have seen are facts being shared and explained. Most of those facts are verifiable through primary sources on the internet. He also takes pains to clearly point out any opinions as such.
No, see his repeated claim that if the pilots had followed standard procedures (for runaway stabilizers and on this thread, arguing that the pilots should have engaged in another standard procedure, “Airspeed Unreliable” by depicting the plane as well above the 1000 foot level, where it would presumably be safe to slow down the plane, when the detailed record of the Ethiopian Air crash show that it had barely crossed 1000 feet before it started getting nosed down, and was almost certainly below that altitude.
In other words, he has repeatedly presented opinion as fact.
All one has to do is read the preliminary report, which clearly vindicates the Travis analysis. I know Gregory and also know for a fact that he is being regularly consulted by Ralph Nader (whose niece was killed in the Ethiopian Airline crash). Fascinating. Note early on the mentions of a “event” associated with the left AOA sensor’s heater. The 737 MAX Master Minimum Equipment List allows the 737 MAX to be dispatched (take off) with both AOA sensor heaters inoperative. The AOA heaters are used to prevent the AOA sensor vanes from freezing in place (from water in the system). This can happen (fairly commonly) if the aircraft gets wet on the ground (i.e it is raining on the ground) or even if something like the aircraft gets washed on the ground. Then the aircraft climbs to altitude, where it is cold, and the water freezes, freezing the AOA sensor in place. I do not know if this is causal in this case but that sequence did cause two AOA sensors to freeze on an A320 (airbus) demonstration flight, causing the aircraft to lose its stall protections. Of note the Master Minimum Equipment List for the A320 will not let it take off if less than two AOA sensor heaters are working.
Also note that the faulty left AOA sensor goes from 35.7 degrees to 74.5 degrees in 3/4 of a second (05:38:44). This is not physically possible (without causing the aircraft to break up) and a correctly designed MCAS software would have identified that as a clearly faulty sensor and disregarded that sensor’s inputs from that point forward.
The sequence between 5:42:54 and 5:43:20 is where things fell apart. The crew correctly identifies the problem “Left Alpha Vane” (AOA sensor). Then, at 05:43:04, the Captain pleads with the First Officer for help in holding the column back (the bitey dog is biting, hard). At this point they have already disabled MCAS (by pulling the cutout switches (see photos, below)) but earlier transcripts show that MCAS had trimmed the aircraft nose-down so much that they could not re-trim manually, using the manual trim wheel (they could not physical move it) (manual trim wheel is #4 in photo, below) (05:41:51).
Now the captain can’t hold the column back himself and is asking the first-officer for help. They’re getting bitten to death.
At 05:43:11 it is clear that they have re-engaged the stabilizer trim cutout switches in desperation, so that they can use their trim switches on the control wheel (#1 in photo, below) (because they cannot keep holding the column back). The stabilizer moves nose-up (ANU) (wheel, #4 in photo below, would have moved on its own) in response to manual electric trim.
Five seconds after they use their electric trim switches to move the stabilizer (05:43:20), MCAS kicks back in and, because the cutoff switches are in their normal position, MCAS is able to reverse the trim input they just made and slams the nose back down (AND) (wheel, #4 in photo below, would have been spinning so fast that it was literally a blur) — taking the stabilizer to its electric nose down stops (0.8 units). At that point the crew loses control of the aircraft.
Here is the full report:
Knowing something and knowing the facts are two different things. While the “737 Pilot” may actually be a pilot and “know something” about the aviation characteristics of the 737, s(he) apparently did not know all the facts. And extolling your aviation chops while somewhat denigrating the aviation skills of the pilots who died in the recent 737 crashes (one of whom had 8000 hours of flying time), doesn’t look good.
Okay, I’ll play.
Which “facts” did I not know?
You don’t have access to the black boxes on both flights, yet keep making definitive statements as if you know everything that is germane and nothing else matters. You do not have a full set of information and are in no position to make the sort of absolute declarations that you do. No expert defending Boeing to the press has said anything as extreme as you do here. You’d expect them to have made comments like yours if they would have stood up to fact-checking.
The flight data recorder information from the Ethiopian flight has been made public. 737 Pilot has only made arguments based on that.
You are now straw manning readers First, the issue is what information “737 Pilot” did not have. Bystander said “black boxes” on BOTH FLIGHTS. No information has yet been released on the Lion Air flight, and on Ethiopian Air, it was the voice recorder only. The official report gave information from the flight data recorder, the other part of what is generally called the “black box” but I haven’t seen any indication that that data has been released in full.
And as I pointed out below, the official Ethiopian Air report contradicts 737 Pilot’s repeated claim that running the trim stabilizer protocol would have saved the plane. So what good is your contention that he argues from the flight recording when it turns out he cherry picked?
Moreover, he later does go on at length about Lion Air.
He similarly confidently asserts that the pilots could/should have slowed airspeed (per another standard protocol) because they were above 1000 feet. But he bases that on the fact that the flaps get retracted then, and MCAS doesn’t kick in until after that. But that does not prove the plane was above 1000″ feet after the nosediving.
And the Ethiopian Air report also says the pilots were getting bad altitude readings:
And MCAS kicked in promptly at not much over 1000 feet, as soon as the flaps were retracted:
So 15 seconds after flap retraction was INITIATED, MCAS kicked in and put the plane into a slight descent.
You can read this for yourself. The pilot got two “DON”T SINK” alerts after that, meaning the plane had descended further. The record also clearly shows what the findings indicated, that the pilots did attempt “runaway stab trim” measures:
In other words, you have accepted 737 Pilot’s convincing cherry picking of the record. Congrats.
The fact that they should have cut airspeed is not just 737’s assertion, I have seen it mentioned a number of times elsewhere.
From the same doc you’re quoting:
“From 05:40:42to 05:43:11 […] the left indicated airspeed increased from approximately 305 kt to approximately 340 kt (VMO) [vlade: VMO is the maximum operating speed. They were flying way too fast]. The right indicated airspeed was approximately 20-25 kt higher than the left
At 05:41:20, the right overspeed clacker was recorded on CVR. It remained active until the end of the recording. [vlade:left clacker went on 10 sec later]”
At that time, they were NOT falling (gaining speed via gravity), it was almost three minutes before the crash happened.
That said, there are inconsistencies in the report. It says that at 5:39:22 “the flaps were retracted”. Only to say at 5:39:45 “requested flaps up […] flap handle moved from 5 to 0 degrees and flap retraction began”. Shortly after that, the captain asks to keep runway heading, indicating he believes they have a problem (confirmed by telling so the FO 7 sec later). This was, if I read the record right, BEFORE the first MCAS engagement (which happened at 5:40:00). Or at least before the first mention of it in the document.
Also, I’d point out that the record confirms 737 pilot’s assertion that MCAS can be overriden by the pilot on the yoke. Namely
“At 05:40:20 […] a second instance of automatic AND stabilizer trim occurred and the stabilizer moved down and reached 0.4 units
At 05:40:27, the Captain advised the First-Officerto trim up with him.
At 05:40:28 Manual electric[vlade: emphasis mine] trim in the ANU direction was recorded and the stabilizer reversed moving in the ANU direction and the trim reached 2.3 units. ”
I.e. 8 sec (not enforced 10, as implied by the 60 minutes) later, the very moment Captain and FO started (electrically) trimming up, they overrode MCAS and put the trim back. Not sufficiently (the initial trim was >5 units) I guess, but at this stage they were already over the safe speed, which might have had some effect.
When then they tried to trim it manually, they failed. But, IMO, at the speed they were going (the overspeed clacker was ON at the time, remember), they had no chance to do so.
Also, reading the report up a bit more.
Yes, the captain had 8000 hours. But only 1400 of them on 373, of which 266 were in the last three months. That is, sixth of his 737 flying time was in the last three months. He got his Pilot-in-charge rating about year and a half ago (Oct 17). Which means he was pretty green, and didn’t fly that much. His FO rating was received 8 years ago, Jan 2011. That means he flew ~1100 hours in about 8 years, which is about 140/hours year.
FO is even worse. He has TOTAL hours 361 of which 207 were on 737, ALL of which were in the last 90 days. He was out of training Jan 31 THIS YEAR.
As a comparison, my commercial pilot friend has > 8000 hours on his airbus (Captain and FO) and when he applied for a position this year as a captain in Japan, he was told he’d have to have >10k on airbus (not just some flying time) to qualify.
Those are all (ex my comparison) facts, from the document you quoted.
To me, they paint a picture consistent with what 737 pilot is saying: A highly problematic piece of equipment leading to emergency which an unexperienced crew was unable to handle.
But I still maintain what I wrote earlier in a comment – failure of a 300USD piece of equipment should not cause even greenish captain to crash.
You chose to pick up a secondary argument about airspeed is a secondary argument made by “737 Pilot” only in this post, and only in an exchange with Synoia, where Synoia questioned cutting air speed at what he said was 600 feet.
You ignore that the claim repeatedly made by “737 Pilot” that executing the “runaway stabilizer” procedure was disproven by the report.
He has made that assertion so many times as a fact rather than an opinion or informed speculation that he has hung his credibility on that.
On the secondary argument about cutting the airspeed to correct the inconsistent readings, “737 Pilot” without saying so in a straightforward manner argued that the MCAS kicked in after flaps retraction (at ~1000 feet) and therefore the plane was higher than that altitude (he implied MCAS kicked in later, giving the impression a bit later, as opposed to just about immediately after).
That undermines the thrust of his argument that the plane was clearly above 1000 feet, ergo cutting speed was OK (he curiously didn’t make your argument, the plane was going so fast that there was room to cut speed, although query whether the protocol works if you merely have a relative drop in speed, or the plane’s absolute speed had to drop below a certain level).
Cutting speed was ok regardless of whether the plane was above or below 1000 feet. The altitude is red herring.
Again, there’s NO altitude where you cannot safely cut speed from overspeeding. They were not doing a few kts more than they should. They were breaking the operational aircraft speeds, on an aicraft that had nose-down tendency.
That is just wrong, and it was, from the report, entirely crew’s fault.
The official report did not say the pilots were “wrong”. The detailed report of the flight says they pilots got two “don’t sink” warnings. The right overspeed clacker didn’t start making noise until after this failed effort to stabilize the trim:
The left overspeed clacker started up shortly thereafter but it was only active “intermittently”.
The plane did go very fast when it was terminal:
This is the full “Initial Findings” section. Please tell me where in this section or anywhere in the report that it finds fault with the pilots for the airspeed:
This is an initial report. The speed might, or might not, be a contributory factor.
I believe it likely, as high speed puts a lot of force on all control surfaces, which is IMO very likely the reason why fully manual trim wasn’t working.
Whether the trim system was, or wasn’t ok, is not something that the preliminary check would not be able to find.
I’m willing to bet that if the trim system was ok, it was the speed which prevented the crew manipulating the manual trim. There was no other reason offered for why the manual trim would not have worked, and really, only mechanical malfunction (which would have to be very specific so that only manual trim would not work, but the motor driven would).
Which would mean that the speed WAS one of the chain links in the crash, and the speed WAS under crew’s control. But for that, we’ll have to wait for the final report.
As a speculation, there’s possibly additional point in the report which I believe favours the speed argument. In the 20 or so secs of pilots trimming up, they get it moved from 2.1 to 2.4 units.
MCAS then trims it down from 2.4 to 0.4 in 10 seconds, much faster than pilots are getting it up.
Assuming that 0 is flat, and that under “normal” conditions the trim up is as fast as trim down, this would be consistent with significant force being applied to the stabiliser, which would slow down trim up, but speed trim-down.
But this is just a speculation – there could be other reasons for that.
Regardless of all above, I stand by my point. Excessive speed (and there was no reason for overspeeding, and I haven’t heard any anywhere) makes any control situations worse. Any maneuvring you make take more space, any reaction you have to make has less time.
> Okay, I’ll play.
There it is.
Are current military pilots reduced to trained automation system operators? Too?
If 737 Pilot was so obviously dishonest it should be trivial for you to cite an example from the comments he left. He was extremely careful to state that training, manuals, and procedures differ between airlines – the video is therefore not representative of all situations. 737 Pilot’s conclusion was still that Boeing should be held liable and to account for the design of MCAS.
It is NOT “trivial” to debunk people who trot out arguments that have a lot of (often accurate) technical information along with the stuff that is a stretch or a sleight of hand. This is why I have a business. I have since 2006 been showing how people in finance and economics make persuasive-sounding arguments that use insider detail and jargon that are in fact bogus.
It isn’t easy to refute this sort of thing. Look at how it takes me paragraphs to debunk simple statements designed to induce nod-your-head agreement like: “We need private equity because it is the only strategy that will beat our return targets.”
Nevertheless, in this comment and subsequent comments, I show that the official initial report on the Ethiopian Airlines directly contradicts “737 Pilot”‘s repeated, central argument, that the pilots simply needed to engage in basic “runaway stabilizer” procedures and all would have bene fine.
And let us be clear: this is blaming the pilots.
On top of that, when presented with this evidence at the very start of a comment further down in their thread,”737 Pilot” ignored it and instead went on about the Lion Air crash (which I mentioned later in the same comment) and good and bad press information, when the press had absolutely nothing to do with the Ethiopian Air official report. This is obvious bad faith argumentation. The Lion Air part of my comment was secondary to the Ethiopian Air evidence, but he dealt with only that to divert attention from the Ethiopian Air showstopper, and even then straw manned parts of what I said.
Perhaps 737 pilot was clued to the operational changed done by Boeing, before or after the crashes. The isolation procedures changed between NG and Max. It seems lots of pilots were blindsided by Boeing’s failure to document the changes, and train pilots on how procedures changed.
The 2 things that I recall that disabled MCAS were flaps down and trim switch actuation. And you had to bump the trim switch every 5 seconds.
If I understand 737 Pilot’s argument correctly, he is saying that on a plane flying as fast as the Ethopian plane was flying that close to the ground, the trim just can not work. Not because of this button or that, but because the forces generated on the plane by air that dense passing by with that much velocity simply overwhelm any and all trim controls.
On the PPRUNE site, that particular argument seemed to be accepted.
I do not know enough to judge whether or not this argument is correct – hopefully some of our bright commentariat does have the requisite background – but I thought it important for that argument to be presented in full.
IIRC, that was what Pilot 737 and another guy (JerryDenim, also saying a commercial pilot) were saying – that even killing the electrics to the stabiliser and trying to trim manually would be impossible at the speed they were going. IIRC, one of the questioned why they were going at the speed they were..
TBH, their main points were sub-par pilot training. Which I buy, but I buy more a riposte in the comments there – “My life as a passenger should not depend 50/50 on whether the pilot at the controls is an average pilot or a good pilot when a single 300 usd part fails.”
As I understand it there is no speed restriction on operating the stabilizer trim motor(s).
Boeing in the 737MAX ganged, linked, the stabilizer off switches: previous 737 have two switches, automatic inputs off, and stab trim motor off.
There seems to be very little about maintenance, failure history of AOA sensors in the news.
it’s more that if you turn off the motors, it’s impossible to use the manual trim wheel past certain speed as the airpressure on the stabilisers is way too high.
I believe that the Ethiopian crew tried that (motor off), but found they could not trim manually (due to too high a speed), so switched the electrics back on, which then made MCAS to kick in again and the situation worse.
Pilot 737 and JerryDenim points were here (IIRC) that even at the point of crash the thrust was full on takeoff thrust.
The absolutly LAST action the Ethopian Air pilpts would do at 600 ft off the ground is to cut airspeed.
That would cause a Pilot assosted crash,.
Airspeed, flying, was their ONLY escape.
The MCAS problem did not present itself until the flaps were retracted, well above 1000′. However, the failed AOA did cause an immediate airspeed discrepancy for which there is also a “memory” procedure. Among other things, that procedure calls for the pilots to reduce the power setting on the engines. Neither accident crew performed this “Airspeed Unreliable” procedure and as a result neither aircraft were as stabilized as they could have been when the flaps were retracted thus fulfilling the final condition for MCAS to activate.
I love the bureaucrat-speak: “The MCAS problem did not present itself”. Lack of agency and lack of specificity.
The excerpt from the Ethopian Air report shows that I cite above shows that MCAS kicked in 15 seconds after the pilots INITIATED the flap retraction, which means about 5 second after it was competed. The nose down started a mere 28 seconds after the plane hit 1000 feet. The first round was only a “slight descent” and the pilots likely thought they had solved the problem and didn’t consider more aggressive responses until the second and third incidents (the report shows two later “DON’T SINK” warnings).
Shorter: the plane may not have been at 600 feet as Synoia suggets, but your really blowing smoke to imply it was comfortably above 1000 feet when the record unambiguously shows otherwise.
MCAS is off until after the flaps are retracted. 5 secs is in line with what is the time for MCAS to kick on.
Synioa’s point is, TBH, wrong. What you want at 600 feet is LIFT. Lift does not equal airspeed.
My point was the thrust was full on. This means they were going way faster than they should have. In fact, the report itself says that the “overspeed” clacker was on for full three minutes before the crash (see my comment above).
For any flight where safety is paramount, there is zero, literally ZERO, reasons to overspeed at 1000 feet, and even less at lower altitudes, especially with a plane that has a tendency to dip down. All it gives you is less time to react, and potentially more problems due to overshooting aircraft’s limitations.
Huh? Greater velocity of a plane = greater lift, all other things being equal. Your argument makes no sense. Google even has posts on how slowing down a plane makes its nose go down!
No. The lift depends on AOA, the wing profile and speed. All these control the lift.
You control wing profile via flaps, AOA via pitch, and speed via thrust. All of those have to work
It’s true that there’s no lift under a certain speed, but there’s no lift over certain AOA either, or with some wing profiles. This is for example why saying “stall speed” has no meaning unless you put it into some context.
Moreover, over certain speeds you get drag competing with lift, which also modifies the properties of the airfoil, potentially substantially (no-slip condition, basically air “sticks” to the airfoil). Which are all reasons why super-sonic aircraft have to have different designs (arguably, super-sonic flight is actually easier, once you get there, and the speed=lift becomes cleaner).
In context, we are taking about the 737 Max, not supersonic or other plane designs.
I started with “all things being equal”. I don’t mean to sound cranky, but this is 6 AM and I am behind on posting. It is a deterrent to participating in comments to have you play “gotcha” for my not having gone back to square zero and thrown in all the qualifiers in the middle of the night on the fine points of a long-running argument.
Using the flaps wasn’t part of any recommended response to this problem, so bringing them in is diversionary. Please tell me how the pilots could have gotten more lift if they couldn’t do it with the trim, as recommend. Absolutely no one that I have read has suggested using flaps. So pray tell, how were they supposed to get more lift?
And the implicit point by Synoia was that cutting the airspeed at a not-very-high altitude did risk loss of altitude. I don’t see you as able to put this concern to bed. We don’t yet have a definitive reading on what the altitude was when the pilots were flying the plane too fast.
This as we discussed at length was a runaway stabilizer problem created by MCAS. The remedy was to deal to try to get control of the trim. The post describes how there appears to have been yet another Boeing-induced impediment to doing that, in that the switch change meant cutting off MCAS also adversely affected the thumb controls for the trim, which would have been the pilot’s first line of response.
“737 Pilot” did not dispute the idea that cutting air speed below normal “flaps down” altitudes might not be a good move. He finessed Synoia’s 600 foot issue by effectively taking the position that the flight was well over 1000 feet. That appears to be debatable at best, and probably not the case.
At the speed they were doing using the flaps would be a major problem TBH, and no-one could argue with that.
Let’s park speed=lift argument (because all things are equal only in lab, if that. I already gave example of “stall speed” as being dumb number if it’s not in a context. You can happily stall at a cruise speed. I’ve done so on simulators a number of times).
Amount of lift you get is usually controlled by AOA. Which, in this case, was problematic due to tipping to nose.
But if you’re tipping to nose, you’ll not get much of move away from the ground by increasing speed. Lift is force perpendicular(ish) to the airflow, not an “up” (as against gravity) force. If you’re nose down a lot, higher speed means that you hit the ground faster, as you’ll never get enough lift to compensate the main thrust vector which is downwards. If airspeed = lift, then the most lift they ever got was just before the impact.
You need lift to stay in the air. That doesn’t mean you need as much airspeed as you can get. And again, there is extremely few situation where you NEED to overspeed, because overspeeding means you’re putting the airframe past of what it can deal with (safely).
In the days past, overspeeding in a dive meant you lost your wings. Literally. Boeing may have substantial safety margins built in, but going into those margins should require a good reason. The wasn’t any given in the report.
You are now straw manning. No one said in this context that increasing the speed of the already too fast 737 Max was a good idea.
That wasn’t the point made by either Synoia implicitly or me taking issue with “737 Pilot”‘s misdirecting response to him.
The issue is that given that the plane was overspeeding, and quite probably at under 1000 feet, was cutting airspeed going to make things better? More specifically, was cutting the airspeed enough to make the indicators behave (due to the hour I’m not looking up the names of the ones affected; 737 Pilot argued for cutting the airspeed as a remedy for the disagreement on the pilot v. copilot altitude and airspeed sensors).
So this isn’t just about cutting airspeed to get it within some sort of normal parameters. This is cutting airspeed to get the sensors to behave. That was the driving reason for making it a priority. But how much of a reduction would it have taken to fix the sensors?
This gets into the general issue with tightly coupled systems, which is arguably part of the picture here: risk reduction measures actually make matters worse. Mind you, that is not what I am claiming, but Synoia raised that issue at 600 feet. “737 Pilot” didn’t dispute the point re this being a risky response at 600 feet, he strongly implied the altitude was much higher than the Ethiopian Air report says it was and at that altitude, cutting the speed would not be a risk.
So we have “737 Pilot” implicitly conceding that cutting airspeed, even high airspeed, at 600 feet was a questionable response. He implicitly says cutting it at comfortably over 1000 feet was warranted. But it appears that the plane was either barely above or just below 1000 feet and “737 Pilot was arguing heights higher than that were safe, not necessarily that altitude (he was silent on that point). Until we have a better altitude reading on that plane after the initial small nosedown, we can’t reach any conclusions.
“The absolutly LAST action the Ethopian Air pilpts would do at 600 ft off the ground is to cut airspeed.
That would cause a Pilot assosted crash,.
Airspeed, flying, was their ONLY escape.”
This is just wrong. It says that they should never, ever, cut speed no matter what speed they had, or what other things were going on.
If I take it to extremis, 737, or any subsonic plane will crash when it tries to breach sound barrier. They would start losing control even before they got there. Again, according to Synoia, the pilots should not decrease the speed.
Similar, after you retract the flaps, your speed increases (flaps have drag). They could have possibly reduce speed and extend flaps, and would be still flying ok (as they were up to the moment they retracted the flaps). No loss of lift, but according to Synoia, it should never happen. I guess that means planes land at the cruising speed, and never abort landings (i.e. decrease speed, then increase, safely)
The reasons why you have max operational speeds are two. One is structural integrity, and the other is ability to control. Let’s ignore the first one for the moment, as it likely played way less here.
The simple fact is the faster you go, the more pressure on your control surfaces. Manual or electric controls can deal only with so much of the force – at some point they stop working, because you just can’t move them.
So overshooting their operational speed – and I’d point out that the 340kts is speed at a cruising altitude, with much less dense air, so again, more pressure on controls – limited their maneuvrability, while giving them NO advantage. If they wanted climb, they could have climbed with less thrust – with a climb, not take-off setting. I bet you noticed how when the plane takes off, at one moment it “slows down”. Well, that’s when they reduce the thrust from take-off to climb (among other things, running on full thrust is very bad for engines, as its the hottest setting)
I have not seen a single mention anywhere which would have said “yep, they were right to run full on thrust”. I have seen numerous people at least politely raising eyebrows on this.
Also, I’d really like to see where that 600 feet figure comes from.
According to the records, the plane was at more than 5000 radio feet until about 15 seconds before impact. You can see the radar profile at page 27, second part (top is atmo pressure alt). You can see that they were at about 1k radio for a long time, until about 40:45, when they started climbing, and were (more or less) climbing all the way till about 43:15.
Looking at the map, the terrain around Addis Abbaba is not much higher than the airport – less than 9k feet atmospheric. Which is a few thousand feet below what the flight was for the last two minutes.
> So overshooting their operational speed
I would very much like to know what Ethiopian Airlines itself specifies as the standards for speed is. Addis Ababa is high altitude, so does thinner air require higher speed on take-off? Also, are we sure the speed sensors were accurate? (I don’t have time to look at the Ethiopian Airlines preliminary report).
Indeed, I just found this:
“The combination of the plane’s speed – edging up towards design limits with the engines still at their take-off power – and the trim setting meant the pilots would have had to exert 50 pounds of force to pull back the control columns, the four experts and one of the pilots said, and moving a backup manual wheel instead was impossible.”
As I recall Boeings instructions were to turn off the trim system to disable the MCAS and then manually adjust the trim. However, at high speeds the air force on the trim parts make it almost impossible to manually adjust the trim – you can’t get enough leverage to overcome the forces. This is actually a known issue with the 737 going back to its launch in the 60’s. In early manuals Boeing instructed pilots to temporarily put the plane into a dive and pull up and repeat. During these dives there will be times when the plane is near zero g’s and the force is reduced allowing the pilots to manually adjust the trim. An interesting thing is that newer manuals have not much more than a footnote about this maneuver.
This video shows it in flight simulation as well as the old manual documents…
Think about the last time you flew. Accelerating down the runway until the plane is going fast enough to fly. V sub r, or velocity to rotate. Is when the tail goes down and the nose comes up and the angle of attack increases and then the plane is flying. I’ll guess that engine noise is proportional to power, and gee whiz it seems like it’s loudest during take off and climb out. Trading fuel for power and power for altitude. To get some distance from the ground if something were to go wrong.
So, you’ve got takeoff power settings, retracted flaps and gear up – configured for least drag climb out. Only, you aren’t climbing anymore. You are trading power for speed, not altitude. Why they were going so fast. When they crashed.
It is counterintuitive to reduce power, lower flaps (just canceled MCAS) put the gear down and land back where you came from. Have you ever experienced this on a jet airplane? Likely not.
Okay, to clarify a few things.
Electric trim works throughout the entire certified flight envelope. I can’t speak for what happens if you go outside that envelope which the Ethiopian pilots did in the final minutes of the flight.
In severe out-of-trim situations, it is very difficult if not impossible to trim the aircraft using the manual trim procedures. This is why it is very important to not let the aircraft get out of trim before cutting off the electric trim. The Ethiopian crew cut power to the electric trim prematurely, thus putting themselves in a positions where they could not trim the aircraft properly. The Runaway Stab Trim procedures has a number of steps that must be done sequentially. Step 2 has the pilot use the electric trim to return the aircraft to a neutral trim state. It is not until Step 5 that they are directed to cutoff the electric trim.
Lest everyone think I’m just beating up on the pilots, let me be clear that these pilots were probably performing to the limits of there training. Professional pilot training has been going downhill across the board for well over a decade. This is not just a third-world country problem either. Like many endeavors, there really isn’t any “faster, better, cheaper” solution. At least one of these three gets sacrificed, and I’ll leave to the audience to guess which.
From the doc:
The RS is included on page 30.
1 Control column .. Hold firmly
2 Autopilot (if engaged) Disengage
Do not re-engange the autopilot.
Control airplane pitch attitude manually with control column and main electric trim as needed.
3 Autothrottle (if engaged) Disengage
4 If the runaway stops after the autopilot is disengaged
5 If the runaway continues after the autopilot is disengaged
STAB TRIM CUTOUT
switches (both) CUTOUT
If the runaway continues
Stabiliser trim wheel .. Grasp and hold
6 Stabilizer Trim manually
7 Anticipate trim requirements.
So, while what you say (cutout only after in neutral) makes sense, it is NOT what is actually in the operating manual.
This is what the document says on the airline procedure (dated Nov 6 2018):
In the event an uncommanded nose down stabilizer trin is experienced […] do the Runaway Stabiliser NNC, ensuring that the STAB TRIM cutout switches are set to CUTOUT and stay in the CUTOUT position for the remainder of the flight.
Initially, higher control forces may be needed to overcome any stabiliser nose down trim already applied. Electric stabiliser trim can be used to neutralise control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT. Manual stabilier trim can be used after the STAB TRIM CUTOT switches are moved to CUTOUT.
Again, no indication that electric trim should be used before CUTOUT.
Ok, now I’m really confused…. how does this:
“Again, no indication that electric trim should be used before CUTOUT.”
agree/disagree with these:
“Control airplane pitch attitude manually with control column and main electric trim as needed.”
“Electric stabiliser trim can be used to neutralise control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT.”
Sorry, it should have read “should be used to get to neutral trim before CUTOUT”. My bad, the “neutral trim” is critical there.
“as needed” and “can” are not the same as “get it to neutral first”.
An experience pilot knows that. Someone who has never been in emergency, under real stress, not so well.
Boeing and its Seattle area work force were once a true jewel of the American economy. Rather than punish the entire company, it would be far more just to remove all those responsible for financializing and crapifying it.
That the rest of the business elite is not up in arms about this is one more proof that we live under a decaying capitalist regime, not a growing one.
The robber barons of the late 1800s and early 1900s and the Chinese leadership of recent decades were not better people, were no less brutal, but at least those autocrats built up the economy rather than run it down.
I think we’re living out the Law of Diminishing Returns, exacerbated by the Law of Unintended Consequences (“there always are some.”)
In particular, entangled with the social failures, is that we’re running out of underlying,physical resources. We’re at the physical limits of the system. This is a situation that all but a few economists actually deny, rendering themselves irrelevant to the real world.
I think that one change that will have to come out of this lethal fiasco is that no longer will safety gear be an optional extra on Boeing aircraft that airlines will have to pay for. Things as basic as the oxygen system for the pilots. All that stuff will have to be included as a standard package with each aircraft but I do not know if the FAA will have the courage to have this change pushed through and made into a law.
An unanswered question in all this is who made the bad decisions and what were they thinking? It feels like amateurs took over Boeing, Jared Kushner types, who don’t know what they are doing, but nevertheless irresponsibly presumed to make design decisions despite their ignorance. My impression is that airplanes are supposed to be designed so pilots are always in control. The MCAS system violated this principle by being able to override the pilot. The idea here was probably to make the plane seem as much as possible like an old 737, throwing away basic design rules and pretending it didn’t matter.
MCAS could not override the pilot. The pilots let MCAS override them. See discussion below.
Let me put it this way: to deactivate the MCAS system the pilot must shut down the electronic systems, especially the electronic trim control. Therefore, the pilot must pay a “price” to take control from the MCAS which jeopardizes the plane. Does this violate the rules of airplane design? Shouldn’t the pilot be able to take control more easily, in a way not exacting a “price”?
The short term way to counter MCAS was to use the yoke trim switch. The long term way was to shut off the electric trim. The smarter way (though not covered by any existing procedure) was to slow down and extend the flaps (MCAS only works with flaps retracted).
Doing any of the above in the proper manner does not jeopardize the aircraft.
My understanding is that the MCAS system overwhelms the electronic trim control. Is it actually possible to counteract the MCAS system with the electronic trim control alone?
Okay, I see you address my question in a later comment:
“As soon as a pilot puts in any amount of trim with his yoke switch, MCAS stops right there – it does not keep going. The pilot can then take out any trim that MCAS put in. After 5 seconds of no input, MCAS will try again, but again, the pilot can stop MCAS in its tracks – none of this letting the trim run for 10 continuous seconds. (BTW, this functional description refers to the original software, not the replacement)”
Perhaps we need to wait for more information from the crash investigations to understand what went wrong.
Yes, though it will be a constant back and forth fight between the pilot and MCAS until the system is disabled.
How much of a safety issue is there in shutting off MCAS; Does the pilot now have to worry about the plane stalling?
As a general rule, commercial pilots try not to get anywhere near a stall.
There are circumstances that either from inattention or some environmental effect like windshear, the aircraft may get dangerously close to a stall, but these are decidedly rare events. Assuming that MCAS was designed right in the first place, it is entirely possible that the entire MAX fleet could operate for years before MCAS was ever needed.
There are numerous systems on any commercial aircraft that can malfunction. When that happens, we execute the appropriate procedures and either continue the flight to destination or land short at a suitable airport. In either case, the procedure will provide guidance on any additional precautions that should be taken. When the MAX is returned to service, I suspect that we will have a new non-normal procedure that will provide this guidance. Offhand, it will probably just advise pilots to exercise greater diligence and/or restrict the flight envelope.
How is the procedure, hardware, and software different between NG and Max.
Who in management didn’t know the difference. The engineers share should have. The pilots would have told you, if the were informed of the changes.
As I understand it, the NG should not have an MCAS system. The engines on the MAX are too large to locate under the wings; they are located in the front. This makes the plane unstable; the nose can tilt up, producing a stall. The MCAS system was introduced on the MAX to counteract this tendency using the elevators. That way Boeing could say the MAX did not require new training for 737 pilots. Incredibly, Boeing hid information about the MCAS system from pilots until the Lion crash.
“My impression is that airplanes are supposed to be designed so pilots are always in control.”
Boeing is known for a philosophy that the pilot should have ultimate authority to over-ride the automation.
Airbus is known for a philosophy that the automation should be able to save the pilot from fatal mistakes.
Neither approach works in every case.
In the 737 MAX crashes, ironically, the aircraft were lost because of a failure to enable the pilots whether through technical errors or training failures, to over-ride the automation in time.
Equally ironically, Air France 447 (an Airbus) was lost in the South Atlantic because a pilot could over-ride the automation. That case also started with a sensor failure, producing incorrect airspeed readings. The autopilot recognized an emergency, and switched from ‘normal law’ to ‘alternate law 2’, granting the pilots more control. The pilots in the cockpit had almost 10,000 hours of experience, over 5,000 on the A330, yet made errors in flying and procedure, apparently missed or ignored alarms, and failed to figure out what was really happening. In any case, a confused pilot held the nose up, keeping the plane in a stall, for minutes until it hit the ocean.
Sometime the pilot can save the plane, and the automation can’t… sometimes the automation can and the pilot can’t. We mostly hear about the failures of both approaches. After reading scores of accident reports (a well written accident report contains a wealth of insight into how systems fail) my impression is that today, the pilot is more likely to break the airplane than the automation.
Of course, in some kinds of failure, there are no ways to prevent a crash.
Certainly it does not appear that increasing automation has increased crash rates overall.
It would be a major oversimplification to credit technology with the gradual improvements in aircraft safety… though in some ways it doubtless helps if only through increased reliability of some systems.
I would like to think that with good training pilots can be counted on to rescue a plane. Should automation be able to take control of a plane from a pilot? What should a pilot do in that situation? Intervene or do nothing?
Anyway, your basic point is that allowing automation to overrule the pilot can be a valid design– O.K. However, the fact that Boeing tried to hide the MCAS system from pilots says to me that they knew they were doing something improper but did it anyway.
I don’t think that Boeing’s top management was stupid. They just had different goals from what we would want. They worked for the short-term rate of return on investment and their own bonuses, even at the cost of the long-term health of the company and the unparalleled workforce. This is what finance capitalism does to industrial capitalism.
Were the 737 MAX crashes predictable? I think they were from the problems with the plane’s design. Therefore there was something wrong with the decisions Boeing was making. Even if the Boeing management was motivated solely by greed these crashes are a disaster for them, a predictable disaster.
As I keep saying, Boeing didn’t have the capital, given (a) the Boeing 787 debacle and (b) stock buybacks and executive compensation to build a new aircraft to replace the 737 (and take any hit to their cashflow). So we got the 737 MAX instead.
So the car, as it were, went over the cliff with the 787. So “predictable” in the sense that the car would roll and roll and parts would come flying off until it finally came to rest, possibly catching fire. Can’t predict the trajectory of any givenflying part, though.
What seems to have really loused up the 737 MAX design was the requirement that pilots not need new training.
Which was a key selling point.
The gist of that latest Seattle Times story was that Boeing failed to provide a single switch for turning off the MCAS–that when you turned off the MCAS you also turned off the powered system that can give motor assist to the mechanism that adjusts the rear trim (and that is controlled by buttons on the control yoke). With the powered system off the pilots control trim via a large wheel connected to a cable that requires many turns to accomplish the same thing. This is why the Ethiopian pilots kept turning the MCAS back on. They were unable to regain level flight via the manual system.
At least that’s how I read it–happy to be corrected. Obviously the design of the MCAS system, not the plane itself, was disastrously bad and Boeing deserves all the lawsuits they get for the mistake and for not confessing it fully as their chief executive sought to cya. He should be fired for his response to the crisis if not for the engineering flaw.
I wonder why criminal charges are not being considered.
The MAX has six different functions that can operated the electric stab trim motor. Only one of them has an “off” switch – the autopilot – and we need that for other reasons.
The Runaway Stab Trim procedure is completely agnostic as to the source of the unwanted trim input. We do not have separate off switches for the different sources, and MCAS is no different in this respect.
The Ethiopian Airlines report says the pilots did run the Runaway Stab Trim checklist, contradicting your repeated claim this would have fixed everything.
737 Pilot has not stated that only running that checklist would magically fix everything. He has always said that they also needed to reduce airspeed, with his only assertion in this regard being that more or different training and experience may have resulted in the coming to this conclusion on their own, aside from the procedure in the manual.
No, he has specifically said (on 4/30) that running either procedure probably have saved the planes but he has weighed heavily on the runaway stabilizer procedure.
He argued upon occasion that the pilots should have used Airspeed Unreliable, but if you look at the Ethiopian Air report, the pilots were already having trim problems before they noticed the differences in the airspeeds. There was hardly enough time to diagnose and implement a response to one problem, let alone two.
The overspeed clacker also came on after they were trying to correct the trim, and per the long extract from the report above, it was one clacker, the one on the side getting the higher airspeed reading from the disagreeing monitors. The report indicates the AOA values were deviating, but it does not say the airspeed values were disagreeing except when it discusses the right clacker starting to make noise first, so it isn’t clear from the report (it would be in the data recorder) when the airspeed monitors started disagreeing.
Why is 737 pilot so adamant that we must listen to his argument? I feel so incensed that you can’t take a minute to think abut all that’s been said that VERY CLEARLY highlights you were wrong. It’s sad when someone cannot re-assess their position, how do we trust we can learn from you if you’re showing signs you don’t know how to?
You have this backwards. I present information in a later comment from the Ethiopian Air official investigation of the crash. It definitively disproves a central claim “737 Pilot” has made repeatedly, not just in this thread but in previous posts, which is if the pilots had simply engaged in standard “runaway stabilizer” procedures, all would have been fine. This is a counterfactual argument. It is refuted by the fact that the Ethiopian Air pilots DID PRECiSELY THAT. Both the findings of the report and the time stamped summary of the flight recordings show that they did follow the runaway stabilizer checklist.
“737 Pilot” claims to have reviewed this information and made arguments from it, yet chose to hide information that flatly contradicts his central claim. I deal with another (see Synoia) where he makes another misleading argument that is contradicted by the Ethopian Air official report.
And you tell me we should trust him, because he’s an expert and sounds convincing?
Yves, from my comments above, reading the document, I’d say that the argument is a bit weaker.
Facts (from the document):
– the crew DID engage in RS procedure as per the manual (page 30 of the repot)
– the crew COULD override MCAS via manual electronic trim (shown in data doing so) (page 26 of the report, and page 10 of the report)
– the crew DID NOT sufficiently trimm electronically (page 10 of the report).
– the crew could NOT sufficiently trim manually post CUTOUT (page 11)
– the aircraft was substantially overspeding, with overspeed warning for at least three minute before the crash (page 11 of the report)
– the aircraft was NOT losing altitude until about 40 seconds before the crash (page 27 of the report)
– the crew could not manually trim due to high airspeed
– the electronic trim might be slower to respond than the crew expected,possibly due to high airspeed – or for some other reason the crew did not trim electronically entirely
– crew was overwhelmed and fixating to one problem (pitch) ignoring the speed [this is a well known human behaviour under stress – fixation on one problem ignoring the environment], which compounded the pitch problem.
Amedmnet to one item:
– the aircraft was NOT losing altitude until about 40 seconds before the crash (page 27 of the report)
This is actually harded to say, because the altitude is pressure-indicated, so calibrated to sea-level. Over-the-ground altitude could vary, and it seems that the terrain was slightly going up (the crash pressure altitude is higher than the take-off one).
What could be said is that until about 40 seconds before the crash there flight was not expriencing dramatic drops in altitude (it seems to climb up till about 43:15, then level for 15 seconds, and then crash).
A propos of nothing at all, but there’s a rather a lot of money at stake, here. Not merely the lawsuits, but reputational damage to Boeing, and in particular to sales of their cash cow, the 737.
So blaming the pilots is something a lot of powerful players have every incentive to do. I wish I knew who Boeing’s PR firm is, so I can get a sense of what kind of methods they use. Does anybody know?
One could of course also argue that someone shorting Boeing stock would use a similar approach….
Help me. This is making shit up.
Shorts don’t use comment sections of blogs. Please. This isn’t a stock jockey site and we don’t make investment recommendations.
We were in the middle of much more visible fights (Lehman, the monolines) with visible shorts in the days when hte econoblogosphere was vastly more influential than it was now. And we didn’t have people coming in comments to try to reeducate us or our reader then.
I guess I missed this whole argument last night but saying that the pilots may have had some responsibility (which is what I said above) is not saying that they are to blame or Boeing is off the hook. As anyone who has ever been in a car accident knows there’s often some negligence on all sides but only one party gets the ticket. For all the discussion and speculation one thing is clear–no MCAS, no crashes. And any “pilot defense” Boeing cares to trot out is negated by their failure to fully inform customers about the MCAS, how it works and why it was there in the first place.
Am I wrong to state that turning off the autopilot on an NG airplane disables that iteration of MCAS? That pulling back on the column disables auto trim inputs on a NG? And disabling a NG’s autopilot allows continued electric/hydraulic trim actuation?
And all of the above statements DO NOT APPLY to the Max?
And the differences were not embraced. They were hidden.
Different between NG and Max?
Was anyone else’s grim humor tickled by this line, ” then tried to downplay its importance when it came out in bits and pieces“?
Thanks for noticing :-)
Boeing took an aerodynamically stable plane and screwed it up. Then they thought they could patch it up with a new layer of programming. If my car is out of line and yawing off the road, I get the alignment fixed. That’s not a problem software can fix.
Until the MAX, the 737 has been a proven airframe with the kinks ironed out. I wouldn’t hesitate to fly on one. However, I won’t knowingly board a MAX, ever.
The MCAS debacle taints the whole certification process for the MAX. It now should be subjected to the same process a clean sheet design would be. And, of course, the FAA process needs examination with a fine tooth comb.
Auto alignement is the best comparison yet. How about a car with lane hold, AKA autopilot engaged?
Only if the Tesla Autopilot kept fighting against the human driver and wouldn’t disengage, forcibly steering into a concrete barrier.
And that has already happened and killed at least one person.
Millions of cars pull to the left or right and yet people still drive them. Actual pilots have written here and elsewhere that the plane is not fatally flawed aerodynamically–just the MCAS which is a system that doesn’t even turn on apparently unless the sensors indicate a near stall condition.
Darius, I’m not sure that you are right about the aerodynamically stable plane. It is a kludge, an engineering term for something that has many new design systems layered over the old ones so that it no longer functions in its primary purpose – to be aerodynamic and act as a flying machine.
The airframe is based on the 707 design from 1952, that was then expanded to accommodate more passengers, without lengthening the wings. Bigger more powerful engines were needed that didn’t fit under the wing so they were pushed farther and farther forward, because the landing gear is not tall enough (not any easy fix btw). So eventually this kludge-ie thinking got a wing that got no lift from its engines.
One pilot called it a big fat lazy pig.
So you have a plane that is not aerodynamic, but gets off the ground with just thrust.
I live in Seattle and come from a Boeing family and my dad was an engineer. We live in the flight path of Boeing field and after the crashes, you could see Boeing testing the new software on the 737-Kludge, but what you notice is that the nose on the plane is tilted at a higher angle, because the pilots have to use full thrust to get the fat lazy pig to altitude.
This is where the MCAS (with its one AOA sensor) makes it a 737-Kludge a death-trap.
Nose up at full speed on take-off is a SOP, which is then gonna trigger an Angle of Attack alert. (Same MCAS-controlled event happened on the Ethiopian plane the day before, but a ride along pilot got the crew out of it) The MCAS comes on and is activated every ten seconds that forces the nose down, and at a much steeper angle than what Boeing told the FAA.
This plane is a modern-era Kludge, so many contradicting systems that CRASHING, not flying, is the inevitable output of its operation.
The angle of attack alerts in the two fatal crashes were triggered by that single defective sensor. In other words the MCAS turned on when it wasn’t supposed to turn on.
And the 707 was a four engine plane. The 737 first appeared in 1967. I think you have this story scrambled a bit.
> And the 707 was a four engine plane. The 737 first appeared in 1967. I think you have this story scrambled a bit.
I have seen other comparisons between the 737 and the 707; they are said to be similar because of the shape of the nose (and the narrow, noisy cockpit). The similarly has nothing to do with the number of engines.
All modern airliners are based on that 707 layout which I believe Boeing may have invented for their military planes. It has no relevance to the recent accident except insofar as the engines are in nacelles below the wing (and therefore that engine placement problem). But I believe that latter was more about the short 737 landing gear.
> It has no relevance to the recent accident
Didn’t say it did. My comment was solely directed to the issue of similarity.
In the video a disc spins when the MCAS system activates. Could the MCAS system be disabled by preventing the disc from spinning?
The disc is the manual control. See my comment above. Pilots can fly the plane manually by reaching over to turn the disc to fine tune the trim while in flight. “737 Pilot” commenter says he does this constantly while flying plane manually. Normally however these planes cruise on autopilot and a computer controls trim while the wheel turns in response.
I’m not a pilot btw but do have some experience with MS Flight Simulator where you can see this happening or there are lots of videos on Youtube showing pilots in action as various airplanes take off and land.
So if the disc is prevented from spinning would that stop the MCAS system?
Also, the pilots were not informed about the MCAS system at first. Would they be in danger of being in the middle of making a manual adjustment to trim when the MCAS takes over the disc? Could the pilot’s hand be damaged by the motorized disc?
There were switches to turn off the MCAS and they worked but they also turned off the motorized control of the trim forcing the pilots to turn that wheel many times before crashing. And air pressure from the high speed dive may have made the turning too hard to do manually. Seattle Times series explains all this better than I can.
Actually, according to this video, this is the last resort to stopping a runaway stabilizer:
The spinning disc is the manual trim wheel. It has a stowable handle that the pilot can use to manually trim the stabilizer if the electric trim is not available. The system is designed such that if that wheel does not turn, the stabilizer does not move. There is a clutch in the system between the electric trim motor and the rest of the system, and this clutch mechanism favors the manual trim wheel. If other methods to cutout the electric trim failed, a pilot could simply place his foot firmly on the wheel and stop its motion, thus stopping the stab from moving further. (The checklist actually says to grasp the wheel with your hand, but the foot works much better).
All that being said, I know of no case in which the above method actually needed to be used with a real malfunction.
deregulation and self regulation are fatally bad ideas
As a rule governments are prone to two errors. They claim authority over all sorts of matters that should be none of their business. And they cock up their duties in matters for which they are, or should be, responsible.
I’m confident that the comment threads on NC will be full of people who manage to believe that (i) the US government has utterly mucked up its regulation of aircraft construction, while also believing that (ii) the answer to any disagreeable condition in American life is to give more power to the federal government.
i’m confident you’re wrong. sure on some issues like climate change, it’s not like the fossil fuel industry is going to lead us away from the abyss, now is it? but i doubt you could get much support her for increasing the power of government to spy on us, or censor the internet, or destroy the first amendment…
Gov used to regulate a lot more, particularly regarding safety, as well in enforcing existing law. The Boeing fiasco is the result of the continuous pressure from industry for less regulation. Other examples are less inspection of meat, dumping into rivers, capping of old oil wells, reviewing rich people’s tax filings, enforcing voting rights act, and on and on. Plus the laws themselves are nowadays written by the industries that were formerly regulated, not least finance, pharma etc.
Some think gov should be limited to a strong military. Our country was better in most ways before the post Vietnam era when we were vigorously enforcing existing regs and antitrust, and willing to toss white collar law breakers in jail.
How does your worldview square with the neoliberal playbook, which is:
1) Take something the govt does well
2) Underfund it till it breaks
3) Complain that govt is broken
4) Sell out to cronies (i.e., privatize)
It was Boeing that certified its own plane. How is that a failure of govt? Boeing literally slapped itself in the face. If the FAA had been responsible for inspecting and certifying everything you could go ahead and argue that it failed, but that’s not what happened.
Actually, there is less of a contradiction then you suggest; government interference is O.K. for those without power but bad for those with power.
Seems like this is another mess that Congress created, granted they ‘pushed’ to deregulate the FAA by Boeing and others, but Congress could have said no. They didn’t, some because they are so tied to their ideology that they couldn’t make a good choice.but we also shouldn’t let the airlines off the hook either
Tied to ideology…
Think donations that usefully encourage and reward anti reg ideology.
There was a period of time in the 2000’s when the FAA budget was not set and the FAA could not hire inspectors. I would suggest looking into that aspect and if it led to a staffing problem that favored self-inspection.
This certainly gives new meaning to the IBG/YBG ethos of American corporate corruption. The people who make the decisions never fly on these planes…1st class is trumped by private corporate jets. Maybe it’s time for corporate passenger hostages.
This is no longer a matter of “real pilots”. This is a matter of someone who has access to the actual 737 Max documentation and is referring to that, or an article that can be trusted to describe it well, or alternatively, input from insiders at Boeing. There are hardly any “real pilots” have had experience with this malfunction.
I don’t have the link but I saw a story about a flight out of Phoenix awhile back where the pilots did have the problem, turned off the MCAS, and continued to their destination. They were reprimanded by the airline for not turning back.
And that Indonesian plane had the problem the day before and a pilot in the jump seat helped them to turn it off and not crash. So obviously the plane can fly with no MCAS at all. It was included so Boeiing could make the false claim that the Max was the same airplane pilot wise as previous versions.
Clearly Boeing is the one at fault here and will doubtless lose any lawsuits.
Grr, this was an “approve and reply” to a comment in the backstage and the underlying comment didn’t come through. Apologies to the commentor. And now because someone replied, I can’t fix this manually (delete my comment, go find where the comment that triggered the “real pilots” response went, and then repost my comment). This is another glitch we’ve been having that I have to see if we can fix.
I thought that a recent Harpers article I read yesterday laid out a reasonable explanation for Boeing self-destruction by linking it to the MIC. Just like you cannot mix oil and water, you cannot effectively merge a defense contractor with a civilian manufacturer and expect the hybrid to produce the best of both worlds. The data are in – Boing got the worst of both worlds.
That was a great article. And part of the point is that it’s not just Boeing. The corrupt Defense Department culture seems to have infected all of these companies with Wall Street, of course, always lurking in the background.
Re: “The only upside to this sordid affair is that it shows why deregulation and self regulation are fatally bad ideas. But what a horrific way to do that.”
I wonder how many unnecessary premature deaths have occurred in less noticeable small quantities, dispersed throughout the population, due to deregulation. Surely the Great Financial Crisis directly or indirectly caused orders of magnitude more premature fatalities. But they weren’t as spectacular.
Hopefully this spectacular consequence of deregulation will function as a warning indicator that is noticed and appropriately acted on by the people who are piloting our civilization.
Thing is, it’s not just the 737 Max that’s crap. Workers assembling the 787 in the Boeing plant in North Charleston, SC were polled on the question “Would you fly on the plane you’re building”? 15 were asked, 10 say they wouldn’t fly on it. Watch this documentary; it’s worth your time and might save your life:
Boeing is apparently scrapping safety for profit. Meantime the FAA is way off-track on their mission…it should be “safety first” but their bureaucratic brains have that scrambled by overriding safety with promotion of the aircraft industry, which shouldn’t even be in the mix!
boeing is probably the worst. i hope.
Somebody has to lead the way…
But I would say that finance is not just leading but is the pied piper luring others to the pursuit of profits above all.
Probably not. Others have likely to do the same
The biggest single root cause, other than the decision to make a derivative 737 Max, was per the first Seattle Times article, that initially the system only had .5 degree of control authority, which would not have been enough to create a catastrophic scenario, and thus the system was categorized as non critical and therefore not subject to scrutiny, but then in flight test they realized it needed 2 degrees of control authority to function properly, and then never revisited the characterization of the MCAS as non critical. If they had they would have categorized it as critical and it would have been subject to a whole bunch of scrutiny.
If there are emails that show Boeing actively ignored revisiting the classification of the system after the flight test changes, they will be criminally liable, if it was somehow just an oversight, even though they followed proper procedure I don’t believe there will be any criminal liability despite the non-criminal negligence that would typically result in a manslaughter charge of an individual.
> The biggest single root cause, other than the decision to make a derivative 737 Max, was per the first Seattle Times article, that initially the system only had .5 degree of control authority, which would not have been enough to create a catastrophic scenario, and thus the system was categorized as non critical and therefore not subject to scrutiny, but then in flight test they realized it needed 2 degrees of control authority to function properly, and then never revisited the characterization of the MCAS as non critical. If they had they would have categorized it as critical and it would have been subject to a whole bunch of scrutiny.
Thanks for reminding me of this.
Well, after that lead in, it seems that this may be a futile attempt, but I’ll give it a go anyway.
Yes, I saw this episode of the Australian 60 Minutes. There was a lot of good material in there, but there were also some things that were highly skewed, probably for dramatic effect – a tendency that is not unknown in the popular media.
My first observation is that this recreation was preformed in a 737NG simulator and not a MAX sim. There is no way to truly simulate a MCAS malfunction in an 737NG because it doesn’t even have MCAS. Thus, there was already a bit of theater in play here.
Next, the pilot was letting MCAS run the full 10 seconds before he intervened. This is exactly what one does not do, so they were actually demonstrating the wrong technique.
Let me give you an analogy that might be helpful. Let’s say you are driving a car that has one of those steering “assist” systems that attempts to keep the car from drifting out of its lane. If that system malfunctioned and started steering the car toward one side or the other, are you going just let it drive you across three lanes of traffic or are you going to try to intervene? I suspect you would take immediate control and put the car back in it’s lane. After 5 seconds, the malfunctioning steering system tries to take you back across those same lanes. Are you going to let it, or are you going to stop it? Won’t you keep doing that for as long as it takes to get the car safely off the road?
As soon as a pilot puts in any amount of trim with his yoke switch, MCAS stops right there – it does not keep going. The pilot can then take out any trim that MCAS put in. After 5 seconds of no input, MCAS will try again, but again, the pilot can stop MCAS in its tracks – none of this letting the trim run for 10 continuous seconds. (BTW, this functional description refers to the original software, not the replacement)
The very first documented case of an MCAS malfunction was Lion Air 610 the day before the first MCAS related accident. That crew responded to the malfunction in just the way I described – and they did not know MCAS even existed. This crew recovered the aircraft safely.
Final observation on the reenactment. The whole “flying the aircraft while paging through a thick manual” was a total, complete, and absolute farce. I’m sorry, there is just no other way to describe this particular scene. First, when performing any checklist (normal or otherwise) on a two-person commercial aircraft, the rule is: one pilot flies, one pilot runs the checklist. If a pilot is trying to fly the aircraft and page through the manual, that person is doing it wrong. Second, there were two non-normal procedures in play – Airspeed Unreliable and Runaway Stab Trim. Depending on the airline, these procedures are done either from memory or by reference to a Quick Reference Card (QRC). There are no other substitutes. The pilot in the reenactment did neither, but rather let go of the aircraft controls and picked up his Quick Reference Manual (QRH, which is something different from the QRC). This scene was profoundly misleading.
Bottom line, the aircraft was flyable. Why the aircrews were not properly trained to do so is another matter entirely.
I know you want to hold Boeing accountable, as do I. I would suggest, however, that promulgating incorrect technical information is not the strongest hand to be played. It would almost be like someone coming on this site to debate MMT and them opening with some obviously false misconception about economics and/or monetary policies – it would be pretty much downhill from there.
So let me say again, there were a number of links in the chain of causation of these accidents – what happened at Boeing is only one of them. I would even go far to suggest that some of the root causes underlying Boeing’s negligence are at work up and down the entire causal chain. The 737 MAX will eventually be fixed or buried, but all those other broken links will still be there if folks chose to ignore them.
The biggest problem is that Boeing tried to fix airworthiness of an airplane with software code. In doing this, they bet the lives of people on a bunch of if…then clauses. Anyone who has written code for a living and has an ounce of ethics in their character, would tell you what a terrible idea this is.
Where do we go from here? The usual – mistakes were made, the execs retire with full severence, the lawsuits get settled, the company engineers add another series of if…then clauses to the code, the airlines get a break on the price and business continues as usual. (See how you cannot get an Airbus delivered in the next 60 months if you ordered one today ).
The underlying airworthiness problems of the plane that required the MCAS do not get fixed in any case.
It is. FAA’s budget gets cut.
It wasn’t the airworthiness of the plane but rather the handling characteristics. The whole saga has been covered in detail by non paywalled Seattle Times articles so no point in going back over it.
But modern airplanes are run by computer code and Airbus more than Boeing since Airbus is completely fly by wire. The point is not that it was code but that it was badly designed code. The code itself is inevitable unless you expect the pilot to sit there and manually fly the plane for thousands of miles.
Read Ralph Nader’s letter. He phrased it better than me. The configuration of the engines on top of the wing makes the plane stall-prone. This directly impacts the airworthiness of the frame and trying to solve the problem by software code is a terrible technical decision.
It is not software per se that is the problem. It is implementing LOGIC in the code to prevent catastrophic failure.
From the official report on the Ethiopian Airlines crash:
So it appears they did run the correct procedure.
I had missed a New York Times article, In Test of Boeing Jet, Pilots Had 40 Seconds to Fix Error, (this must be the story that Lambert said he saw that went tin the details of how the MCAS would kick in repeatedly for 10 second intervals). It reviewed the Lion Air crash in detail, with input from pilots using a flight simulator.
It also points out that the Lion Air pilots indeed resorted to the manual to try to fix the problem, as did the Ethiopian Air pilots. And I disagree with your spin on the 60 Minutes video clip. When the pilot was flipping through the manual, it was obvious he was not simulating flying at that point but demonstrating how long it would take to find the section, which he said was only 2 seconds which frankly I found to be optimistic. And your two pilot point is irrelevant if the pilots thought that further corrective action at that point depended on getting input from the manual.
It is not a reasonable viewpoint to claim, even if your “they didn’t try the right remedy” point is correct, that the pilots were therefore to blame if they only had 40 seconds, or charitably, a minute in which to initiate that process. Diagnosis is not instantaneous.
A Seattle Times article also describes at length how Boeing misinformed the FAA as to how aggressive the MCAS intervention would be when the system kicked in:
The articles you cite contain a mixture of both good information and bad, and I know it can be hard for a layperson to separate out the two. This is not a phenomenon that occurs just in reporting about aviation. I bet you can recall more than a few times that a news article misreported on an issue that you were already very informed about, and it just drove you crazy that the reporter didn’t get the facts straight. You are also probably aware that some reporters will overly dramatized some information to make for more compelling storytelling. Same here.
Most of the references you cite for Lion Air actually occurred the day before on the same aircraft with the same problem. This crew landed safely because they were able to effectively counter the MCAS input until they cutout the trim.
We actually have limited knowledge as to what the crew of the Lion Air accident flight did because at the time of the preliminary report the Cockpit Voice Recorder (CVR) had not been located. It has now been found, but the Indonesian investigation authority has said it will not release an official transcript until it issues its final report sometime this fall.
Even so, we can still infer certain actions were or were not taken by looking at the Flight Data Recorder (FDR) traces that have been released for the accident flight. There is no sign on the FDR output that the Lion Air crew performed actions that were consistent with either of the non-normal procedures that were in play. We certainly have no evidence that anyone was looking through a manual.
We do know that the Lion Air 610 crew flew for about five minutes while MCAS was active. If you look at the FDR output, you can see where MCAS put in nose down trim followed by one of the pilots (likely the Captain) taking it back out again. There is no limit for the amount of time this back and forth could have continued, but for some reason the pilots ultimately did not fight back long enough.
I don’t know how much weight you wish to give to this next bit, but there have been some unofficial leaks of the crew conversation from the CVR. This (again unofficial) release indicates that the Captain was initially flying the aircraft, but then decided to turn the aircraft over to the First Officer shortly before the final dive. There is a definite inflection point on the FDR output that shows a break from a point where the stab trim was being managed effectively to a condition where the MCAS inputs were not being fully countered. This inflection would be consistent with a change of aircraft control from an experience pilot who knew how to maintain control to an inexperienced one who didn’t.
I don’t care how many sources say otherwise, the Ethiopian crew did not correctly perform either the Airspeed Unreliable or the Runaway Stab Trim checklists. We know this because we have enough of the FDR and CVR outputs to make this determination. Yes, they cutout the electric stab trim. I will say again, this is Step 5 of the procedure, and it is very important to do the other steps first. The fact that they did not neutralize the trim before using the cutout switches is what ultimately doomed this flight.
As far as all the various references as to how much time either crew had to figure out this problem, let me say again, in the strongest possible way, that the crew had as much time as they needed – as long as someone was flying the plane.
As long as one pilot was actively trimming the aircraft against the MCAS input, they could have literally flown until they exhausted the fuel on board – a few hours at least. I think we can assume in that amount of time they would have figured out a way to properly isolate the trim system – even if one of them had to get on the radio and get instructions from someone on the ground.
No disagreement on the royal f$#kup between Boeing and the FAA as to the MCAS design parameters, or Boeing’s general tendency to take the cheap way out. The FAA oversight role has been eroded tremendously. We should also be very troubled that pretty much every airline on this planet has conspired with just about every manufacturer to water down pilot training requirements. We have an entire generation pilots coming through the pipeline who are not being given the training or the experience to deal with anything but the most mundane emergencies.
In the end, MCAS will be fixed. I have my doubts about the rest of it.
Correction to the above. I was re-reading the “leaked” CVR transcript and it does make reference to the pilots looking through the the Quick Reference Handbook (QRH) on the Lion Air accident flight. However, as I previously mentioned, the procedure in a case like this is for one pilot to fly and one pilot to look through the QRH. At no time does one pilot fly AND read.
If this transcript is to be taken at face value, it appears that the Captain was flying the aircraft and the First Officer was trying to find an appropriate procedure to follow. After a few minutes, the Captain decided to look himself and asked the First Officer to fly. It was shortly after this change in control that the aircraft started its fatal descent. The implication is that while the Captain was able to maintain control, the First Officer wasn’t. The Captain was likely “heads down” in the book, and by the time he sensed that the First Officer had lost control, it was too late to do anything about it.
Link to Reuter’s article
This is now OBVIOUS bad faith argumentation. I gave an exact quote from the official initial report on the ETHIOPIAN Airlines crash which said they ran the runaway stabilizer checklist and this didn’t resolve the problem
You utterly straw man the official report, to which I provided a link, by falsely depicting it as a “press story” and having “good and bad information”. This is an utter crock as a claim, as anyone who clicks through on that link can see.
You instead go on a diversion by ignoring the evidence from the Ethiopian Air official report, which directly disproves your repeated claim that running normal runaway stabilizer procedures would have save the 737 Maxes from crashing.
You instead spill tons of pixels on the Lion Air crash, which is a clear effort to divert attention from how devastating the Ethiopian Airlines finding is to your now-very-well-established line of argument.
You further straw man the simulator presentation in the 60 Minutes video, and basically ignore/talk over what I wrote. No one would see that clip as saying the pilots stopped flying the plane. The entire point of pulling out the manual was to demonstrate that it took two seconds of the only five second between when MCAS nose-down attempts, as in it ate into very scarce time.
As I clearly stated, the pilot who was flying was waiting for input from the manual the other pilot was reading!!! He was waiting for guidance. He wasn’t going to stop flying the plane, but he was not going to make a new intervention while waiting to get input.
I think we can establish a two things from this discussion:
737 Pilot and Yves are in full agreement that the FAA, Boeing, and the airlines utterly failed in many points along the chain in providing a safe plane.
737 Pilot and Yves are in significant disagreement regarding the ability of the pilots in these specific incidents to reestablish control of their aircraft.
Given this, I’m however having trouble figuring out what nefarious motives 737 Pilot would have in consistently agreeing with Yves on the first part but disagreeing on the second part while continuing to provide patient expert insight as to the technical and operational details of flying a 737.
I suggest you reread what “737 Pilot” said with much more care than you have. You have already misread and them mischaracterized what he said by inaccurately claiming that he only presented fact, when he repeatedly gave his personal opinion, with no qualifiers like “likely” or “I believe” as fact.
So quite honestly, you have shown yourself to be not careful and exacting in reading text, at least as far as when a presumed expert like “737 Pilot” holds court.
“737 Pilot” very clearly and repeatedly has said the pilots, executing standard procedures, could have prevented the crashes, That is tantamount to blaming the pilots for the crashes (or at best the airlines for not requiring more seasoning). By contrast, as you can see immediately above, he depicts what Boeing and the FAA did as “fuckups”. He doesn’t spend anywhere near as much time on Boeing’s culpability, and “fuckup” is very ambiguous in terms of how much blame for the crash he is assigning to them. But is comes off as far less definitive than what he pins on the pilots.
Well, he has a point. Moving the electric trim switch seems to disable MCAS. Continuing to disable MCAS would likely have saved the plane.
But Boeing didn’t see fit to inform the pilots that the system control parameters had changed, how they changed, or why.
The 2 different checklists (NG & Max) might have been the same, but control functions had been changed. And not documented.
It all comes down to a simple question for me – why is the MCAS even necessary. It comes down to a fundamental problem with the plane design, which gets patched by a mechanical solution, which creates another problem, which is then patched by a software solution.
Do you not see ho brittle this makes the whole system?
The pilots may have done everything possible or they may have screwed up.
But the system design does not set them up for success.
The MCAS was necessary to satisfy certain control feel characteristics required of commercial aircraft when approaching a stall. This is not a new requirement, and the 737 is not the only aircraft to use some type of control augmentation system to meet this requirement.
The method that Boeing chose to meet this requirement was not particularly fault tolerant, and a single-point failure in the AOA system created a novel and hazardous condition. I have my own speculations as to why Boeing chose this particular solution and why they didn’t properly foresee the potential failure modes (and yes, it was driven by cost and time pressures), but that gets into some very technical discussions that are beyond the scope of this forum.
737 Pilot: Thank you for your calm and clear comments. They are very valuable and highly appreciated.
I think that many pilots and experts who say the accident flight pilots should have been able to handle the MCAS failure would agree that they are not blaming the pilots per se. As modern accident investigations have shown, pilots are only as good as the system that produces them and supports them. And as you say, that system, despite improving in some ways, has been degrading recently in others.
As this episode shows, ensuring safety trumps dollar signs is a constant struggle. Let’s hope aviation gets back to the right side of this equation.
Let me start by saying I apologize for impuning your motives, I should know better.
My argument with you was triggered by what I saw as an effort, in part, to blame the deceased, and, in part, to defend what I see as a badly flawed product.
Boeing’s reaction, understandable, but sad, has been to defend that product.
I am sorry for misinterpreting your calm, dispassionate manner for bias in favor of a machine, and a corporation, and disregard for your fellow pilots, crew and passengers.
After consideration, I understand that is not your intent.
There’s a lot of contention going on here, so let me jump in and thank you for the clarification. I am not a pilot but I thought it was very strange that the pilot in the video was going through a paper manual. On the other hand the claim is that the procedure is not included on the Quick Reference Card. It was stated in the video that the procedure is only in the paper manual.
There are a number of key non-normal procedures for which the first steps are considered so important that they are executed either from memory or by reference to a laminated Quick Reference Card (QRC) depending on the airline. Once the initial steps are complete, then the pilots will then refer to the paper (or electronic) manual for any additional actions. Every procedure that is on the QRC is also in the full manual.
The Runaway Stab Trim procedure is one of these QRC/memory items, but it is technically not incorrect to say it is in the paper manual. It is located in both places.
I went back and re-watched the segment you reference. It was actual the narrator/reporter, and not the pilot, who said the procedure could only be found in the paper manual. This could have been a simple misunderstanding on her part.
This is another bad faith comment. You ignored the Ethiopian Air official report finding, that the recordings show the pilots did run the runaway stabilizer procedures. You keep imputing a failure on their behalf that is dispriven by an authoritative report. You are now doubling down on a line of argument that has been shown to be false. Your failure to even acknowledge the Ethiopian Air finding, which was at the very top of a comment you replied to with two long comment, and are instead repeating a now-debunked line of argument is not on.
So, that scene was totally make believe, as in they programed the 737NG simulator to display the behavior of a 737MAX in order to make a TV show?
So the pilots interviewed, allowed themselves to be part of a charade, letting the MCAS that didn’t exist on the simulator they were sitting in, run in a manner that was easily remedied, and thus convinced us dullards that it wasn’t pilot error and lack of training, but a fundamentally flawed MCAS system that led to these ‘accidents’?
These pilots helped perpetrate this hoax because why, they wanted to see themselves on TV?
I suppose it was on 60 Minutes Australia because only they could get away with slandering the reputation of one of the largest aircraft manufacturers in the world?
Well, I can’t read minds, so I can’t really say what motivations were at work, but let me ask you this:
Can you think of any other case in the history of TV where someone was paid/encouraged to say things that weren’t quite true in order to promote a particular story or get better ratings? Even on 60 Minutes?
The sad thing is that they could have done a proper and informative reenactment even with the 737NG sim, but for some reason they chose otherwise.
Bottom line, the 737 MAX (and all the decisions made to get it into the air) has led to bits and pieces of 300+ people to be scattered across the earth & ocean.
The aircraft was flyable under certain conditions, but seemingly only by extraordinary pilots. That’s a no-go for commercial aircraft.
(As a state official in a younger life, I had to fly regularly in a twin-engine Beechcraft King Air with a state pilot to a small airstrips in eastern Nevada for scheduled meetings. The pilot (a former AF fighter pilot) always checked the plane thoroughly then checked the weather at our destination. Several times he cancelled these flights due to cold weather. State business was not more important than risking our lives, he said. I thank him every time I seem him in retirement.)
The 737 Max crashes were directly caused by deregulation and the drive to increase C-suite income and shareholder value. Work conditions at Boeing are toxic. Problems that would slow down production ignored. Laws and regulations disregarded. This is universal in Western multi-national corporations from Bayer to Volkswagen. When brought to court, juries will have a field day awarding billion-dollar judgments as they have in cases of non-Hodgkin’s lymphoma linked to glyphosate. The fix is restoration of the rule of law and regulation. But both the US and UK governments have been transformed in bungling incompetency. Ruled solely by conflicting corporate bought influence. Besides restarting the Cold War with Russia, the USA is threatening a hot war with Iran, and has started an economic war with China. The only conclusion from these articles is that Boeing and the FAA are unlikely to design a software fix, include hardware redundancy, flight testing, and pilot training that would make the 737 Max safe to fly again. The only sure thing is to stay bunkered up at home but that is not viable. Life has become a hell of lot more dangerous. Denial is about the only way to coup except that doesn’t work anymore if one is halfway connected to reality.
I don’t know how deep the hole Boeing is in but the hole made by the Ethiopian Airlines 737 Max crash was 32 feet deep.
Indeed, I too am not a fan of what Boeing has done with the 737 MAX. And yet…
Unlike most American corporations that are increasingly into pure rent-seeking behavior, Boeing makes very sophisticated pieces of machinery that must operate flawlessly day in and day out and where every single failure is put under a microscope. Most of Boeing’s planes fly flawlessly. The difficulty of this accomplishment is shown by how hard the Chinese, Japanese, Russians, etc. have had trying to match this technology. Cars, satellites, atomic bombs.. these are children’s toys next to modern jet airliners. Seriously: North Korea can make atom bombs. A reliable commercial airliner? In your dreams! Castigate Boeing all you like, but can you imagine a plane designed by Goldman Sachs? (hahahah!). Or Microsoft? (“Error cannot access drivers” KAPOW!)
Of course, Boeing is outsourcing key technologies, and sooner or later the Chinese/Japanese or some other country with a real industrial policy will push them aside – but by then the insiders will have cashed out and so it’s all good.
Pilot error … ummm…
You’ll get that flying air frames with issues ….
If it is true, as 737 Pilot contends and contrary to some press reports, that activating the electronic trim control turns off the MCAS, then it does sound to me like the pilots could recover from an MCAS malfunction, assuming that was the only malfunction taking place. I am not sure what happened in the Ethiopian Airlines crash.
Perhaps NC might want to allow 737 Pilot to write a rebuttal, if he is willing, explaining what he believes is true and false in the reporting on the 737 MAX, but then also invite a response from other experts or the journalists being criticized.
A friend just pointed me to this thread. A lot of interesting information. May I offer an insight?
I’ve worked around aviation on and off for a few years. I’m always fascinated by the accident investigation process. Quite often in the early stages there is a lot of misinformation flying around, and the findings in the final report are frequently very different from the initial impressions. There are a lot of statements being made based on incomplete information, so my recommendation would be to let the investigators do their job and not jumped to too many conclusions. It may take longer than everyone likes, but they usually get to the right answers.
Yes, this is good to remember.
The “Anon” above is not the same commenter as the one earlier, Anon May 15, 2019 at 5:26 pm
Although I’m in agreement, I don’t think the incident investigation is going to change the impact of the Seattle Times reporting on events at corporate Boeing.
One good that can come out of this sad story is that perhaps people will fly less.