Go Back  PPRuNe Forums > Flight Deck Forums > Rumours & News
Reload this Page >

Ethiopian airliner down in Africa

Rumours & News Reporting Points that may affect our jobs or lives as professional pilots. Also, items that may be of interest to professional pilots.

Ethiopian airliner down in Africa

Old 28th Apr 2019, 06:44
  #4481 (permalink)  
 
Join Date: Oct 2017
Location: Tent
Posts: 333
Originally Posted by Cows getting bigger View Post
Supplemental - would you have expected the pilots to 'deal with it' if you knew it ran away at 2.5 rather than 0.6 deg?
Yes very good fact!
Bend alot is online now  
Old 28th Apr 2019, 08:07
  #4482 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 57
Posts: 378
737 Driver

What MCAS will then attempt to do is run the trim nose down for 9 continuous seconds and spin the trim wheel about 37 times. I’ve asked this before, and I’ll ask this again: Starting from a stabilized, in-trim platform, exactly how long should a qualified 737 type-certified Captain who is hand-flying the aircraft let the trim run in one direction before he/she does something about it?
A long time ago I asked a critical question, to which I did not get a satisfactory answer. Perhaps you can indulge me.

On the B737, the flight computers (and the FDR post-crash readout**) know exactly how much nose-down trim has been applied to the horizontal stabiliser, but the pilots do not.

The only way for them to determine this is to look at the sliding scale between the manual trim wheels, or count the number of rotations, or infer from the elevator feel forces.

Stabiliser trim position is not shown on any of the primary flight displays, and AFAIK is not part of any routine instrument scan.

It seems that pilots have to infer runaway trim, rather than a big flashing display that could easily warn them of that fact. How un-ergonomic is this setup, that the most powerful flight control, is the one they have least information about?

Edit: **In much of the post-crash discussion, the question is asked why did the pilots stop trimming nose-up? Those comments were based on the FDR readouts, while the pilots had to infer the position of the horizontal stabiliser trim, relative to column forces and elevator feel, etc.

Edit: Or to put it even more bluntly: The pilots had to make a life-or-death decision about when to hit the trim cutoff switches, when they did not know the actual position of the horizontal stabiliser. Much of the post-crash "hindsight" analysis is based on the FDR readouts, which contain information that the pilots flying the aircraft did not have access to!?

Last edited by GordonR_Cape; 28th Apr 2019 at 08:52.
GordonR_Cape is offline  
Old 28th Apr 2019, 08:30
  #4483 (permalink)  
 
Join Date: Jul 2014
Location: Harbour Master Place
Posts: 572
Originally Posted by GordonR_Cape View Post
737 Driver



A long time ago I asked a critical question, to which I did not get a satisfactory answer. Perhaps you can indulge me.

On the B737, the flight computers (and the FDR post-crash readout) know exactly how much nose-down trim has been applied to the horizontal stabiliser, but the pilots do not.

The only way for them to determine this is to look at the sliding scale between the manual trim wheels, or count the number of rotations, or infer from the elevator feel forces.

Stabiliser trim position is not shown on any of tbe primary flight displays, and AFAIK is not part of any routine instrument scan.

It seems that pilots have to infer runaway trim, rather than a big flashing display that could easily warn them of that fact. How un-ergonomic is this setup, that the most powerful flight control, is the one they have least information about?
I don't think i've ever looked at the trim setting in flight whilst hand flying. Throw in the stick shaker and IAS disagree, it would probably the the last thing that would get my attention too. Even in a runaway trim, just trying to communicate to the person sitting beside me what I thought was going on and what I was going to do would be tough.

Remember the trim is there to reduce workload. The first lesson of flying is to trim trim trim. Smooth small pitch and thrust adjustments & keeping the aircraft in trim significantly frees up the mind to process other details. An out of trim aircraft consumes ALL your cognitive resources, the instant you take you eyes off the attitude the aircraft will move significantly away from where you want it. You have to start putting in large inputs and things feel very out of control. That is why trainees from the Airbus struggle initially, they have to consciously remember to trim, as the aircraft won't sit where they left it.

It would be an interesting experiment in the sim to do a difficult sequence without touching the trim at all from takeoff to landing. I suspect even flying a bog standard circuit would be a very challenging exercise with takeoff trim set, particularly if in the forward region for a F25 takeoff. Perhaps someone with a bit of spare time in the sim could try this and report.
CurtainTwitcher is online now  
Old 28th Apr 2019, 09:10
  #4484 (permalink)  
 
Join Date: Oct 2007
Location: Here
Posts: 621
Originally Posted by 737 Driver View Post
Found them, thanks! The reports were misfiled with the 737NG's for some reason. They appear to be reporting the same anomaly. Report numbers are ACN 1597286 and ACN 159380 for anyone looking for them.
https://asrs.arc.nasa.gov/search/database.html

For example:

Choose 737 NG
then text search
(max OR max8) AND pitch

27 hits

I do not have the expertise to read them and reliably comment.

Also, 5 possibly relevant reports: (PDF is easy to search)
https://www.documentcloud.org/docume...-737-max8.html
https://assets.documentcloud.org/doc...r-737-max8.pdf
Still available as of 28 April 2019

"Contributed by: Cary Aspinwall, The Dallas Morning News"
HTML has some summary material (metadata) not in the pdf.

ASRS Reports for 737 max8
- A 737 Max 8 captain noted problems on takeoff
- An unidentified captain says the Airworthiness Directive does not address the
problem in November 2018.
- An airline captain called the flight manual for the Boeing 737 Max 8 "inadequate and
almost criminally insufficient."
- A co-pilot reported an altitude deviation in November.
- Co-pilot said after engaging autopilot, aircraft pitched nose down.
- Co-pilot reported that aircraft pitched nose down on departure.
- A Boeing 737 Max 8 goes nose down suddenly during takeoff, pilot reports incident.

Last edited by jimjim1; 28th Apr 2019 at 09:14. Reason: Added Quote which I had forgotten initially
jimjim1 is offline  
Old 28th Apr 2019, 09:33
  #4485 (permalink)  
 
Join Date: Dec 2001
Location: Leeds, UK
Posts: 281
I'm still trying to get my head around how a certification fail of how the yoke feels, somehow led to a bodged computer controlling flight surfaces? Why not introduce a feel system to the yoke to pass certification? Scary how it morphed from one thing to another without a blink.

I'd also love to hear the B engineers side of how MCAS was specced and built and whether they were put under pressure to "just do it". As somebody earlier said, where's the integrity gone?

G
groundbum is offline  
Old 28th Apr 2019, 10:11
  #4486 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 57
Posts: 378
Originally Posted by jimjim1 View Post
Also, 5 possibly relevant reports: (PDF is easy to search)
https://www.documentcloud.org/docume...-737-max8.html
https://assets.documentcloud.org/doc...r-737-max8.pdf
Still available as of 28 April 2019

"Contributed by: Cary Aspinwall, The Dallas Morning News"
HTML has some summary material (metadata) not in the pdf.

ASRS Reports for 737 max8
- A 737 Max 8 captain noted problems on takeoff
- An unidentified captain says the Airworthiness Directive does not address the
problem in November 2018.
- An airline captain called the flight manual for the Boeing 737 Max 8 "inadequate and
almost criminally insufficient."
- A co-pilot reported an altitude deviation in November.
- Co-pilot said after engaging autopilot, aircraft pitched nose down.
- Co-pilot reported that aircraft pitched nose down on departure.
- A Boeing 737 Max 8 goes nose down suddenly during takeoff, pilot reports incident.
All of those reports are suggestive, but inconclusive, in the absence of detailed corroborating evidence such as FDR readouts. The most likely explanation that matches anything to do with MCAS, would a transient high AOA value, due to turbulence soon after takeoff, or some non-permanent mechanical or electrical glitch.
GordonR_Cape is offline  
Old 28th Apr 2019, 12:06
  #4487 (permalink)  
 
Join Date: Feb 2015
Location: The woods
Posts: 1
How it should have been done...

Originally Posted by groundbum View Post
I'm still trying to get my head around how a certification fail of how the yoke feels, somehow led to a bodged computer controlling flight surfaces? Why not introduce a feel system to the yoke to pass certification? Scary how it morphed from one thing to another without a blink.

I'd also love to hear the B engineers side of how MCAS was specced and built and whether they were put under pressure to "just do it". As somebody earlier said, where's the integrity gone?

G
Thank you Groundbum - exactly my view too. As to why it was done the roundabout route via Stab tinkering - this seems long to have been a "fix-all" at Boeing and it was probably the obvious adjustment to them.
And while on the subject - refering to AoA input - what could possible go wrong? a few posts back - well a lot more can go wrong with a rogue input to the stab - as documented over many pages here... than to a feel spring.
If false AoA info were to trigger a false spring resistance, the only result would be a one time hardening of pull input - one time - which could be trimmed out.

bill fly is offline  
Old 28th Apr 2019, 12:08
  #4488 (permalink)  
 
Join Date: Apr 2019
Location: USA
Posts: 217
Originally Posted by GordonR_Cape View Post
737 Driver
A long time ago I asked a critical question, to which I did not get a satisfactory answer. Perhaps you can indulge me.

On the B737, the flight computers (and the FDR post-crash readout**) know exactly how much nose-down trim has been applied to the horizontal stabiliser, but the pilots do not.
You can be forgiven your unfamiliarity with the 737 cockpit layout, so let me help here. There is a direct reading of the stab position right next to the trim wheel:




This is the view from the First Officer side, but there is an identical trim wheel and index on the Captain's side. You can also see the stab trim cutout switches just aft of the trim index.

The stab trim setting is not in the pilot's forward scan, but it is fairly easy to check at a quick glance. The green band is the takeoff trim zone. Normal takeoff trim is usually between 4.5 to 6.0 units. From personal experience, having the trim lower than 4.0 units or greater than 7.0 units in flight is highly unusual. If you told me I had to operate a flight at one constant trim setting (assuming a normal C.G.), I would say somewhere around 5.0 units would work. We do have procedures for a jammed stabilizer, and any 737 pilot should have been exposed to this non-normal as part of their initial training.

That being said, except for your initial takeoff setting, you don't operate the trim by reference to the index. You operate by feel. If you are holding in any sustained control pressures around any axis (pitch, roll, yaw), you would normally trim away those pressures until the controls were neutral. I say "normally" because there are some flying techniques that involve leaving in some control pressures in certain situations, but that is beyond the scope of this discussion. So in reality, you could tape over the index, and as long as you trimmed out the control pressures, then maintaining aircraft control should pose no problem. Again, this is assuming that you do not have an aircraft that is loaded outside its C.G. envelope.

In terms of increased awareness of a runaway trim situation, I will say that I liked the approach used on the MD-80. There was a aural tone that sounded for every degree of stab movement. A continuous movement of the stab, would produce a steady series of tones. This tone was active with the autopilot on or off, and actually was most useful in a situation where the autopilot was engaged, but the speed was bleeding off (i.e. autothrottles inop or disengaged while autopilot was attempting to hold altitude). Now, before anyone thinks this is a surefire fix, I'll just make the observation that flight crews have, on multiple occasions, demonstrated the ability to not hear aural warnings as well (most notably, gear up landings).

If there is one point I would like to hammer home, it is this. A pilot's primary reference to the aircraft's trim state is through the feel of the controls. If the control pressures are heavy, then they are not trimmed for whatever maneuver is being performed. For an experienced pilot, trimming properly should reside somewhere near the unconscious level - I believe procedural memory is the proper term. Learning to trim correctly is one of those psychomotor skills that can only be developed by constant practice - like riding a bike, throwing a baseball, or learning to dance. If you are already under a heavy cognitive load, and you have to think about trimming to actually do it, then you are miles behind the aircraft.

Last edited by 737 Driver; 28th Apr 2019 at 13:09.
737 Driver is offline  
Old 28th Apr 2019, 12:13
  #4489 (permalink)  
 
Join Date: Apr 2019
Location: USA
Posts: 217
Originally Posted by GordonR_Cape View Post
All of those reports are suggestive, but inconclusive, in the absence of detailed corroborating evidence such as FDR readouts. The most likely explanation that matches anything to do with MCAS, would a transient high AOA value, due to turbulence soon after takeoff, or some non-permanent mechanical or electrical glitch.
Just an observation.... you need to filter out all the reports in which either the flaps are extended (takeoff to about 1000') or the autopilot was engaged. MCAS is inhibited by these conditions.
737 Driver is offline  
Old 28th Apr 2019, 12:32
  #4490 (permalink)  
 
Join Date: Jan 2008
Location: Reading, UK
Posts: 10,648
Originally Posted by 737 Driver View Post
Just an observation.... you need to filter out all the reports in which either the flaps are extended (takeoff to about 1000') or the autopilot was engaged. MCAS is inhibited by these conditions.
Quite so.

And that leaves precisely zero events in ASRS where MCAS was a factor.
DaveReidUK is online now  
Old 28th Apr 2019, 12:52
  #4491 (permalink)  
 
Join Date: Apr 2019
Location: USA
Posts: 217
Originally Posted by wonkazoo View Post
Specifically to your point about being a forum for professional pilots. You are completely correct. Professional pilots who fly airplanes made and certificated by large companies and in the US the FAA. If this isn't the place to opine about a substandard or frankly lethal product being provided to you on which you will earn your living I don't know what is.

Look, I agree with pretty much everything you say, up to the point where it was the crews' fault for what happened. They had no control over their training. They had no control over Boeing's clear abuse of the certification process, and they had no input on a system which (as I have described previously) would try to kill you if you didn't get the diagnosis right. And that's an important distinction. In an earlier reply you spoke of numerous failures that you experienced which were nearly catastrophic. IIRC you mentioned engine failures as an example of a potentially fatal event. The problem is twofold: First, you need engines to fly. Like, I mean, you can't live without them right?? And second: Yes, an engine failure can be (and many times is) fatal. But it doesn't have to be, and if you goof a little bit the outcome isn't a 100 percent fatal conclusion. Even if you goof a lot the outcome is frequently far less than fatal.

If an engine failure happens there are a gazillion possible outcomes, all of which rely on you, the pilot, to select and control the final result.

But with MCAS there are only two possible outcomes, one is you keep flying if you line up the dots within roughly a minute (40 seconds if you believe media reports) the other is you are d-e-d dead.

I know of no other binary system/operational outcome that aligns with this and I once again submit this to the community: Anyone got an equivalent they can share??

Here is another attempt to try to illustrate how unique this circumstance is, and how profound Boeing's errors were in bringing this airplane forward for commercial use. (And how culpable they are as a consequence)

Imagine if your airplane had a third, previously unneeded engine that contributed nothing to the performance, stability, safety or functionality of the aircraft. I'm even going to give us the benefit of the doubt and say you know this third engine exists. If engines 1 or 2 fail you just do everything like you always have. Pull out the proper checklist, do your memory items and find someplace to land. But if engine #3 fails, well then you have 30 seconds to a minute to identify the correct engine, diagnose it and shut it down using an exact mechanism that has zero tolerance for deviation. If you fail to do this exactly right your third engine explodes and rips off the tail in the process and you and your airplane are toast on a stick.

That's what I mean when I say MCAS will try to kill you (it will...) and that's why I believe this is a unique circumstance and finally: That's why I place the responsibility for the entirety of the outcome for both flights at the feet of Boeing and the FAA.

As professional pilots I would think you would be equally interested (for the sake of your own preservation) in holding the manufacturer that created this Rube-Goldberg piece of utter BS to account, certainly as much, if not more than you want to hold a deceased and clearly under-trained crew to account. And I cannot imagine a more powerful lobby or forum- the pilots who fly the planes themselves for sharing your views and insights into what happened and how utterly unnecessary both hull losses were.

We can go round and round picking apart the numerous ways both crews could have and should have done better. What I would rather see is a discussion on how the community is going to hold the feet of Boeing and the FAA to the fire- for real (even incremental) change going forward. All so that the next "MCAS" never even gets to the drawing board because the concept was shot down in flames earlier in the process due to the manufacturer's desire to build safe airplanes no matter the cost.

Warm regards,
dce
Wonkazoo,

I understand your position, and I fully support what you say about the need to hold the manufacturers, regulators and airlines accountable. However, I guess we will have to just agree to disagree on how much control the crews had on the outcome.

To me, the great tragedy of these two accidents is not in the complexity of the malfunction, but rather in the simplicity of the appropriate response.

Since my very first days as a student pilot, one primary commandment has been repeated over and over and over again. This commandment applied to all operations, normal and otherwise. Following this commandment may not always save the day, but disregarding it will almost always lose it.

So I'll say it again and again, for as long as it takes: FLY THE AIRCRAFT, first, last, and always.

For all situations, for all malfunctions, for all weather conditions, for all regimes of flight, some pilot must be actively monitoring, and if necessary, actively flying the aircraft. Whenever there is undesired or unexpected aircraft state, the pilot's first and most important priority is NOT to figure out what it is going wrong. The pilot's first priority is to FLY THE AIRCRAFT. Set appropriate attitude and power, monitor the performance, trim as necessary, adjust as appropriate.

In all of the discussion of these accidents, there has not been a single shred of evidence that the primary flight controls or trim were not responding to pilot inputs. There has been no credible argument that if the pilot flying had simply set a reasonable pitch attitude, set a reasonable power setting, and trimmed out the control pressures, that the plane would not have been flyable. The fact remains that in the heat of the moment, these crews forgot or disregarded the first commandment of aviation - FLY THE AIRCRAFT, first, last, and always. WHY they forgot and what corrective measures can be taken to train future crews should certainly be part of a serious post mortem, but we cannot remain in a state of denial regarding what happened.

Sadly, no matter how many times this commandment is repeated, and tacitly acknowledged by just about every pilot, we still seem to have difficulty applying it in practice. Any professional pilot here likely has access to various incident/accident reports. We can read the narratives and easily determine in which cases there was someone actively monitoring or flying the aircraft, and in which cases they were not. Fortunately, most of the "not's" do not wind up as a smoking hole somewhere, but it is still somewhat distressing how often the first commandment is forgotten.

FLY THE AIRCRAFT, FLY THE AIRCRAFT, FLY THE DAMN AIRCRAFT

Last edited by 737 Driver; 28th Apr 2019 at 13:18. Reason: grammar, clarity
737 Driver is offline  
Old 28th Apr 2019, 13:36
  #4492 (permalink)  
 
Join Date: Mar 2018
Location: Central UK
Posts: 283
I'm still trying to get my head around how a certification fail of how the yoke feels, somehow led to a bodged computer controlling flight surfaces? Why not introduce a feel system to the yoke to pass certification? Scary how it morphed from one thing to another without a blink.
Please explain why the current solution should be seen as "bodged" when you already have a speed-trim system (isn't that an electronic 'bodge' too?) that this can ride off the back of, and why it would be 'better' (ie in your terminology somehow less bodged) to add a completely new artificial feel system - despite the fact that MCAS is just that anyway - artificial feel. On the KISS principle MCAS seems remarkably rational to me though as we now know with attendant and unforseen failure modes. I'm far from saying Boeing and the FAA (and by association every other authority that has the MAX on it's register) haven't made errors, but to suggest they are criminal, negligent, collusion or whatever is unfair and irrational without evidence that they are more than mere honest errors.

Despite ample - frankly overwhelming evidence to demonstrate the pilots were guilty of 'bodges' galore no one here is using that sorrt of loaded, accusative and frankly rather hysterical terminology against them, despite their many failures to act correctly being the ultimate causes of the accidents? Rather they are rationally and logically discussing pilots' 'errors' and 'omissions'. It seems only when Boeing and the FAA are involved this hysteria ramps up and wild, totally unsubstantiable accusations of bodgery, incompetence, colliusion, criminal negligence, lethal design etc get flung around when the worst we can see they did from the info we have avaiable is underestimate a consequence, albeit one that imo they quite correctly and reasonably took to be a given as the entire industry always has done from day one. That is; design aircraft to be safe in the hands of - in your terminology, any non 'bodging' crew to sort out using existing procedures? Boeing quite reasonably believed that a trim runaway brought about by MCAS would be seen for what it is - a trim runaway and dealt with accordingly. As, I submit, would 98% of commentators here had that proposition question been put to them before the accidents. I have no doubt they were as gobsmacked as the rest of us when that proved not to be the case. But then they didn't factor in what's beginning to look like sub-standard training and operating procesures and airlines that don't adhere to international standards of promulgating safety bulletins - and why should they have done?

It is abundantly clear that a lot of commentators here are neither pilots nor much related to the industry and clearly many are subject to a '100% of everything must always be 100% safe" delusion or of the muck-stirring 'always assume evil and corruption in business and politics' brigade but some of the shrill and baseless accusations made are nothing short of scurrilous, especially in view of the lack of evidence to support them. Despite the amount of unequivocal evidence of misdeeds by the pilots that could not unreasonaby be couched in that same shrill accusatory language they are notably much more understated and kept to facual levels as befits Professional Pilots discussing Professional matters. The input of conspiracy-theorey activists is most unwelcome and quite inappropriate in an accident investigation discussion, especially so when they continue to use inflammatory language and wild baseless accusations - it merely flags up their unpleasant agenda and irrational, random thought processes which advances our knowledge and understanding not one jot.

In other words, present evidence and discuss it rationally - please!
There are shades of grey, not everything is black or white.
Present, discuss, conclude, don't just shoot from the hip...

Last edited by meleagertoo; 28th Apr 2019 at 14:05.
meleagertoo is offline  
Old 28th Apr 2019, 13:47
  #4493 (permalink)  
 
Join Date: Jun 2003
Location: UK
Posts: 2,181
Originally Posted by 737 Driver View Post

FLY THE AIRCRAFT, FLY THE AIRCRAFT, FLY THE DAMN AIRCRAFT
Your mic may be stuck.

Could you answer this? Why do you think "the flying of the aircraft" did not occur satisfactorily in the ET accident?
HundredPercentPlease is online now  
Old 28th Apr 2019, 13:59
  #4494 (permalink)  
 
Join Date: Aug 2004
Location: US
Posts: 428
Originally Posted by HundredPercentPlease View Post
Your mic may be stuck.

Could you answer this? Why do you think "the flying of the aircraft" did not occur satisfactorily in the ET accident?
Because the airplane ended up in a smoking hole?
20driver is offline  
Old 28th Apr 2019, 14:01
  #4495 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 57
Posts: 378
meleagertoo
Please explain why the current solution should be seen as "bodged" when you already have a speed-trim system (isn't that an electronic 'bodge' too?) that this can ride off the back of, and why it would be 'better' (ie in your terminology somehow less bodged) to add a completely new artificial feel system - despite the fact that MCAS is just that anyway - artificial feel. On the KISS principle MCAS seems remarkably rational to me though as we now know with attendant and unforseen failure modes. I'm far from saying Boeing and the FAA (and by association every other authority that has the MAX on it's register) haven't made errors, but to suggest they are criminal, negligent, collusion or whatever is unfair and irrational without evidence that they are more than mere errors.
I am not a pilot, but from an engineering and computer programming viewpoint, the items listed by BluSdUp (edit: in the other thread) are exactly those I would be concerned about the in the revised MCAS implementation. See: Boeing 737 Max Software Fixes Due to Lion Air Crash Delayed

Any flight control system that relies on an "unreliable" input (AOA disagree threshold of 5.5 degrees), and a 10 second delay lag in implementing nose-down trim, is asking for trouble.

They (collectively) have a much lower probability of killing anyone than the original MCAS, but IMO should never be present on an aircraft that relies on manual flight controls, where precise and direct authority is expected, and pilot/machine induced oscillations could be a killer.

P.S. I have already made the same point several times in this thread (as have some others), and am not going to repeat the whole argument.

Last edited by GordonR_Cape; 28th Apr 2019 at 14:19.
GordonR_Cape is offline  
Old 28th Apr 2019, 14:10
  #4496 (permalink)  
 
Join Date: Feb 2006
Location: USA
Posts: 389
WSJ

https://www.wsj.com/articles/boeings...ff-11556456400

Boeing’s Enduring Puzzle: Why Certain Safety Features on 737 MAX Jets Were Turned Off

Some midlevel FAA officials contemplated, but then quickly dropped, idea of grounding Boeing 737 MAX jets last year


By Andy Pasztor
April 28, 2019 9:00 a.m. ET

Plane maker Boeing Co. didn’t tell Southwest Airlines Co. when the carrier began flying 737 MAX jets in 2017 that a standard safety feature, found on earlier models and designed to warn pilots about malfunctioning sensors, had been deactivated.

Federal Aviation Administration safety inspectors and supervisors responsible for monitoring Southwest, the largest MAX customer, were also unaware of the change, according to government and industry officials.

Boeing had turned off the alerts which, in previous versions of the 737, informed pilots if a sensor known as an “angle-of-attack vane” was transmitting errant data about the pitch of a plane’s nose. In the MAX, which featured a new automated stall-prevention system called MCAS, Boeing made those alerts optional; they would be operative only if carriers bought additional safety features.

Southwest’s cockpit crews and management didn’t know about the change for more than a year after the planes went into service. They and most other airlines operating the MAX globally learned about it only after the fatal Lion Air crash last year led to scrutiny of the plane’s revised design. The FAA office’s lack of knowledge about Boeing’s move hasn’t been previously reported.

“Southwest’s own manuals were wrong” about the status of the alerts, said Southwest pilots union president, Jon Weaks. Since Boeing hadn’t communicated the modification to the carrier, the manuals still reflected incorrect information.

Following the Lion Air crash, Southwest asked Boeing to reactivate the alerts on planes already in its fleet. This move, along with questions about why they had been turned off, prompted FAA inspectors overseeing Southwest to consider recommending that the airline’s MAX fleet be grounded while they assessed whether pilots needed additional training about the alerts. Those internal FAA discussions, however, were brief and didn’t go up the chain, according to documents reviewed by The Wall Street Journal.

Boeing hasn’t specifically addressed why it turned off the feature, called “AOA disagree alerts,” without informing customers. Questions surrounding that move have remained unanswered since October, when the Lion Air accident killed 189 people, followed by an Ethiopian Airlines crash in March of the same model that took 157 lives. MAX planes remain grounded. Boeing recently said it would book $1 billion in expenses tied to the groundings and related business disruptions.

In previous 737 models, the computer-generated alerts appear as colored lights in the cockpit when a plane’s twin angle-of-attack sensors provide significantly different data from each other. In the MAX, they serve the same purpose but additionally are intended to warn pilots that MCAS, the new automated system implicated in both accidents, could misfire because of faulty sensor data.

MCAS commands that automatically push down the nose of a plane can overpower a pilot’s efforts to get out of a dive. In the Ethiopian jet, which lacked the disagree alerts, it took four minutes for the pilots to realize that incorrect data was coming from one of the sensors, according to investigators’ preliminary report.

A Boeing spokesman said that from now on, “customers will have the AOA disagree alerts as standard” on all MAX aircraft, including those coming out of the factory and already delivered to airlines. Boeing is currently devising a new software package that aims to fix MCAS by making it less powerful, while also restoring the alerts. The moves are among the safeguards the plane maker and FAA have embraced to make MCAS less hazardous if it misfires, and to get the fleet back in the air.

A Southwest spokeswoman said that before the loss of Lion Air Flight 610, the carrier had assumed the alerts “as operable on all MAX aircraft.” Boeing “did not indicate an intentional deactivation,” she said. Today, the reinstated feature offers “an added cross-check on all MAX aircraft,” even though none are flying.

Although the alerts were reactivated, some midlevel FAA officials who oversaw Southwest briefly considered the possibility of grounding its fleet of roughly 30 of the 737 MAX aircraft until the agency established whether pilots needed to receive new training, according to documents reviewed by the Journal.

Less than a month after the Lion Air jet went down, one FAA official wrote that AOA-related issues on 737 MAX jetliners “may be masking a larger systems problem that could recreate a Lion Air-type scenario.”

Roughly two weeks later, other internal emails referred to a “hypothetical question” of restricting MAX operations with one message explicitly stating: “It would be irresponsible to have MAX aircraft operating with the AOA Disagree Warning system inoperative.” The same message alluded to the FAA’s power: “We need to discuss grounding [Southwest’s] MAX fleet until the AOA Warning System is fixed and pilots have been trained” on it and related displays.

The email discussions, previously unreported, were fleeting red flags raised by a small group of front-line FAA inspectors months before the Ethiopian jet nose-dived last month. The concerns raised by the FAA inspectors never progressed up the agency. Within days, they were dismissed by some involved in the discussions who concluded that the alerts provided supplemental pilot aids rather than primary safety information, and therefore no additional training was necessary. During that stretch and beyond, Boeing and the FAA continued to publicly vouch for the aircraft’s safety.

These very concerns, however—ranging from potential training lapses to confusion by many aviators about the specifics of angle-of-attack alerts—have now emerged as high-priority items as Boeing’s design decisions face scrutiny. The issues are among those being pursued by various congressional, criminal and Transportation Department investigators, say people with knowledge of their lines of inquiry.

On Wednesday, a Boeing spokesman said that while the internal FAA discussions were under way last year, “there was no data that indicated the fleet should be grounded.”

An FAA spokesman said the agency expects to mandate that all 737 MAX aircraft include working disagree alerts. But government and industry officials said questions about why the alerts were turned off in the first place remain central to uncovering the history and safety problems surrounding the MAX fleet.

Testifying before a Senate panel last month, acting FAA chief Daniel Elwell said one important factor is prioritizing data pilots receive. “Every piece of real estate in a cockpit is precious,” he said. “You put one gauge up there, you are sacrificing another.”


At American Airlines Group Inc., one of the few carriers that initially had working angle-of-attack alerts as part of a broader array of optional MAX safety features for which it paid extra, pilots are still anxious to see Boeing and the FAA get all the steps right to end the grounding.

In a meeting about a month after the first crash, a Boeing executive told American Airlines pilot union officials that American’s MAX cockpit warning lights would have helped them avoid problems like those encountered by the Lion Air pilots, union officials who attended the meeting said. A Boeing spokesman has previously said the executive didn’t recall making that statement.

“Our minds are not at ease on this,” said Dennis Tajer, the pilot union spokesman.

—Andrew Tangel, Robert Wall and Alison Sider contributed to this article.

Write to Andy Pasztor at [email protected]
Zeffy is offline  
Old 28th Apr 2019, 14:12
  #4497 (permalink)  
 
Join Date: Jun 2003
Location: UK
Posts: 2,181
Originally Posted by 20driver View Post
Because the airplane ended up in a smoking hole?
No.

My question is: WHY did the "flying of the aircraft" not actually happen? Not "prove that it didn't happen".

The crew at some point did not keep the fundamental parameters in check. I'll give you some examples as to why this may happen:

Didn't know that you had to fly the DAMN aircraft (737 Driver style)
Misdiagnosis
Distraction - both pilots
Volume of failure information leading to cognitive overload
Conflicting information leading to cognitive overload
Insufficient training, resulting in difficulty diagnosing, leading to cognitive overload
Lack of systems knowledge, resulting in difficulty diagnosing, leading to cognitive overload
Multiple failures, some not able to be cancelled, resulting in cognitive overload.
etc (looking for 737 Drivers viewpoint)

If I asked WHY the crew just did not "fly the aircraft" in the Eastern 401 accident, everyone knows the answer. So WHY did it happen here?
HundredPercentPlease is online now  
Old 28th Apr 2019, 14:26
  #4498 (permalink)  
 
Join Date: Mar 2015
Location: Washington state
Posts: 149
Originally Posted by 737 Driver View Post
Wonkazoo,
In all of the discussion of these accidents, there has not been a single shred of evidence that the primary flight controls or trim were not responding to pilot inputs. There has been no credible argument that if the pilot flying had simply set a reasonable pitch attitude, set a reasonable power setting, and trimmed out the control pressures, that the plane would not have been flyable. The fact remains that in the heat of the moment, these crews forgot or disregarded the first commandment of aviation - FLY THE AIRCRAFT, first, last, and always. WHY they forgot and what corrective measures can be taken to train future crews should certainly be part of a serious post mortem, but we cannot remain in a state of denial regarding what happened....

FLY THE AIRCRAFT, FLY THE AIRCRAFT, FLY THE DAMN AIRCRAFT
There is evidence from the voice recorder (of the second accident) that the co-pilot was unable to move the stabilizer. Now if you are convinced that the accident was the pilot's fault then you may disregard it as insufficient knowledge of how to move the backup manual wheel connected to a cable connected to the stabilizer (what year was this plane built, anyway?) but there it is. Now perhaps the definition of "primary flight control" does not include the stabilizer, but as a creature of the water rather than a dragon rider, I'd consider anything that has the ability to put the bow of the flying ship into the ground to be a "primary flight control".

Perhaps to bridge the gap between us engineer types and the dragon riders, I have a question. I have seen some very well thought out responses to the accident scenario. This is good, it helps safety to imagine what should have been done (even if it turns out not to be relevant young dragon riders can learn something more about their craft from reading it.) Now can somebody working out this scenario give me an estimate of how many feet of altitude they think that they would have lost in the accident scenario? And then, for us poor fish men down on the water, how much altitude can you afford to lose in the worst case scenario at the worst case airport? And then, how much altitude would they have lost when they were a young dragon rider and had just been qualified?

Murphy's law is actually a serious engineering proposition, not a cynical joke. Humans have a hard time understanding scale and probability. Our brains just can't comprehend tens of millions of flight hours a year, so when we judge the probability of "x AND y" happening we tend to vastly understimate the probablity of "x AND y" EVER happening. Science shows us that the new plane has a fatal flaw, it nearly killed one set of pilots and did actually kill two other sets of pilots. Was it a coincidence that it happened on a new instance of a new model of plane? Statistics can be used to model this, but intuition says that in this social media age, we would know if it was a fairly common occurence for airplanes to suddenly pitch down at takeoff. Heck, we get front page CNN coverage when there is an emergency landing because of a little bit of smoke in an aircraft cabin, and even the best dragon riders probably would have to admit that in the MCAS scenario the pasengers might have felt a bit of discomfort.

A large number of ship accidents are caused by not knowing where you are. A friend tells me that one of his captains had a habit of walking up to the most junior officer and asking "where are we, mister?" Damn good advice, and we could say that every time somebody hits (charted) rocks it was their fault and what they really needed was a guy behind them saying "WHERE ARE WE, MISTER?" Or, after a few ships have hit the rocks and caused massive oil spills, we could put a lighted bouy on the rocks to warn them.
Water pilot is offline  
Old 28th Apr 2019, 14:26
  #4499 (permalink)  
 
Join Date: Mar 2018
Location: Central UK
Posts: 283
Originally Posted by GordonR_Cape View Post
meleagertoo

Any flight control system that relies on an "unreliable" input (AOA disagree threshold of 5.5 degrees), and a 10 second delay lag in implementing nose-down trim, is asking for trouble.

They (collectively) have a much lower probability of killing anyone than the original MCAS, but IMO should never be present on an aircraft that relies on manual flight controls, where precise and direct authority is expected, and pilot/machine induced oscillations could be a killer.
Perhaps you'd add your aviation-based criticism of the Speed Trim system, the Elevator Feel system, and the Rudder pressure limiting system to name just three - that already exist, unremarked, on this manual flight control aircraft?

And explain why/how it is unacceptable to have safety augmentation devices in a manual control aeroplane yet you appear to believe it is perfectly OK for an entire aeroplane's control systems to be electronic? That doesn't seem rational.

How does the mechanical method of controlling the aeroplane ('manual' cable controls or fly-by-wire) have any bearing on the matter?
meleagertoo is offline  
Old 28th Apr 2019, 14:34
  #4500 (permalink)  
 
Join Date: Feb 2000
Location: Pacific
Posts: 734
Almost every accident is caused by or contributed to by pilot error. This accident is no different. Apart from the failure to pull the throttles back, which alone would have caused the airplane to be lost due to the excess speed, the pilot could have followed the checklist for the trim failure. He did not, or did not do it as per the checklist. The loss of the aircraft was not due to the failure of the MCAS, it was due to the failure of the crew to follow the correct emergency checklist procedures. It matters not what caused the trim to run away; in this case it was the MCAS but next time it could be a stuck trim switch, or a broken wire: are we supposed to accept that it is OK for a professional pilot to simply fly the aircraft into the ground because he was unaware of the cause of the runaway trim?

I cannot follow the logic that because Boeing did not give enough information on the MCAS to the crews (who obviously would not have understood it anyway) it is OK for a pilot to kill everyone on board his aircraft. There is no excuse for not knowing how to fly your aircraft, especially when the procedure that would have allowed the crew to safely land was in the QRH and undoubtedly had been trained, was ignored or not followed.

And more so in this case, when the Lion Air guys had already demonstrated the wrong way to handle this emergency.

Don't gang up on Boeing, sure they could have handled the introduction of the MCAS on the 737 better, but it is patently obvious that if they had done a better job, this crew would not have understood or changed anything in the way they flew and would have flown into the ground anyway. Or are we all lawyers now, going after the deep pockets? If so, how are we as pilots or the airlines or customers supposed to benefit? If we don't face up to the fact that pilots are not being trained properly and fix that, this type of accident will continue to happen.
boofhead is offline  

Thread Tools
Search this Thread

Contact Us Archive Advertising Cookie Policy Privacy Statement Terms of Service

Copyright © 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.