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 19th Mar 2019, 08:17
  #2041 (permalink)  
 
Join Date: May 2016
Location: Nantes
Posts: 63
Originally Posted by Sailvi767 View Post


The problem with the above is that at least in the Lion Air accident it did not happen that way. The MCAS commanded 2.5 units of nosedown trim which was countered by the CA retrimming the aircraft. 5 seconds later MCAS again commanded 2.5 units nose down trim and again it was electrically countered by the Captain. This cycle repeated itself 21 times. Thatís a lot of time to diagnose the problem. MCAS does not drive the stab full nose down and cause a instant plunge into the ground. It moves in 2.5 degree increments and electrical trim is available to counter it. Not that difficult to retrim electrically and then disable the trim motors. Your now in trim and have manual trim available.
So, under your criteria, the Captain was competent and the first officer not. But what about the MCAS ? After 21 times MCAS actions were countered by the Captain, MCAS wasn't able to "understand" it was doing something wrong ? Idem when the first officer was pulling desperately the column aft ??? No need for complex AI here, just a few lines of code...
deltafox44 is offline  
Old 19th Mar 2019, 08:28
  #2042 (permalink)  
 
Join Date: Jan 2008
Location: Irvine, CA
Posts: 90
AI, AS, PG ... who needs computers if you have such humans

Originally Posted by deltafox44 View Post
So, under your criteria, the Captain was competent and the first officer not. But what about the MCAS ? After 21 times MCAS actions were countered by the Captain, MCAS wasn't able to "understand" it was doing something wrong ? Idem when the first officer was pulling desperately the column aft ??? No need for complex AI here, just a few lines of code...
Talking about AI... I think by now it is a reasonable conclusion to make, that MCAS is an AS system. Artificial Stupidity. Conceived, implemented and certified by humans. Driven by PG. Pathological greed.
Interflug is offline  
Old 19th Mar 2019, 08:38
  #2043 (permalink)  
 
Join Date: Apr 2009
Location: Hotel Gypsy
Posts: 2,830
Originally Posted by FCeng84
While the MCAS software update developed after the Lion Air accident that is almost ready to go to the fleet will likely remove reliance on the three MCAS design assumptions listed above and thus would have greatly improved the likelihood of a safe outcome for the Ethiopian event we are left with a huge elephant in the room. After making the planned update we still must address the following:
A. How many other key points in the 737MAX safety story are based on pilot response assumptions that may not be valid?
B. How about other airplane models? Are they deemed safe based on faulting assumptions regarding pilot action?
- For instance, how may current 737 crews (all models) would not respond quickly enough to a classic stabilizer runaway that was not arrested by column cutout (i.e., pulling the call far enough)? I know this is covered in simulator sessions for 737 pilots, but is that enough?
C. Moving forward with the current status and future of commercial aviation have we gotten to the point where basic flying skills and system awareness are so low that we are at risk throughout the whole industry?
D. Can current and future pilot reaction short falls be addressed through training? If so, what kind, how much, and how often?
E. How will we know that we have achieved a sufficient industry wide level of safety?
Yes, yes, yes - excellent post. These are the real questions that need to be answered. The technical side of MCAS will be sorted as hands-on engineers are very good at that sort of thing, especially once an issue has been identified. The real problem is that we could be in Donald Rumsfeld territory (Unknown unknowns). I think it may be a slight turn-of-phrase, but FCeng84 has felt the need to use the 'assumption' word a lot. As an aviator, I've always been aware of assumptions, especially lots of them.

My view is that Boeing, FAA and probably a whole pile of other people engaged in aircraft manufacture (including my side of the Atlantic) may want to take a very close look at how their entire design, manufacture, training, introduction to passenger flying and oversight systems have evolved over the decades. This sort of thing simply should not happen on 21st century aircraft; we should have enough layers of safety such that the hole in the final, pilot's, slice of cheese is acceptably small. If the Ethiopian crash has the same flavour as Lion Air, some pretty big and incorrect assumptions have been made.
Cows getting bigger is offline  
Old 19th Mar 2019, 08:50
  #2044 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 1,848

The posts from FCeng84, #2018, Ethiopian airliner down in Africa
and Aloha KSA, #2022, Ethiopian airliner down in Africa
… … essential reading for everyone before considering posting to this or related threads - ‘sticky link’ mods?

FCeng adequately identifies the problems associated with ‘assumption’, but less so it’s association with, and the effects of hindsight.
The outcome of an event (success) does not change the risk of the decision before the event.
Similarly, outcome and assumption in piloting ability. Regulations might imply that there is ‘an average pilot’ (as I recall the wording - ‘without exceptional skill or force’), the danger is if the industry judges or classifies people in a mathematical or mechanical manner; we are not machines, computers, but variable and fallible individuals.

Aloha, et al,
An increasing difficulty in modern aviation is how experience is to be gained, communicated, and heeded by everyone. What might have been true for earlier versions of the aircraft and operational environment in the widest sense, is no longer so.
Differences training is only forward looking; design and regulation must look over back and consider the history, if not then history will ‘find all of us out’.
safetypee is offline  
Old 19th Mar 2019, 08:59
  #2045 (permalink)  
 
Join Date: Jan 2003
Location: Mud Island
Posts: 64
FAA missed certification issues

https://www.seattletimes.com/business/boeing-aerospace/failed-certification-faa-missed-safety-issues-in-the-737-max-system-implicated-in-the-lion-air-crash/

Last edited by offa; 19th Mar 2019 at 09:06. Reason: Typo
offa is offline  
Old 19th Mar 2019, 09:05
  #2046 (permalink)  
 
Join Date: Feb 1999
Location: MAN
Posts: 788
Originally Posted by Aloha_KSA View Post
As a long-time 737 driver I'll just chime-in a few points. Regarding the trim wheel ability to move the stab at high speed from full deflection: I seriously doubt that there would ever be too much force on the elevator that the trim wheel could not move it. There is a lot of leverage in the jack screw, and the turns of the trim wheel make very minute changes to the angle of the stab.

But know that moving from full deflection to neutral would take a painfully long time. On the -200 we used to wind the trim full fore and aft as part of the preflight checks. (We stopped doing that on the NG.) . Using the motor-driven trim this took about a minute to go full forward, full aft, then back to about 4 units. It is not physically possible to wind the trim wheel that fast for that long manually, especially when your aircraft is lurching about like a rodeo bull. Also note that it doesn't take much movement to change the aircraft's attitude significantly, and the aircraft is very controllable using manual trim. Back in the cargo days I once did a 20 minute flight using only the rudder and stab trim and manual power (on the -200) from after flaps up to 10000' on descent. That's right - I didn't touch the yoke, autopilot-off from climb-out to descent. It was very controllable and stable. (That was a good training exercise, too.)

My First Point: If we don't catch this mis-trim early, un-doing it manually will take a very long time and maybe more time than is available when your aircraft is only 1000' AGL. AND to use the trim wheel for more than small changes, one has to fold the handle out. A handle that has injured many a knee in simulator sessions because combined with the trim motor's speedy rotation of the wheel, it can leave one with a permanent limp.

Second: For all the arm-chair Monday morning QB's who are saying: "Oh, they should have recognized it immediately and disconnected the trim:"

(1) Just after takeoff there is a lot going on with trim, power, configuration changes, and as noted above, the darn speed trim is always moving that trim wheel in seemingly random directions to the point that experienced NG pilots would treat its movement as background noise and normal ops. Movement of the trim wheel in awkward amounts and directions would not immediately trigger a memory item response of disconnecting the servos. No way.

(2) The pilots could very reasonably not have noticed the stab trim movement. Movement of the stab trim on the 737 is indicated by very loud clacking as the wheel rotates. On the -200 it was almost shockingly loud. On the NG, much less so. HOWEVER, the 737 cockpit is NOISY. It's one reason I am happy to not be flying it any more. The ergonomics are ridiculous. Especially at high speeds at low altitudes. With the wind noise, they may not have heard the trim wheel moving. The only other way to know it was moving would be yoke feel and to actually look at the trim setting on the center pedestal, which requires looking down and away from the windows and the instruments in a 'leans'-inducing head move. On the 717, for example, Ms. Douglas chimes in with an audible "Stablizer Motion" warning. There is no such indication on the 737.

(3) The fact that they were at high power and high speed tells me the stick shaker was activated. With that massive vibrator between your legs, alternating blue sky and brown out the window, your eye balls bouncing up and down in their sockets as the plane lurches up and down in positive and negative G's, it would have been a miracle if one of the pilots calmly reached down, flicked off the stab servo cutout switches, folded out the trim handle, and started grinding the wheel in the direction of normalcy. These pilots said over the radio that they had "unreliable airspeed". So they did not even know which instruments to rest their eyes on for reliable info. Their eyes were all over the cockpit looking for reliable info, the plane is all over the place like a wild boar in a blanket not behaving in any rational way. And the flying pilot may have been using the tiny standby IFDS for airspeed and attitude. Ouch.

Finally, runaway stab trim is a very, very rare occurence up until now. We trained it about once every other year in the sim because it is so rare. And when we did it was obvious. The nose was getting steadily heavier or steadily lighter with continuous movement of the trim wheel. That is a VERY different scenario than what these pilots faced.

We also trained for jammed stabilizer, the remedy for which is overcoming it with force. The information they were faced with could very reasonably have been interpreted that way, too.

An URGENT AD from the FAA/ Boeing after Lion Air would have helped get it back to the front of the pilot's minds for sure. Extra training by the airline or an urgent pilot memo would have helped. Maybe one was issued, we don't know yet.

A better question might be: given this nose down attitude, high speed, and fully nose down or almost fully nose down stab, how much altitude would they have had to have to be able to recover. I'm thinking at least 10000 feet to recognize the problem, disconnect the switches, fold out the handle and start frantically winding the stab back to normalcy while the flying pilot tries to gain control via the elevator. It's entirely possible that this scenario, if not recognized early on, is unrecoverable at any altitude.

Ask the expert.com - could you explain TCMA and itís functionality on the MAX8?

Dogma is offline  
Old 19th Mar 2019, 09:50
  #2047 (permalink)  
 
Join Date: Sep 2017
Location: Europe
Posts: 1,418
Similarly, outcome and assumption in piloting ability. Regulations might imply that there is ‘an average pilot’ (as I recall the wording - ‘without exceptional skill or force’), the danger is if the industry judges or classifies people in a mathematical or mechanical manner; we are not machines, computers, but variable and fallible individuals.
Thanks Safetypee.

This is relevant. At some point a subtle shift occurs; moving ever so slightly away the assumption changes, more specifically the definition shifts.
It is not noticed and to the casual observer the base assumption is unchanged. Over time, the foundation upon which the assumption was founded bears little resemblance to the original.
Subtle language shifts will give a clue, but the regulator must be looking for them.
Self regulation, regulatory capture.

The safety analysis Boeing sent to the FAA reported that the MCAS could only move the plane’s horizontal tail 0.6 degrees (out of a physical maximum of a little less than five degrees). But during later flight tests, Boeing discovered that 0.6 degrees of movement wasn’t enough to avert a high-speed stall, the Seattle Times reported. Boeing eventually increased the limit to 2.5 degrees.Despite quadrupling the amount that the MCAS could move the plane’s tail, Boeing never updated the documents it sent to the FAA. FAA engineers only found out about the change after the Lion Air crash, when Boeing sent a notice to airlines explaining how the system worked.

This encapsulates organisational deviance from normal.
Although this change is more substantial, it is often missed, so the outcome isn't adjusted either.
It is an incremental tweak here, a subtle shift there.
None of it is really nefarious, over time it isn't questioned. Call it 'assumption precession'. Small changes unnoticed do not alter the course of the event however over time the outcome will be nothing like what was envisaged. They can deliver a big shock when they are finally noticed.

“The FAA believed the airplane was designed to the 0.6 limit , and that’s what the foreign regulatory authorities thought, too,” an FAA engineer told the Times. “It makes a difference in your assessment of the hazard involved.
Bad assumption, insufficient cirtique.

Economic models are built on assumptions, like rational expectations and reasonable man.
Does Bill Gates perceive a change in say interest rates and act the same as a minimum wage worker in Chicago? Of course not but these assumptions are what underpin modern economic theory and they are in part the reason why the model fails: Bad assumptions.

Last edited by Rated De; 19th Mar 2019 at 10:30.
Rated De is offline  
Old 19th Mar 2019, 10:25
  #2048 (permalink)  
 
Join Date: Jul 2009
Location: NI
Posts: 1,012
Originally Posted by unworry View Post
But to address a point on capitalism, the competition between A & B certainly stoked innovation in our industry
I'm not sure about that point. With two notable exceptions ( the A320 and 777 )* there is generally just-enough-innovation to keep up with the other. Neither wants to do anything really risky like a BWB, like the A380 should have been, because why bother?

Now that we're down to two global manufacturers the pace can slacken whilst they chip away at their 12,000 aircraft backlogs. That's not healthy competition. Neither is the careful dovetailing of widebodies so that one is not directly competing with the rival's.

It doesn't look like the Russians are gaining much traction as a competitor but if the Chinese can use their African and Asian influence to sell the C919 then perhaps we'll get back to a more innovative playing field. How will B and A compete with C when they're all using the same engines? They'll need to raise their airframe game.

* Interestingly in both cases the innovator was competing against TWO active rivals, not one.
El Bunto is offline  
Old 19th Mar 2019, 11:11
  #2049 (permalink)  
 
Join Date: Dec 2006
Location: Florida and wherever my laptop is
Posts: 1,297
A question not asked on this thread yet..

For a safety system it makes no sense to have a 'single point of failure' this is a sine qua non of any safety related engineering. MCAS is obviously a safety system and (despite disparaging comments here) Boeing engineers will not have deliberately chosen to make it unreliable. So this begs a question.
First there are a LOT of 737's flying; from Wikipedia:
" As of 2006, there were an average of 1,250 Boeing 737s airborne at any given time, with two either departing or landing somewhere every five seconds"
A large number of those 737 (if not all) will have the same AoA feeding similar ADIRU and will get an Unreliable Airspeed alert plus stick shaker if AoA disagree. Out of 50 or so Max 8 they have already had two errors only months apart. If that was the rate in the whole fleet there would be continual unreliable airspeed reports and no engineer in their right mind would hang a safety system onto a single AoA if they were that unreliable. From that one can only assume that AoA is normally very reliable to the extent that failures are very rare on previous 737 models.
So the question is - why are the AoA systems on 737Max8 failing at a rate higher than acceptable? Is this just a sad coincidence that when a safety related system is designed to use the AoA output - two failures arise in 6 months -or- is there something different about the Max that is leading to AoA problems?

Even 'ramp rash' could be affected by the different position of the engines when push crews and catering used to NG have to work with the Max.

Any thoughts?
Ian W is offline  
Old 19th Mar 2019, 11:53
  #2050 (permalink)  
 
Join Date: Nov 2007
Location: Hong Kong
Posts: 342
Originally Posted by Flap 80 View Post
It is very easy to make suggestions as to how each crew should/ could have handled the situation and I suspect the CVR/FDR readout will be disturbing.


Agreed.Not looking forward to it.
CodyBlade is offline  
Old 19th Mar 2019, 12:19
  #2051 (permalink)  
 
Join Date: Jan 2008
Location: Herts, UK
Posts: 733
Originally Posted by Interflug View Post
Talking about AI... I think by now it is a reasonable conclusion to make, that MCAS is an AS system. Artificial Stupidity. Conceived, implemented and certified by humans. Driven by PG. Pathological greed.
Yeh. Think I made that point earlier based on such a blatantly naive solution to an Alpha ~ Stick force condition, messing so brutally with stab trim.

Literally, we seem to have a black and white mandatory certification requirement, ignored or overlooked by the project engineers who postulated moving these large engines up and forwards, developing into a mad rush for an affordable workaround... to a situation that would RARELY be met; would be UNDERSTANDABLE to a crew if it was; and could be ANSWERED COMPREHENSIBLY by a pre-existing and time-honoured stick shaker (or re-engineered shaker/ pusher pair)
HarryMann is offline  
Old 19th Mar 2019, 12:24
  #2052 (permalink)  
 
Join Date: Sep 2008
Location: Australia
Posts: 32
Originally Posted by Dogma View Post



Ask the expert.com - could you explain TCMA and itís functionality on the MAX8?

I'm no systems whiz. Just trying to generate some empathy for the tech crews in these scenarios.
Aloha_KSA is offline  
Old 19th Mar 2019, 12:26
  #2053 (permalink)  
 
Join Date: Jan 2008
Location: Irvine, CA
Posts: 90
Originally Posted by HarryMann View Post
Yeh. Think I made that point earlier based on such a blatantly naive solution to an Alpha ~ Stick force condition, messing so brutally with stab trim.

Literally, we seem to have a black and white mandatory certification requirement, ignored or overlooked by the project engineers who postulated moving these large engines up and forwards, developing into a mad rush for an affordable workaround... to a situation that would RARELY be met; would be UNDERSTANDABLE to a crew if it was; and could be ANSWERED COMPREHENSIBLY by a pre-existing and time-honoured stick shaker (or re-engineered shaker/ pusher pair)
Wasn't the problem with that, that - with the new bigger engines attached to the old wings - the elevator alone is too small to begin with to come to the rescue in a perceived approaching stall scenario, particularly at high speed at altitude scenarios, and that a redesign of the tail section was to be avoided, to save money and time?

Also, as far as we know, MCAS itself might have worked as intended, had the single AoA sensor it depends on not acted up. The problem apparently is, that against fundamental state of the art aerospace engineering principles, it was designed without redundant sensor data input, thus making it irregularly vulnerable as a system, that in case of malfunction potentially leads to outcomes in the "catastrophic" category.

Anyway, the jackscrew was found in the HST full nose down position, and it's highly unlikely the pilots put it there. Pilots and passengers are dead, but Boeing's order book is still full.
Insurances rationally calculate a fatality in aircraft accidents somewhere between 2 to 4 million US $ per victim.
So we stay at about 1 billion US $ for lost lives. Plus another 1-2 billion payments to customers for lost revenue due to the grounding.
Versus 600 billion US $ for Boeing's backlog for the 737 MAX?
Of course there are also highly unpredictable non tangible risks in that equation, image related, fear of traveling public etc.
Probably manageable though, and overall, the cost analysis is still in favour of the hasty design and pseudo-certification procedure. And looking at the numbers, Boeing management still deserves their yearly bonus, no?

(if you find cynicism, you can keep it.)

Last edited by Interflug; 19th Mar 2019 at 13:42.
Interflug is offline  
Old 19th Mar 2019, 12:31
  #2054 (permalink)  
gmx
 
Join Date: Nov 2018
Location: Brisbane
Posts: 25
Although it may well turn out to be the case, it is still an assumption that a malfunctioning AoA vane induced MCAS-trim-response is involved in both incidents.

The Lion incident in particular is intriguing in terms of three things:
  1. the Captain repeatedly and successfully counteracting MCAS nose-down commands with electric trim but failing to activate the STAB TRIM CUTOUT.
  2. the Captain transferring control to the F/O while just barely managing to keep the aircraft in "level" flight himself.
  3. the apparently calm interactions of the crew with ATC.
This contrasts with what appears to have been the panicked response of the crew to the emergency encountered on ET602. It would appear that the two situations were quite different, or at least manifested in a significantly different way, or less likely (hopefully), the crew was not adequately briefed on how to correctly respond to a malfunctioning AoA vane induced MCAS-trim-response.
gmx is offline  
Old 19th Mar 2019, 12:32
  #2055 (permalink)  
 
Join Date: Jan 2003
Location: Mud Island
Posts: 64
When it comes to safety, never allow the manufacturer to set and mark its own exams!
offa is offline  
Old 19th Mar 2019, 12:43
  #2056 (permalink)  
 
Join Date: Jan 2012
Location: Welsh Marches
Posts: 58
Originally Posted by gmx View Post
This contrasts with what appears to have been the panicked response of the crew to the emergency encountered on ET602. It would appear that the two situations were quite different, or at least manifested in a significantly different way, or less likely (hopefully), the crew was not adequately briefed on how to correctly respond to a malfunctioning AoA vane induced MCAS-trim-response.
But I think the Lion Air flight had a bit more altitude to play with when the problem manifested itself, hence less panic??

Alchad is offline  
Old 19th Mar 2019, 13:06
  #2057 (permalink)  
 
Join Date: Jan 2008
Location: Herts, UK
Posts: 733
Originally Posted by Interflug View Post
Talking about AI... I think by now it is a reasonable conclusion to make, that MCAS is an AS system. Artificial Stupidity. Conceived, implemented and certified by humans. Driven by PG. Pathological greed.
Yeh. Think I made that point earlier based on such a blatantly naive solution to an Alpha ~ Stick force condition, messing so brutally with stab trim.

Literally, we seem to have a black and white MANDATORY CERTIFICATION requirement, ignored or overlooked by the project engineers who postulated moving these large engines up and forwards, developing into a mad rush for an affordable workaround... to a situation that would RARELY be met; would be UNDERSTANDABLE to a crew if it was; and which could ARGUABLY have been answered by a pre-existing and time-honoured stick shaker (or re-engineered shaker/ pusher pair)

Last edited by HarryMann; 19th Mar 2019 at 13:29.
HarryMann is offline  
Old 19th Mar 2019, 13:13
  #2058 (permalink)  
 
Join Date: Apr 2005
Location: UK
Posts: 1,863
Originally Posted by Sailvi767 View Post
They had a much easier choice, comply with the QRH for runaway trim, disconnect the trim system, manually trim and fly to destination as airline crews have been doing since the invention of electric trim.
Quite simply none of this is true in any sense.
Just This Once... is offline  
Old 19th Mar 2019, 13:20
  #2059 (permalink)  
 
Join Date: Jan 2008
Location: Herts, UK
Posts: 733
Thumbs up

Originally Posted by Nomad2 View Post
The single most powerful control surface on a transport jet is the THS.
Why Boeing would give control of it to a subsystem, like the MCAS is hard to understand.

No amount of hauling (or pushing) on the elevator is going to save you if the THS isn't where it should be.
Yup. that's where Ive been coming from too Nomad, for as long as I've been aware of MCAS
HarryMann is offline  
Old 19th Mar 2019, 13:27
  #2060 (permalink)  
 
Join Date: Dec 2013
Location: US
Age: 61
Posts: 399
Originally Posted by fizz57 View Post
I notice that in endlessly repeating your point of view you never mention the fact that the pilots were contending with unreliable airspeed and the distraction of a stick shaker. Care to illuminate us on why?
I have read the reports. There was nothing mentioned about unreliable airspeed. All 3 airspeed indicators appear to have matched. The press has said airspeed but they confuse it with AOA and the pilot report he made probably because of the stickshaker. As any pilot should know the stickshaker or stall warning is generated off the AOA probe. They had it because the AOA was damaged. Twice I have had the stick shaker activate at rotation because the AOA probe on the left side was damaged by jetways. Not a big deal. Cross check airspeed and flap/slat configuration and continue.
Sailvi767 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.