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 5th Apr 2019, 23:02
  #3381 (permalink)  
 
Join Date: May 2016
Location: Nantes
Posts: 63
Originally Posted by nevillestyke View Post
But reducing power pitches nose down, due to the aerodynamics of the aircraft. How much would airspeed reduce? Parking brake is of little use.
The nose up moment of the engines thrust may be important at lower speeds, but negligible at higher speeds, since the thrust is constant but any other aerodynamic moments vary as the square of speed.

What about the pitch of airbrakes ?
deltafox44 is offline  
Old 5th Apr 2019, 23:06
  #3382 (permalink)  
 
Join Date: Mar 2014
Location: Birmingham, UK
Posts: 57
Does anyone else think that maybe the very fact that they're trying to solve this through a software update may be part of the problem?
AndyJS is offline  
Old 5th Apr 2019, 23:14
  #3383 (permalink)  
 
Join Date: Apr 2015
Location: Under the radar, over the rainbow
Posts: 707
Originally Posted by AndyJS View Post
Does anyone else think that maybe the very fact that they're trying to solve this through a software update may be part of the problem?
I'm pretty sure that, as we learn more and more about the underlying issues, more people think just that, every day. I think a lot of us who have been following this closely will be surprised if the MAX gets back in the air quickly, with only a new software patch.
OldnGrounded is offline  
Old 5th Apr 2019, 23:17
  #3384 (permalink)  
 
Join Date: Nov 2010
Location: USA
Posts: 98
Aviation Week: MAX Saga Spotlights Flight-Deck Human Factors

MAX Saga Spotlights Flight-Deck Human Factors

MAX Saga Spotlights Flight-Deck Human Factors

Sean Broderick, Bill Carey and Ben Goldstein Washington

Details emerging from investigations into two fatal Boeing 737-8 MAX accidents in five months are fueling a heated debate over whether the pilots involved were adequately prepared to face their inflight emergencies or simply could not overcome failure modes rooted in a flawed design. Either scenario implicates flight-deck human-factors shortcomings that will reverberate far beyond the software upgrades Boeing is counting on to help get the 737 MAX fleet flying again.

While the investigations into the Oct. 29, 2018, crash of Lion Air Flight 610 (JT610) and the March 10 Ethiopian Airlines Flight 302 (ET302) accident are ongoing, links between the two have been established. In each case, the flight crews battled to keep a new 737-8 aloft while the aircraft’s Maneuvering Characteristics Augmentation System (MCAS) pushed the nose down by applying stabilizer trim. The MCAS, which was added to the 737 speed trim system to help the new model handle like its 737NG predecessor in certain flight profiles, relies on data from one of the MAX’s two angle-of-attack (AOA) vanes. In each accident, investigators have confirmed the aircraft was getting unreliable data from an AOA vane, which triggered repeated MCAS nose-down inputs.

Boeing is developing a software upgrade that will prevent the updated system from activating if it is fed erroneous data (see page 24). It also gives pilots ultimate elevator authority by limiting the degree of automatic nose-down stabilizer. Additional training and updated flight manuals will also be provided. These changes will be part of safety regulators’ demands to lift 737 MAX revenue-service operations bans that have grounded the 376-aircraft fleet since March 13.

The changes are a de facto admission that the MCAS needed improvement. Beyond that, questions about how well-prepared pilots were to deal with the system’s failure remain paramount in many circles. Boeing did not include any MCAS information in 737 MAX flight manuals, which some point to as an egregious oversight. It was only after JT610 that Boeing provided pilots with extensive details about handling the MCAS.

Many pilots say that though they do not agree with Boeing’s philosophy of keeping the system in the background, they acknowledge Boeing’s logic that an MCAS failure would be recognized as uncommanded stabilizer input and managed via the common “stabilizer runaway” checklist was reasonable. The checklist, which is the same on the 737NG and MAX and includes a step that cuts power to the stabilizer, is supposed to be common knowledge for airline pilots.

“Pilots of large aircraft are trained from Day 1. When the pitch of the aircraft is doing something you’re not telling it to do, you do a runaway pitch trim checklist,” Acting FAA Administrator Dan Elwell, a former airline pilot, told a U.S. Senate subcommittee during a March 27 hearing about the MAX. “In every plane I’ve ever flown, it’s called a memory item. You’re not fumbling through books. It’s a time-critical procedure, and you go to that.”

The European Aviation Safety Agency Executive Director Patrick Ky, speaking to European Parliament members on March 19, said the procedure “is not that complicated.” But the fact that it was not followed by the Lion Air crew suggests they were confused. “If they knew what was really happening, they would not have done what they did, and they would not have crashed,” Ky said.

The ET302 crew had the benefit of knowing about the MCAS, and investigators determined that they followed the prescribed procedure—at least in part. The MCAS activated and pushed the aircraft’s nose down. The crew responded with manual inputs via column-mounted trim switches, which countered only a portion of the MCAS nose-down input. The automated system, still receiving erroneous data, activated two more times, dropping the nose even more. After the third MCAS nose-down input, the crew toggled the cutout switches. Struggling to maintain altitude, they turned the system back on, which triggered the MCAS again.

Investigators are looking closely at how the ET302 crew reacted, and why they reactivated a system that they identified as central to their problem. One possibility: With power to the stabilizer cut off, they would have needed to move it by cranking a center-console-mounted wheel attached to cables and pulleys. This may have taken more time than they believed they had, or been too difficult, so they opted to reengage stabilizer power and try the column-mounted switches.

Another possibility: They may not have fully understood what they were facing. Boeing’s approach of keeping the MCAS in the background means its activation did not result in any cockpit warnings. Boeing’s assumption: Unwanted nose-down inputs, signified in part by a spinning trim wheel, would alert pilots to a runaway stabilizer and prompt them to execute the checklist.

“If I had been flying a MAX with stickshaker activation at liftoff after the Lion Air accident, shutting off the trim would have been accomplished in a matter of seconds, not minutes,” says one U.S.-based MAX pilot. “I probably would have activated the stabilizer trim cutout switches before the gear was even up. Why that didn’t happen on the Ethiopian flight is a mystery to me.”

The stickshaker warning, or artificial vibrating of the control column that signifies a stall is imminent, activated on both JT610 and ET302 because of the faulty AOA data. Meant to alert pilots of a problem, it can be more of a distraction than a help in certain scenarios,the MAX pilot says.

“Not only would the noise mask the operation of the trim [wheel], it is such a significant warning that it would command a lot of attention,” the pilot says. “Ultimately, it is not telling you anything useful, but it makes recognizing the trim runaway more difficult,especially since the trim was not continually running.”

The two MAX accidents underscore a larger concern: Is automation is beginning to supplant, instead of augment, basic flying skills? While airlines have long used it safely, pilots who typically fly with automation who were involved in accidents “made errors when confronted with an unexpected event or [when] transitioning to manual flying,” the Transportation Department Inspector General’s office found in a 2016 report to Congress.

“As a result, reliance on automation is a growing concern among industry experts, who have questioned whether pilots receive enough training and experience to maintain manual flying proficiency,” Transportation Department Inspector General Calvin Scovel said in the March 27 hearing. The initial results from the ET302 accident “raise concerns about pilots’ abilities to recognize and react to unexpected events,” he added.

The FAA now requires that Part 121 pilots be trained in specific abnormal flight conditions, including stall and upset recovery and loss of reliable airspeed, and that training on the responses be performed in full-flight simulators. But the recent 737-8 accidents have raised questions about the availability and capabilities of simulators, Scovel says. According to the FAA, “existing simulators do not fully replicate the 737 MAX aircraft, and no U.S. airline currently has a MAX simulator,” he says.

Most airline standard operating procedures “recommend and encourage” using full automation to control an aircraft for safety and efficiency reasons, observes Hassan Shahidi, Flight Safety Foundation (FSF) president and CEO. Automated systems can improve pilots’ management of the flightpath, particularly during reduced weather minima, relieving them from repetitive tasks. But by depending on automation, they revert to monitoring the system rather than actively flying the aircraft. There are various theories about how complacency affects pilot performance, says Shahidi, a former Mitre Corp. senior executive who started at FSF in January.

With all of the advantages it confers, automation is not a substitute for the function of the pilot, who ultimately is responsible for flying the aircraft, says Shahidi. With respect to pilot training, “there needs to be sufficient understanding for the basis of the automation—why there is automation in the first place—and what happens with partial or full-use” of a system, he says. Also, pilots should understand the importance of monitoring an expected function so they can take timely and corrective action if there is a malfunction.

Training that focuses pilots on abnormal situations is important, whether in a simulator or non-simulator environment, Shahidi advises. For example, pilots who fly with the autothrottle engaged, even in small aircraft, may lose the habit of regularly scanning the speed indicator. When the autothrottle disengages for some reason, the pilot may not readily notice or react to even large speed
deviations.

“Automation has the potential to cause significant issues if it is misunderstood,” says Shahidi. “Poor automation can reduce the pilots’ situational awareness and create significant workload as they are trying to figure out what the automation is doing, especially if the system fails. This certainly can lead to an aircraft getting into an undesirable state from which it is difficult or sometimes impossible to recover.”

The FSF issued a position paper on pilot training and competency in March 2018 saying the commercial aviation industry has reached a “crossroads” in determining how pilots should be trained and mentored and questioning whether the current approach can produce a “sustainable quantity and quality of pilots” for the expected future demand. Boeing has forecast a need for 790,000 new civil aviation pilots over the next 20 years. The Asia-Pacific region leads demand with a requirement for 261,000 new pilots over that time, the manufacturer predicts.

Shahidi concurs when asked if there is a need for more standardized pilot training across airlines that have different standard operating procedures and training requirements beyond what is minimally required by aviation authorities and manufacturers.

“Moving forward, especially in light of the fact that we’re looking at significant demand for new pilots in many of these regions, it is important to harmonize the level of training that is required,” he says.

Among the recommendations the FSF white paper calls for are “competency- or evidence-based” training programs that are not solely hours-based and for maximized use of simulation devices. Pilot performance criteria should be universally recognized, and the International Civil Aviation Organization tapped to define guidelines for the performance required of flight academies.

Shahidi points to the role of what he calls nontechnical competencies of communications, analysis, problem-solving, decision-making and leadership in piloting an aircraft. “It is not just understanding [which buttons to push] or disengage, but—as part of a holistic approach to training—these basic skills are important,” he says.

Fundamentally, mastering these skills is both more realistic and more valuable than practicing scores of failure-scenario combinations.

“We are not trained in many potential emergency situations,” says the U.S.-based MAX pilot. “But we are trained to fly the airplane, prioritize, work together and diagnose whatever problem is facing us. In fact, I think just about every diversion I’ve had for mechanical problems has been for something we did not train for.”

The FAA is working on guidance that stems from a 2013 working group report on flightpath management system usage that includes 18 recommendations. Much of the report’s focus is on how automation is both helping and hindering pilots. The anticipated guidance is expected to address several recommendations, including one on creating policy that airlines can integrate seamlessly into their own operations.

“The policy should highlight and stress that the responsibility for flightpath management remains with the pilots at all times,” the report says. “Focus the policy on flightpath management, rather than automated systems. Note that this policy would contain what has previously been named an ‘automation policy’ and would be broader, to emphasize flightpath management.”

Aviation Week & Space Technology April 8-21, 2019 page 26-28
airman1900 is offline  
Old 5th Apr 2019, 23:18
  #3385 (permalink)  
 
Join Date: May 2016
Location: Nantes
Posts: 63
Originally Posted by OldnGrounded View Post
I'm pretty sure that, as we learn more and more about the underlying issues, more people think just that, every day. I think a lot of us who have been following this closely will be surprised if the MAX gets back in the air quickly, with only a new software patch.
People got used to software updates... Windows needs updates after each security fault detected, Boeings need an update after each crash
deltafox44 is offline  
Old 5th Apr 2019, 23:28
  #3386 (permalink)  
 
Join Date: Apr 2019
Location: EDLP
Posts: 13
Fly Dubai 981

As edlb pointed out, there are some similarities. It is, however, a non max and a 12 sec AND trim.

wikipedia.org/wiki/Flydubai_Flight_981
Artus KG is offline  
Old 5th Apr 2019, 23:46
  #3387 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 58
Posts: 424
Originally Posted by LowObservable View Post
Woof woof, it's dark here....

So the dog that hasn't barked, as far as I know, is this:

Aside from the three-crew Lion Air flight that preceded the mishap flight, where are all the crews who might have experienced an AoA failure that triggered MCAS at flaps-up, and handled the situation safely?

For this to happen means that one of the AoA sensors has to be bad from the start, that it has to be bad enough to call a spurious stall warning, and it has to be the sensor that is driving MCAS on this flight (the last being a 50:50 chance). How often has this happened in the MAX's history? I have not seen a single such account in this long thread (I stand corrected if I missed one).

Clearly, that number would provide some valuable context to the two occurrences when the result has been fatal.
I agree that such data would be very useful, but suggest two reasons why it has not been forthcoming:
1. If AOA fails nose down (50% odds), there will be no stick shaker or stall warning or MCAS activation, and it would probably be a non event, written up for maintenance, or reported to the NASA hotline. Conversely if AOA fails nose up on the 'non-MCAS' side (25% odds), there will be stick shaker activation, but it will not be flight critical, and the fault will be repaired. In neither case will the airframe be examined in any detail.
2. In the event of AOA failing nose up and MCAS activating (25% odds), there is an overwhelming fatality rate. There are no known survivors other than the preceding Lion air flight, and unfortunately that airframe was destroyed the very next day, so there is little evidence to strip down and examine (apart from the FDR).
Does this provide some context on the relative scarcity of such occurrences?
GordonR_Cape is offline  
Old 5th Apr 2019, 23:47
  #3388 (permalink)  
 
Join Date: Apr 2015
Location: Under the radar, over the rainbow
Posts: 707
Originally Posted by deltafox44 View Post
People got used to software updates... Windows needs updates after each security fault detected, Boeings need an update after each crash
Since we're at the stage where we can accept at least small injections of humor related to this deeply-depressing mess, here's an old one, which may be more amusing if you were a geek in a now-bygone computing world:

If Operating Systems Were Airlines
OldnGrounded is offline  
Old 6th Apr 2019, 00:01
  #3389 (permalink)  
 
Join Date: Dec 2018
Location: Nyc
Posts: 49
Originally Posted by yanrair View Post
The isolation for MCAS is the STAB OFF switches beside pilots knee. Anything that moves the stab goes through these switches. But, you would have to recognize the problem. Clearly the pilots knew nothing about MCAS like the rest of us. All they may have seen is the STAB running AND - nose down.
The isolation for MCAS is reverse what it did then operate the switch before it does it again. Do that at high speed and low level while you're dealing with multiple critical warnings including IAS disagree, ALT disagree and stick shaker.

This is video game territory and I'm lost why we're still debating it.

Boeing are revising activation thresholds, adding signal redundancy, reducing maximum authority, updating the manual (FCOM, QRH and AMM) and adding training.

Why is anyone left debating whether the implementation of the automation is at fault?
Turbine70 is offline  
Old 6th Apr 2019, 00:03
  #3390 (permalink)  
 
Join Date: Nov 2010
Location: Denver
Age: 52
Posts: 70
Originally Posted by BluSdUp View Post
Working for the B Team lately
Never was a Nr Man as in memorizing a heap of Pitch Pwr.
But I love Whatever It takes , then BallPark !
Let me give it guess:
94 N1 to say 70% N1 ca 240kts F5, ehhh. 1 to 2 units ANU say 1.5 ( Trim CHANGE)
Pure guess
Thanks! That doesn't sound too bad, I still think reducing power would have mead the situation easier to handle (yes, typing from my armchair)
hans brinker is offline  
Old 6th Apr 2019, 00:06
  #3391 (permalink)  
 
Join Date: Nov 2010
Location: Denver
Age: 52
Posts: 70
Originally Posted by pchapman View Post
Not sure about jumping into this 3300+ post thread but here goes:
I'm concerned about the crew's actions both for the manual trim attempt, and with the electrical trim attempts just before they lose control. And I can't help but distrust the inexperienced First Officer more.

To summarize:
- they had the MCAS pitch control problems
- got the Cut-out switches to stop MCAS from trimming down any more
- they're flying OK, got some altitude banked, although still in crisis mode, still having to hold back pressure, deal with the one stick shaker, overspeed clackers after some point, and what not
- there's a discussion about the manual trim not working. We don't know the exact words used.
- Interpretation: Might have to do with high forces on the manual trim wheel, but it is also unclear whether the First Officer really doing what was expected. Was he just hitting the thumb trim switch, or trying to roll the trim wheel (without much experiences of the forces that might be involved) or did he actually have the fold-out handle open and still find he was unable to turn it?
- either way, they keep flying OK for another minute or so
- then there are the 2 very short electrical up trim attempts (as others have also commented on). I agree that the graphs show a small up trim effect. Hard to see on the graph but really does look like a little bit of nose up trim change. Barring any totally unexpected electrical issue, this is only possible with electric trim turned back on!
- very soon after, MCAS kicks in, trimming a bunch nose low, and the pilots lose control and nose over.

The preliminary report gives no indication that the pilots discussed turning electric trim back on. Nothing about, "Despite the AD saying to leave it off, you turn electric trim back on, I'll trim out these awful control forces, you turn it back off before this thing tries to pitch down again."
Did the Captain try turning it back on? Or was the First Officer fiddling around on his own? Trim gets turned back on with no consultation and but only 2 ineffective short trim activations are made. Were jackscrew forces so high that electric trim wasn't working well? Who knows, I'll leave that to the experts. But the electric trim attempts were two short blips, not like trying mashing that button down a full 5 seconds to see if they got some effect out of it.

That really messed them up, ineffective trim attempts without killing the electric trim system again afterwards.
So, before you bad mouth the FO, AFAIK he was the one suggesting using the stab cut out switches, inhibiting the third MCAS input. We don't know who switched the trim back on, or who made the insufficient inputs close to the crash, so if I were you I would give it a rest.
hans brinker is offline  
Old 6th Apr 2019, 00:07
  #3392 (permalink)  
 
Join Date: Nov 2010
Location: Denver
Age: 52
Posts: 70
Originally Posted by ManaAdaSystem View Post
68 tons of -800 at 10000 ft with FL5 and 240 kts will need about 72-73% N1, but the trim will stay around 4 units.
I don't think you'll ever see the trim near full AND during normal ops.
Thanks, and as replied already by mana, it was the change, not the absolute value.
hans brinker is offline  
Old 6th Apr 2019, 00:09
  #3393 (permalink)  
 
Join Date: Nov 2010
Location: Denver
Age: 52
Posts: 70
Originally Posted by Bleve View Post
I don’t fly the 737 Max, but after doing a bit of ‘what-if-ing’, I’ve come to the conclusion that if I were a Max pilot and had stick shaker on take-off, I would not fully retract the flaps. I would retract to Flaps 1. My experience on other Boeing’s (767 & 744) is that Flaps 1 provides the best manoeuvre margin to the stall and in the 737 Max would prevent MCAS doing it’s (evil) thing.
Yeah, considering they switched the trim off, it sounds like they were aware of what was happening as far as trim. However they never reduced power and going 350kts with the flaps out is probably going to end bad too.
hans brinker is offline  
Old 6th Apr 2019, 00:10
  #3394 (permalink)  
 
Join Date: Nov 2010
Location: Denver
Age: 52
Posts: 70
Originally Posted by deltafox44 View Post
Attempted level was FL320 !

With stall warning and attempting to clear the terrain, more speed means more potential energy, so I think overspeed was not on top of their priorities.

If Boeing NNC for trim runaway had included "reduce speed if manual trimming is too hard", perhaps they would have done it...
They set a lower alt in the window, so not trying to climb to F320
hans brinker is offline  
Old 6th Apr 2019, 00:39
  #3395 (permalink)  
 
Join Date: Jan 2019
Location: Belfast
Posts: 36
Originally Posted by deltafox44 View Post
The nose up moment of the engines thrust may be important at lower speeds, but negligible at higher speeds, since the thrust is constant but any other aerodynamic moments vary as the square of speed.

What about the pitch of airbrakes ?
🧐

Thrust also varies as a square of speed?!

Acceleration is "metres per second squared"

I think I see where you may be trying to come from... thrust and drag are balancing out, increasing thrust will result in an ever decreasing change in velocity. (Or less acceleration)

But if you agree with this, you must also accept... in that state, a marginal reduction in thrust will result in an ever increasing change in velocity? (Or more deceleration)

My point is, doesn't matter where on the scale you want to pick, drag and thrust are always going to be proportional to eachother.

Last edited by positiverate20; 6th Apr 2019 at 01:18.
positiverate20 is offline  
Old 6th Apr 2019, 00:43
  #3396 (permalink)  
TLB
 
Join Date: Mar 2009
Location: Ottawa, Canada
Posts: 64
Originally Posted by hans brinker View Post
Yeah, considering they switched the trim off, it sounds like they were aware of what was happening as far as trim. However they never reduced power and going 350kts with the flaps out is probably going to end bad too.
Don't know what you mean by "flaps out" hans brinker. If you mean the flaps were 'down', they were not. The flaps were up. If they were down, the MCAS would not have activated.
TLB is offline  
Old 6th Apr 2019, 00:57
  #3397 (permalink)  
 
Join Date: Feb 2009
Location: Cambridge (the original one)
Age: 72
Posts: 53
>"Firing blanks"

That's not uncommon in avionics software. It simplifies code - and that, more often than not, makes it more reliable. Plus, the not-engaged FCC is then a hot spare, ready to engage without further logic.
Gegenbeispiel is offline  
Old 6th Apr 2019, 01:11
  #3398 (permalink)  
 
Join Date: Jun 2001
Location: Rockytop, Tennessee, USA
Posts: 5,820
Originally Posted by deltafox44 View Post
Attempted level was FL320 !
From the narrative excerpted above:

At 05:41:21, the selected altitude was changed from 32000 ft to 14000 ft.
Addis airport elevation is 7657 feet.

Originally Posted by deltafox44 View Post
If Boeing NNC for trim runaway had included "reduce speed if manual trimming is too hard", perhaps they would have done it..
Or if it said 'don't forget to fly the plane!' Avoiding the overspeed by reducing power is pretty basic in my opinion.
Airbubba is online now  
Old 6th Apr 2019, 01:11
  #3399 (permalink)  
 
Join Date: Feb 2009
Location: Cambridge (the original one)
Age: 72
Posts: 53
Originally Posted by AndyJS View Post
Does anyone else think that maybe the very fact that they're trying to solve this through a software update may be part of the problem?
Since the MCAS is just software (running in the FCCs) there is no other way.

Gegenbeispiel is offline  
Old 6th Apr 2019, 01:12
  #3400 (permalink)  
 
Join Date: Nov 2010
Location: Denver
Age: 52
Posts: 70
Originally Posted by TLB View Post
Don't know what you mean by "flaps out" hans brinker. If you mean the flaps were 'down', they were not. The flaps were up. If they were down, the MCAS would not have activated.
Yes, I should have said down, but was a direct reply to someone suggesting he would have left the flaps F1 in this situation, and seeing as we are up to 3000+ post already I would suggest you read before you post, so we don't get endless back and forth like this.
hans brinker is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

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