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 11th Apr 2019, 17:21
  #3861 (permalink)  
 
Join Date: Feb 2015
Location: The woods
Posts: 1
Originally Posted by alf5071h View Post
L39 Guy,

“… sadly this all points to …”
… the persistence, the difficulty in avoiding hindsight, prejudging, assuming; -

How would a crew know that stick-shake just after lift off is erroneous. How is the accuracy of AoA established, of attitude, or at that instant, speed, until the accuracy of these is confirmed all that remains is assumption.
Whatever is argued, the correct AoA cannot be determined even with an EFIS display - which one is correct.
The on-side airspeed will be inaccurate because AoA is used in pressure error correction. Similarly the EFIS low speed awareness - based on AoA, perhaps adding to a belief that stick shake is valid.
Add distractions of Speed and Altitude Disagree alerts, and Feel Diff Press, together with the surprise of an unexpected event, or even higher stick force due to Feel Diff, then all available thinking ability is required to manage what exists.
In time, crosschecking speed, attitude, ‘feel’ (experience) might judge otherwise.

Assumptions such as “… it cannot be a bona fide stall but an erroneous indication” can be just as hazardous as an abnormal technical situation.
And thus any conclusion based on such assumption fails because of false reasoning.

Back to the links in Ethiopian airliner down in Africa - background reading.


Well, you can have that problem on all aeroplanes which have an AoA dependent stall warning system. And there are a lot of those certified.
bill fly is offline  
Old 11th Apr 2019, 17:32
  #3862 (permalink)  
 
Join Date: May 2010
Location: Boston
Age: 69
Posts: 442
From GordonR
Article summarises the fact that evidence is all in the FDR readout:
- Initial wild gyration of AOA sensor immediately after takeoff.
- Thereafter stable at implausibly high AOA.
- Moments before the final crash, the aircraft entered a negative-g bunt, AOA sensor flipped 180 degrees, stick shaker stopped.
Already hinted before in this thread, but the only physical explanation consistent will all of these data points is vane detachment. The event is clearly different from Lion Air, which had a constant 20 degree offset, but the end result was the same.
The other item that strongly supports the departing vane theory is the master caution anti icing alarm and AoA heat trace off at the same time.
This would not (likely) not happen if some coupling/gear inside the sensor failed

Last edited by MurphyWasRight; 11th Apr 2019 at 17:35. Reason: typo
MurphyWasRight is offline  
Old 11th Apr 2019, 17:43
  #3863 (permalink)  
 
Join Date: Jan 2008
Location: Reading, UK
Posts: 12,292
Originally Posted by Loose rivets View Post
Is the vane detachment - the very bizarre coincidence of two different failures - solely based on the different angle split readouts?
Two independent failures would indeed be a bizarre coincidence.
DaveReidUK is offline  
Old 11th Apr 2019, 18:45
  #3864 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 59
Posts: 424
Originally Posted by DaveReidUK View Post
Two independent failures would indeed be a bizarre coincidence.
To understand the prior probability of two independent failures, would require use of Bayes Theorem, and statistical data on each failure mode. It is clear from the (lack of) response to queries about rates of AOA failure, that this data is scarce.

I have stated previously that AOA fail nose-down would give interesting information, but if the vane departs, the failure will always be nose up. Isn't that even more ironic, that the only known failure mode is 100% guaranteed to trigger MCAS!?

The fact that the MAX is a new aircraft may have something to do with the Lion Air fixed offset error, but presumably bird-strike frequency has nothing to do with the model type.
GordonR_Cape is offline  
Old 11th Apr 2019, 18:54
  #3865 (permalink)  
 
Join Date: Jul 2003
Location: An Island Province
Posts: 1,154
bill fly, #3910,
The point was not about whether a system can be certificated or not; it is about the human reaction and ability to manage the failure.
The 737 Max has several consequential ‘failures’ associated with AoA inaccuracy; in many other aircraft there is little or no interaction due to different system design or redundancy.
The certification judgement should involve crew workload, conflicting alerts, ambiguity, and distraction; consideration of the wider situation, risk of assumption, and comparison with additional tasks on top of a range of normal procedures. Thus it could be argued that the 737 requires a different approach, this could involve training or from what has been learnt for these accidents, reconsidering the overall system - AoA, speed correction, low speed awareness, alerts …
alf5071h is online now  
Old 11th Apr 2019, 19:19
  #3866 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,069
Gordon et al.
Would the scenario of a loose vane about the shaft fit all incidents.

Re Lion, the vane rotated on the shaft as far as 22 deg (grub screw); the vane aligns with the airflow, but misreads due to the repositioned shaft under the influence of the balance weight. This also enabled the vane to be checked by maintenance without fault, because the vane was still free to move around the shaft.

With Ethiopian, the vane became completely free to rotate during takeoff, then the balance weight positioned the shaft equivalent to full up ? (down?). The vane remained free in the airflow, but without connection it was irrelevant. The shaft remained free such that the balance weight reflected the effects of ‘g’ at the end of the flight.
Have the ‘errant’ AoA vane units been found; loose vane, or no vane.
safetypee is offline  
Old 11th Apr 2019, 19:28
  #3867 (permalink)  
 
Join Date: Jan 2008
Location: Reading, UK
Posts: 12,292
Originally Posted by GordonR_Cape View Post
To understand the prior probability of two independent failures, would require use of Bayes Theorem, and statistical data on each failure mode.
Isn't Bayes' Theorem a way of calculating conditional probability? If two failures are truly independent of each other, you don't need Bayes to work out the probability of both happening together.

Having said that, I have no idea what are the two independent failures the poster is referring to - one is the AoA vane detachment, but what is the other one ?
DaveReidUK is offline  
Old 11th Apr 2019, 19:50
  #3868 (permalink)  
 
Join Date: Jul 2009
Location: France
Age: 58
Posts: 44
Originally Posted by DaveReidUK View Post
...what are the two independent failures ...
https://www.bloomberg.com/news/artic...lure-data-showA review of public databases by Bloomberg News reveals the potential hazards of relying on the devices, which are mounted on the fuselage near the plane’s nose and are vulnerable to damage. There are at least 140 instances since the early 1990s of sensors on U.S. planes being damaged by jetways and other equipment on the ground or hitting birds in flight. In at least 25 cases in the U.S., Canada and Europe, the damage triggered cockpit alerts or emergencies.



spornrad is offline  
Old 11th Apr 2019, 20:10
  #3869 (permalink)  
 
Join Date: May 2010
Location: Boston
Age: 69
Posts: 442
Originally Posted by safetypee View Post
Gordon et al.
Would the scenario of a loose vane about the shaft fit all incidents.

Re Lion, the vane rotated on the shaft as far as 22 deg (grub screw); the vane aligns with the airflow, but misreads due to the repositioned shaft under the influence of the balance weight. This also enabled the vane to be checked by maintenance without fault, because the vane was still free to move around the shaft.

With Ethiopian, the vane became completely free to rotate during takeoff, then the balance weight positioned the shaft equivalent to full up ? (down?). The vane remained free in the airflow, but without connection it was irrelevant. The shaft remained free such that the balance weight reflected the effects of ‘g’ at the end of the flight.
Have the ‘errant’ AoA vane units been found; loose vane, or no vane.
Not sure why a common AoA failure cause is needed, in any case there are very significant differences between the 2.

Lion Air :Incorrect (offset) values present for entire (2!) flights, the offset shows as soon as speed is sufficient.
Not sure of aerodynamics but doubt that a damaged vane would have a consistent offset.
Since this was a refurbished unit it is possible that it was mis-assembled or damaged on installation/test, only other possibility is a wiring issue, which might fit with prior unit being intermittently bad. One can be sure the investigators will be looking at this in depth.

Ethiopian: Values ok at first then a sudden and dramatic shift to full over and at the same time the heater circuit opens.
This can be explained by bird strike or prior (ramp rash) damage.

Recovery of either unit intact enough for analysis is unlikely due to high energy of crashes and position of sensors.
Finding the missing vane is also unlikely due to small size and other factors, an all out search might find it but would at most just confirm one of the 2 causes, prior damage or bird strike.

Last edited by MurphyWasRight; 11th Apr 2019 at 20:16. Reason: Added note on finding vane.
MurphyWasRight is offline  
Old 11th Apr 2019, 20:34
  #3870 (permalink)  
 
Join Date: Jan 2008
Location: Hotel Sheets, Downtown Plunketville
Age: 73
Posts: 0
Originally Posted by MurphyWasRight View Post
Not sure why a common AoA failure cause is needed, in any case there are very significant differences between the 2.

Lion Air :Incorrect (offset) values present for entire (2!) flights, the offset shows as soon as speed is sufficient.
Not sure of aerodynamics but doubt that a damaged vane would have a consistent offset.
Since this was a refurbished unit it is possible that it was mis-assembled or damaged on installation/test, only other possibility is a wiring issue, which might fit with prior unit being intermittently bad. One can be sure the investigators will be looking at this in depth.

Ethiopian: Values ok at first then a sudden and dramatic shift to full over and at the same time the heater circuit opens.
This can be explained by bird strike or prior (ramp rash) damage.

Recovery of either unit intact enough for analysis is unlikely due to high energy of crashes and position of sensors.
Finding the missing vane is also unlikely due to small size and other factors, an all out search might find it but would at most just confirm one of the 2 causes, prior damage or bird strike.
If the rwy was closed after the accident and if someone then bothered to examine it for FOD after the event, it is unlikely any feathers will turn up along with the missing AoA vane.
Chronus is offline  
Old 11th Apr 2019, 21:07
  #3871 (permalink)  
 
Join Date: Jul 2000
Location: sunny side
Posts: 8
I had a look again at the data recorder readout in the context of the AoA being detached and have doubts about the coincidence of AoA going rogue exactly at rotate.
Perhaps the stall warning was genuine.
The RTOW was 72400 and actual TOW was 71896.
Possibly more baggage than standard weights with a lot of long haul pax transferring.
The rotate induced 1.5 G and peaked at 20 deg pitch swiftly reduced to less than 10 deg and 0.8G.
Somehow this kicked off the erroneous AoA disparity and subsequent activation of MCAS.
This was a take-off at max weight and high altitude that demanded careful handling.
Apologies if this has been covered.
firewall is offline  
Old 11th Apr 2019, 21:22
  #3872 (permalink)  
 
Join Date: May 2010
Location: Boston
Age: 69
Posts: 442
Originally Posted by firewall View Post
I had a look again at the data recorder readout in the context of the AoA being detached and have doubts about the coincidence of AoA going rogue exactly at rotate.
Perhaps the stall warning was genuine.
The RTOW was 72400 and actual TOW was 71896.
Possibly more baggage than standard weights with a lot of long haul pax transferring.
The rotate induced 1.5 G and peaked at 20 deg pitch swiftly reduced to less than 10 deg and 0.8G.
Somehow this kicked off the erroneous AoA disparity and subsequent activation of MCAS.
This was a take-off at max weight and high altitude that demanded careful handling.
Apologies if this has been covered.
The left AoA went rogue about 6 seconds after wow changes. The right one did not, stick shaker only on left side.
Hard to see how it was genuine stall since right AoA looks nominal.
MurphyWasRight is offline  
Old 11th Apr 2019, 21:26
  #3873 (permalink)  
 
Join Date: Dec 2014
Location: Hungary
Posts: 26
Originally Posted by GordonR_Cape View Post
Article summarises the fact that evidence is all in the FDR readout:
- Initial wild gyration of AOA sensor immediately after takeoff.
- Thereafter stable at implausibly high AOA.
- Moments before the final crash, the aircraft entered a negative-g bunt, AOA sensor flipped 180 degrees, stick shaker stopped.
Already hinted before in this thread, but the only physical explanation consistent will all of these data points is vane detachment. The event is clearly different from Lion Air, which had a constant 20 degree offset, but the end result was the same.

Re-posting annotated FDR readout:

All 3 sensors on the left side mess up at the same time , just not as bad as the AOA senser and also in the flip between 05:43:15 and 05:43:30 there is a influence on altitude speed and AOA so it seems the problem is not just with the one sensor but with the wiring or the adiru.
and on both flights they never look further then one of the sensors and then just clear the errors.
Is this poor acceptance checks or something ? just take the plane out for a circle and thats it and then 4 months later errors show up and the solution is to just spray some wd40 on the connector and clear the error log
maxxer is offline  
Old 11th Apr 2019, 21:36
  #3874 (permalink)  
 
Join Date: Jun 2003
Location: UK
Posts: 2,324
Originally Posted by L39 Guy View Post
If one rolls down the runway with both engines providing the advertised thrust (94%), the airspeed cross check at 80 indicates no disparity, and the aircraft is rotated normally at Vr with both airspeed indicators working normally then rotated to a climb attitude and the engines continue to turn and burn, then any erroneous stall warning has to be ignored and treated as such.
.
Really?

You rotate and get a stall warning, and because your engines are "turning and burning", then you just ignore the stall warning? Has it not occurred to you that you may have your high lift devices incorrectly set, and that the fault was the takeoff config system (Spanair)? Or you have been loaded incorrectly? Or you have an engine indication issue (Air Florida)? Just because your engines are "turning and burning" does not mean that any stall detection on rotation is erroneous. Obviously.
HundredPercentPlease is offline  
Old 11th Apr 2019, 22:08
  #3875 (permalink)  
 
Join Date: Aug 2008
Location: Moscow
Age: 41
Posts: 43
Originally Posted by maxxer View Post
All 3 sensors on the left side mess up at the same time , just not as bad as the AOA senser and also in the flip between 05:43:15 and 05:43:30 there is a influence on altitude speed and AOA so it seems the problem is not just with the one sensor but with the wiring or the adiru.
AoA data is used to adjust airspeed and altitude, so errorneous AoA signal will affect both of these.

AlexGG is offline  
Old 11th Apr 2019, 22:29
  #3876 (permalink)  
 
Join Date: May 2010
Location: Boston
Age: 69
Posts: 442
This is from the Bloomberg article, seems that pressing on with a stick shaker is not unheard of. SInce this was in Nasa safet database unlikely to be a 'third world crew', correct me if I am wrong on that.

As soon as the plane got airborne, the captain, seated on the left side, got the loud thumping noise and vibrating control column warning that the plane was about to stall, according to the NASA report. The captain’s airspeed and altitude displays disagreed with the copilot’s, indicating an error and setting off additional alerts. All of those symptoms occurred on the two recent Max crashes.

The pilots opted to continue onto their destination in spite of the multiple failures. Both the captain and the copilot said that they regretted continuing the flight and didn’t realize that they had violated their airline’s procedures by disabling the stall warning.

“A return, while considered, should have been accomplished,” said the captain.

Only after they landed did they realize that the captain’s angle-of-attack vane was bent for unknown reasons.
MurphyWasRight is offline  
Old 11th Apr 2019, 22:46
  #3877 (permalink)  
 
Join Date: Jan 2008
Location: Reading, UK
Posts: 12,292
Originally Posted by MurphyWasRight View Post
This is from the Bloomberg article, seems that pressing on with a stick shaker is not unheard of. Since this was in Nasa safety database unlikely to be a 'third world crew', correct me if I am wrong on that.
Correct. US carrier.
DaveReidUK is offline  
Old 11th Apr 2019, 23:45
  #3878 (permalink)  
 
Join Date: Jun 2001
Location: Canada
Posts: 23
Not sure whether this has been posted previously...

https://seekingalpha.com/instablog/3...se-pilot-error
Ace McKool is offline  
Old 12th Apr 2019, 00:11
  #3879 (permalink)  
 
Join Date: Nov 2018
Location: Vancouver
Posts: 67
Originally Posted by L39 Guy View Post
If one rolls down the runway with both engines providing the advertised thrust (94%), the airspeed cross check at 80 indicates no disparity, and the aircraft is rotated normally at Vr with both airspeed indicators working normally then rotated to a climb attitude and the engines continue to turn and burn, then any erroneous stall warning has to be ignored and treated as such. That is why manufacturers provide nominal pitch/power settings to insure a stall free climb or cruise while it gets sorted out. In fact, I had the same stick shaker after take-off many years ago on the 737-200 after lift-off; yes, it gets your attention but if the attitude of the aircraft is ok and the engines are running fine then it cannot be a bona fide stall but an erroneous indication.
If the crew thought it was a stall, then the stall recovery should have been implemented - it was not and in fact they (Ethiopian) tried to engage the autopilot, a definite no-no in a stall.
So even setting aside the MCAS issue later, sadly this all points to training and experience to handle a pretty basic emergency. And that points back to the airline and the CAA who are responsible for that.
Maybe you should read about the background a little bit?

ADD, Addis Ababa Bole Airport, 2334 meter [7625 ft] elevation/ASL.
DIA, Denver International Airport, 1655 meter [5431 ft] elevation/ASL.

ADD is more than 2200 feet higher than DIA. It is among the top for the list of notoriously "HOT AND HIGH" airports in the world.

It goes without saying that the aircraft there will have a different TO routine, notably requiring a higher thrust and a longer run, plus high terrain avoidance.

We have to also remember the MCAS "deadly potion" as well: AOA False high & Autopilot Off & FLAP Zero.

105 seconds into their flight, the AOA was already gone mad [IAS disagree, ALT unreliable, stick shaker], but they MUST go FLAP ZERO ASAP to gain Altitude at this higher elevation. And, they needed all the thrust.

They knew they had to play around with the flap and/or the Autopilot to keep MCAS at bay- at least one of these had to be ON to hold off the MCAS "deadly potion".

So, these are their only choices:
1. Going with with Flap On|Autopilot Off.
2. Going with Flap Off|Autopilot On.
3. Going with Flap On|Autopilot On.

Obviously going with Flap On (i.e. [1] and [3]) would go against their effort to gain altitude. That would leave them with (2). So, they turned on the AP, not because they wanted the AP, but it was because their circumstance FORCED them to turn AP on.

When the AP was turned on, they bought themselves about 60 seconds [actually only about 20 "clean" seconds because during FLAP ZERO] to continue gaining the altitude but they were still only about 300-500 feet off the ground [8000+ ft - 7626]...

In other words, the MCAS "deadly potion" constraint made their already difficult TO situation [a high elevation TO, a long run, a stick shaker, an IAS disagree, an ALT unreliable, other warnings] much much worse.
patplan is offline  
Old 12th Apr 2019, 00:45
  #3880 (permalink)  
 
Join Date: Jul 2004
Location: Found in Toronto
Posts: 610
Originally Posted by Ace McKool View Post
Not sure whether this has been posted previously...

https://seekingalpha.com/instablog/3...se-pilot-error
Thanks for the link. That is an EXCELLENT read.

Summary

The pilots (crew) mismanaged engine thrust and airspeed.

Excessive airspeed rendered manual trim ineffective.

The crew deviated from the emergency procedure.

Crew experience and competency a major contributing factor.

CONCLUSION

Examining both accidents separately provides valuable insights—it’s easy to understand how these unrelated airlines and crew may have responded in similar ways—but the overall conclusion in our previous article, “Boeing’s Grounding: Catastrophic Crashes, and Questions About Boeing’s Liability And 737 MAX Aircraft Viability,” still stands—the major contributing factor to these accidents was pilot error.

After a more comprehensive analysis of each of the two accidents, especially Lion Air Flight 610, we are persuaded more than ever that the case for pilot error—as well as inadequate training—are the dominant contributing factors in both accidents, not the only ones but the most serious factors.

The LA 610 accident is somewhat excusable since the pilots were not privy to MCAS and its challenges. Even so, there were surprising pilot practices and judgment shortfalls as well as concerns with appropriate MAX training. The Ethiopian accident, however, is more confounding since it was verified by the airline that the pilots were trained in accordance with Boeing (and FAA) recommendations. Perhaps the company’s training verification should be scrutinized.

As we have highlighted, the ET 302 pilots did follow the runaway trim procedure, at least initially. However, questions remain as to why the pilots mismanaged the airspeed and deviated from company and Boeing procedures. These actions led to an unrecoverable dive resulting in the loss of crew and passengers. We believe that the final accident report will (or should) reflect this finding.

Last edited by Lost in Saigon; 12th Apr 2019 at 01:16.
Lost in Saigon 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.