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

MAX’s Return Delayed by FAA Reevaluation of 737 Safety Procedures

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

MAX’s Return Delayed by FAA Reevaluation of 737 Safety Procedures

Old 8th Jul 2019, 14:10
  #1221 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 65
Posts: 7,358
Received 544 Likes on 343 Posts
Originally Posted by yoko1
And so have I, but that is a very different situation when the Captain/Aircraft Commander deems it is necessary to intervene. In the JT601 event, we are talking about a situation when the Captain voluntarily and with intent hands off the controls to his First Officer. If someone wishes to argue that the Captain gave his First Officer a handful of out-of-trim aircraft, I’m just going to have to point out that this would have been another crew error - which works against the narrative that there were no significant crew errors that led to the loss of control. You can’t have it both ways.
*i really shold just lurk ...*
Yes, that would most likely be outside of what the Captain's training and normal behaviors would be. (The second part).
That the crew seemed to be having difficulty in parsing just what the heck was wrong, fundamentally, points back to the training put into place by their company, and by the manufacturer -- and also what training wasn't put into place. (And perhaps compounding this issue, that their company did not buy the "option" that included the AoA miscompare alert? hard to say. Would that indication have given the Captain the "aha" he needed to better sort out the situation? Unknown)

@PJ2: Evening, sir. Thanks for that link. Interesting stuff.
Lonewolf_50 is offline  
Old 8th Jul 2019, 14:23
  #1222 (permalink)  
 
Join Date: Jul 2013
Location: Norway
Age: 57
Posts: 140
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by robocoder
Interesting take on the matter. So you're in quite a tight maneuvering spot and suddenly you get a "runaway trim" on top. Which, as the system was undisclosed, cannot look anything else than a genuine runaway trim to the pilots! Or perhaps the idea was that pilots are already used to STS, so trim going its merry way is now unsurprising... for how long?

It seems that in their zeal to leave no pilot error unturned, someone has uncovered another disturbing aspect of MCAS.

Now, for the real questions the above musings raise in a non-pilot (in case some of you have then answers on hand):
  1. This 3"-seconds-reaction-time-to-runaway figure comes from where? And is it further qualified with "in every possible situation", or "in straight and level flight", ...?
  2. How long are typical STS bursts?
Very interesting.

So, lets say the pilots mismanages the speed, for example autothrottle disconnected without pilots noticing. Speed decayes, AOA increases, passes MCAS treshold, MCAS start to put in 9 sec of AND trim. So, how should the pilots react now? Should they hit trim cutout after 3 or 4 seconds? How can they know this is a genuine MCAS event and not trim runaway?

This problem clearly needs attention. Something along the lines of a MCAS active aural message, do not hit cutout?
SteinarN is offline  
Old 8th Jul 2019, 14:43
  #1223 (permalink)  
 
Join Date: May 2010
Location: Boston
Age: 73
Posts: 443
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Lonewolf_50
*i. (And perhaps compounding this issue, that their company did not buy the "option" that included the AoA miscompare alert? hard to say. Would that indication have given the Captain the "aha" he needed to better sort out the situation? Unknown)
It is important to remember that the AoA miscompare alert was not an option.

It is a standard feature that Boeing knew was not working unless the optional AoA display was installed.

Boeing knew this for a significant amount of time -before- the Lion Air accident.
Had it been working on the penultimate Lion Air flight it might have led to fixing AoA sensor system before the accident flight.
MurphyWasRight is offline  
Old 8th Jul 2019, 15:14
  #1224 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 65
Posts: 7,358
Received 544 Likes on 343 Posts
Originally Posted by MurphyWasRight
It is important to remember that the AoA miscompare alert was not an option.

It is a standard feature that Boeing knew was not working unless the optional AoA display was installed.

Boeing knew this for a significant amount of time -before- the Lion Air accident.
Had it been working on the penultimate Lion Air flight it might have led to fixing AoA sensor system before the accident flight.
Bravo, Murphy, thank you for clearing that up. I had misrepresented how that all fit together.(And honestly, I think I misunderstood that it was only optional in terms of "see it or don't" based on a months ago discussion on "what does one put on the already scarce real estate of pilot displays" ....) Appreciate the course correction.
Lonewolf_50 is offline  
Old 8th Jul 2019, 15:16
  #1225 (permalink)  
 
Join Date: May 2019
Location: Somewhere over the rainbow...
Posts: 0
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Bend alot
If MCAS will still have a maximum 9 second run time and a pilot should cut it in 3 seconds - what is the other 6 seconds of MCAS operation still required for?
The short answer is that the other 6 seconds was driven by the FAR 25 certification requirement. There is a specific region approaching a stall over which the control forces must have a linear slope gradient whether or not there is a realistic expectation that the pilots would ever get into this corner of the flight envelope. That was the entire reason for MCAS’s existence.

Functionally, I think this 6 seconds would be superfluous in any actual approach to stall because there is also an expected reaction time to a legitimate stall warning. I don’t actually know what the FAA standard is, but a pilot shouldn’t need 9 seconds to figure out whether the aircraft is actually stalling when the stick shaker activates. Part of the stall recovery procedure involves unloading the aircraft (release back pressure at a minimum, but likely a strong push) accompanied by trim as necessary. Both of these actions would terminate the MCAS input before its full run.

On the other hand, if the pilots are somehow so distracted that they do not register what the stick shaker and aircraft energy state are telling them, I doubt they are going to pick up on the stab movement either. In that case, a full MCAS run will likely save their bacon.

yoko1 is offline  
Old 8th Jul 2019, 15:16
  #1226 (permalink)  
 
Join Date: Apr 2018
Location: Sudbury, Suffolk
Posts: 269
Received 5 Likes on 2 Posts
Originally Posted by yoko1


Singing my song here.

I haven’t had a Runaway Stab Training drill in the sim in ages, though I suspect that will change soon. I willing to be the accident crews had similar experiences. There are a whole plethora of things that we no longer train for largely driven by the desire to keep training costs to an absolute minimum. This year it shows up as an ineffective response to MCAS. It also showed up this year in the SJ-100 accident as a lack of familiarity of alternate law flight control properties. In previous years, it was an ineffective response to loss of airspeed. Sometime in the next few, it will show up as something else.

That is reasonable in itself. I would add two observations:

1. This phenomenon is not limited to pilots from any particular system, on your account and supported by the UK CAA Notice
2. Irrespective of the lack of general training for Runaway Stab Trim or other unusual flight conditions, the unexpected activation of MCAS, and possible crew responses were UNTRAINABLE prior to either accident, and remain so (so far as we are aware)

You have asserted that false MCAS activation should be detected by crew as an instance of Runaway Stab Trim in a straightforward way. Others disagree.
Maninthebar is offline  
Old 8th Jul 2019, 15:22
  #1227 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,487
Likes: 0
Received 25 Likes on 12 Posts
#1234,
There is no evidence that changes in training are directly related to deliberate cost savings.
Recent changes may be the introduction of evidence based training, new requirements, schedules.

Supposition (betting) on some another operators training programmes - crew’s experiences, similarly are unsupported by fact. Linking these mistaken beliefs to accident outcome is covert blame.

Re other aircraft accidents, supposition that crews’ lack of familiarity, or knowledge of alternate control laws, is an unjustified slight on the operator and aircraft involved. Newspaper reports do not replace official accident investigation and reports.

Backwards looking to airspeed issues is classic hindsight bias, again without explanation.
Foresight, what comes next is much more difficult to manage, whatever ‘it’ is.
How might industry train for ‘it’.
safetypee is offline  
Old 8th Jul 2019, 15:31
  #1228 (permalink)  
 
Join Date: Jun 2008
Location: Cambridge UK
Posts: 198
Likes: 0
Received 13 Likes on 2 Posts
Originally Posted by MurphyWasRight
It is important to remember that the AoA miscompare alert was not an option.

It is a standard feature that Boeing knew was not working unless the optional AoA display was installed.

Boeing knew this for a significant amount of time -before- the Lion Air accident.
Had it been working on the penultimate Lion Air flight it might have led to fixing AoA sensor system before the accident flight.
Does this indicate that the MAX maintenance procedures failed to detect that this "standard" feature wasn't working?
Peter H is online now  
Old 8th Jul 2019, 15:34
  #1229 (permalink)  
Thread Starter
 
Join Date: Apr 2015
Location: Under the radar, over the rainbow
Posts: 844
Received 22 Likes on 12 Posts
Originally Posted by SteinarN
Very interesting.

So, lets say the pilots mismanages the speed, for example autothrottle disconnected without pilots noticing. Speed decayes, AOA increases, passes MCAS treshold, MCAS start to put in 9 sec of AND trim. So, how should the pilots react now? Should they hit trim cutout after 3 or 4 seconds? How can they know this is a genuine MCAS event and not trim runaway?

This problem clearly needs attention. Something along the lines of a MCAS active aural message, do not hit cutout?
The original question in this subthread:

Originally Posted by Bend alot
If MCAS will still have a maximum 9 second run time and a pilot should cut it in 3 seconds - what is the other 6 seconds of MCAS operation still required for?
As originally conceived, MCAS was intended (effectively) to "correct" pilot input. Since pilots were -- deliberately -- not told even of its existence, it seems pretty clear that it was never anticipated that pilots might need or want to "cut it" -- unless, of course, it was somehow recognized as runaway trim, despite the uncharacteristic behavior.

The awful engineering and terrible operations management associated with this mess are just breathtaking.
OldnGrounded is offline  
Old 8th Jul 2019, 15:39
  #1230 (permalink)  
 
Join Date: May 2019
Location: Somewhere over the rainbow...
Posts: 0
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Maninthebar

You have asserted that false MCAS activation should be detected by crew as an instance of Runaway Stab Trim in a straightforward way. Others disagree.
Actually what I have asserted, and continue to assert, is that 1) the Main Electric Trim controlled by the pilot’s yoke switch trumps every other electric trim input including MCAS, 2) and if the pilot does not like where the stab trim is or where it is going, then the solution is immediately available under his/her thumb.

Furthermore, this trim state is not some number read off some gauge - it is an always present tactile sensations. If the pilot is really hand-flying the aircraft, if the pilot is really trying to place the aircraft attitude in a specific place to achieve a specific result, then the pilot will feel the control pressures. If the control pressures are not what the pilot wants, then the pilot needs to do something about it. Once the aircraft is stabilized, then the crew has the opportunity to address the malfunction. We can disagree on whether the crew would recognize the Runaway Trim procedure as the best tool to handle the malfunction, but I’m pretty sure they would find their way to those cutout switches eventually. The key is to do so from a stabilized platform.

A good analogy to this situation would be attempting to ride a strong-willed horse who wants to go somewhere different than the rider. Just because you are holding the reins doesn’t mean you are in control of the horse. An experienced rider will keep a firm hold on those reins and make the horse go where he/she wants. An inexperienced rider will be driven around by the horse. What’s the solution? Give the rider more training in how to drive a stubborn horse.
yoko1 is offline  
Old 8th Jul 2019, 16:04
  #1231 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,487
Likes: 0
Received 25 Likes on 12 Posts
#1236,
The speculation about the time that MCAS could run in normal operation is misleading, without foundation.
Previous technical discussions identified a maximum time proportional to the required stick force required from tail trim to meet stability requirements. Most flight conditions would require less run-time to meet the need, where need is defined by an elevated AoA and required stick force; not stall - perpetuating the myth.
https://www.boeing.com/commercial/73...e-updates.page

Associating stall awareness, stick-shake, with an inability to determine that the trim operation of a previously undisclosed sub-system had malfunctioned is both illogical and misleading.
Stick-shake had been operating continuously since take off; erroneous MCAS trim only after the flaps were up - suggesting that the crew had a revised appreciation of speed, even if not all of the other distracting alerts.
Supposition that the abnormal trim operation should have been identified, not requiring xx time because yy; has no substantiating argument, or apparent awareness of the limitations of human performance.

Generic procedures for an approach to, and at stall, do not relate to ‘unloading’ the aircraft; a term synonymous with normal acceleration. The primary objective is to reduce AoA, nose down pitch control, which in many circumstances will reduce normal acceleration.
—————-
Re AoA Disagree
https://boeing.mediaroom.com/news-re...ts?item=130431

Re #1239, the implication is that the AoA Disagree system as configured, did not flag a comparison fault in the maintenance system.
safetypee is offline  
Old 8th Jul 2019, 16:09
  #1232 (permalink)  
568
 
Join Date: Aug 2002
Location: Castletown
Posts: 241
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by BedakSrewet
Since it was a brand new aircraft, and assuming that Boeing has a Jakarta based tech. rep , was he ever consulted after the aircraft landed from Den Pasar, and assisted in the trouble shooting ?
Boeing provide FSR's (field service representatives) for Companies introducing new aircraft.FSR's may be based permanently in Asia or be based at the Companies operating base for a period of time, to deal with technical issues.
568 is offline  
Old 8th Jul 2019, 16:17
  #1233 (permalink)  
 
Join Date: Jan 2008
Location: Reading, UK
Posts: 15,979
Received 301 Likes on 156 Posts
Originally Posted by 568
Boeing provide FSR's (field service representatives) for Companies introducing new aircraft.FSR's may be based permanently in Asia or be based at the Companies operating base for a period of time, to deal with technical issues.
Bear in mind also that, by the time of the Lion Air accident, they had been flying the Max for well over a year.
DaveReidUK is online now  
Old 8th Jul 2019, 16:30
  #1234 (permalink)  
 
Join Date: Dec 2002
Location: UK
Posts: 2,487
Likes: 0
Received 25 Likes on 12 Posts
#1241, Is an unusual description of trim - stick force and aircraft control.
Most professional pilots already have an adequate understanding of this and the various relationships, including speed - thrust, and configuration.

The primary objective of control is to achieve the required flightpath (not where the trim is going), using stick input to change pitch. When a stable condition is achieved, including speed, then trim can be applied to reduce stick force.

Differences in views about the ability to identify a trim runaway, perceive, comprehend, relate, and choose action (checklist) appear to be fundamental to the current delay in return to service.

See issues of trim awareness Loss of Control In-Flight - Flight Crew training U.K. CAA safety Notice http://publicapps.caa.co.uk/docs/33/...ice2019005.pdf
safetypee is offline  
Old 8th Jul 2019, 17:03
  #1235 (permalink)  
 
Join Date: Feb 2015
Location: The woods
Posts: 7
Likes: 0
Received 4 Likes on 3 Posts
Originally Posted by safetypee
#1241, Is an unusual description of trim - stick force and aircraft control.
Most professional pilots already have an adequate understanding of this and the various relationships, including speed - thrust, and configuration.

The primary objective of control is to achieve the required flightpath (not where the trim is going), using stick input to change pitch. When a stable condition is achieved, including speed, then trim can be applied to reduce stick force.

Differences in views about the ability to identify a trim runaway, perceive, comprehend, relate, and choose action (checklist) appear to be fundamental to the current delay in return to service.

See issues of trim awareness Loss of Control In-Flight - Flight Crew training U.K. CAA safety Notice http://publicapps.caa.co.uk/docs/33/...ice2019005.pdf
Quite right. That is normal piloting.
However if the trim should for any reason be so far out, that the elevator has too little authority to correct the pitch, it may be necessary to trim simultaneously with pulling back.
And right there, in an extreme case there can be a problem with overloading of the trim mechanism in the tail (irreversible worm drive/screw jack), which is the reason for the unloading or sea saw procedures previously promulgated in jet-upset recovery advice. And as we know unloading at treetop height - even if you know how to do it - puts you in the ground.
B

bill fly is offline  
Old 8th Jul 2019, 17:05
  #1236 (permalink)  
 
Join Date: Nov 2010
Age: 56
Posts: 981
Received 15 Likes on 8 Posts
Originally Posted by Bend alot
If MCAS will still have a maximum 9 second run time and a pilot should cut it in 3 seconds - what is the other 6 seconds of MCAS operation still required for?

The pilot will not like where it is going, as it is in an unfamiliar part of the flight envelope that it operates. So the pilot should cut MCAS when it is needed?
Great post.

You can't have it both ways: either the pilot is supposed to kill any trim running for over 3 seconds, so MCAS wouldn't achieve it's goal when it's required, or, pilots should let the trim run until it stops in case MCAS is correct and could end up too AND to correct.
If you expect the pilots to instantly know which one it is, you probably don't think we need EGPWS, TCAS, stall warning or overspeed warning either.......
hans brinker is offline  
Old 8th Jul 2019, 17:17
  #1237 (permalink)  
 
Join Date: May 2010
Location: Boston
Age: 73
Posts: 443
Likes: 0
Received 0 Likes on 0 Posts
Quote:
Originally Posted by MurphyWasRight
It is important to remember that the AoA miscompare alert was not an option.

It is a standard feature that Boeing knew was not working unless the optional AoA display was installed.

Boeing knew this for a significant amount of time -before- the Lion Air accident.
Had it been working on the penultimate Lion Air flight it might have led to fixing AoA sensor system before the accident flight.

Originally Posted by Peter H
Does this indicate that the MAX maintenance procedures failed to detect that this "standard" feature wasn't working?
Working from memory, since I don't have time to track original posts at the moment, I believe this was reported by Southwest and possibly others.
I don't think we know if this was discovered by Boeing testing before it was reported, possible it was missed since it does work if the optional AoA display is pressent.
It was deemed non critical and was scheduled to be fixed in a standard SW upgrade cycle.

It would actually be a bit unusual to notice that a disagree warning was not working unless there was a case where an AoA failure happened (bird strike/whatever) and someone noticed the lack of warning. A human factors factor is that it is usually harder to notice something missing than an an unexpected occurrence.

Since it is implemented in software I doubt it would be a line item in a maintenance check unlike the old days when a similar alarm might have relays and light bulbs in the mix there would be nothing specific to the AoA disagree alarm that could fail. With a big 'assuming' it worked at all of course.

Last edited by MurphyWasRight; 8th Jul 2019 at 17:19. Reason: typo
MurphyWasRight is offline  
Old 8th Jul 2019, 18:51
  #1238 (permalink)  
 
Join Date: Jul 2019
Location: itsthewater
Posts: 19
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by bill fly
Quite right. That is normal piloting.
However if the trim should for any reason be so far out, that the elevator has too little authority to correct the pitch, it may be necessary to trim simultaneously with pulling back.
And right there, in an extreme case there can be a problem with overloading of the trim mechanism in the tail (irreversible worm drive/screw jack), which is the reason for the unloading or sea saw procedures previously promulgated in jet-upset recovery advice. And as we know unloading at treetop height - even if you know how to do it - puts you in the ground.
B

Uhh- not too sure where you get the worm drive screw jack bit - but the 737 series ( and others 767,777, etc - use a double recirculating ball " NUT " on the course threaded screw jack. Which has about a 90 percent efficiency and can therefore be back driven absent a braking system. While the electric motor MIGHT be overloaded at the extreme position (AND) and high speed, it is more likely that the manual (trim wheel- handle system ) is simply unable to produce enough torque absent superman along with the large number of turns required per degree and all by ONE person.
walkon19 is offline  
Old 8th Jul 2019, 18:58
  #1239 (permalink)  
 
Join Date: Aug 2012
Location: Scotland
Age: 54
Posts: 282
Received 82 Likes on 23 Posts
Originally Posted by yoko1

Actually, it is much more of an observation than an accusation......

Certain posters frequently attack both my character and my motivations in response to the facts and reasoned arguments that I present......

I’m not the one catering to emotion and spreading hate here........

I may get frustrated with some posters reluctance to deal with the facts that are staring them in the face........
Maybe if you were to dial back on the self assuredness & the absolute fact statements (& number of which are only opinions!) you might get on better?

What sort of book of truth is it that you have that allows you to think you can dismiss the valid opinions & comments of other obviously knowledgable contributors with opinions that you are pedalling as fact?

A controlled handover in a steady state... protocol.... procedure..... "That would be a significant error."... Blah...Blah... "Why the Captain did not pick up on this struggle is a bit of a mystery".... I means seriously, are you reading some of this junk before you post it - or is your schedule just too busy?

I may just have found a new use for that scroll wheel.
Thrust Augmentation is offline  
Old 8th Jul 2019, 19:17
  #1240 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 62
Posts: 424
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MurphyWasRight
Working from memory, since I don't have time to track original posts at the moment, I believe this was reported by Southwest and possibly others.
I don't think we know if this was discovered by Boeing testing before it was reported, possible it was missed since it does work if the optional AoA display is pressent.
It was deemed non critical and was scheduled to be fixed in a standard SW upgrade cycle.

It would actually be a bit unusual to notice that a disagree warning was not working unless there was a case where an AoA failure happened (bird strike/whatever) and someone noticed the lack of warning. A human factors factor is that it is usually harder to notice something missing than an an unexpected occurrence.

Since it is implemented in software I doubt it would be a line item in a maintenance check unlike the old days when a similar alarm might have relays and light bulbs in the mix there would be nothing specific to the AoA disagree alarm that could fail. With a big 'assuming' it worked at all of course.
My understanding is that testing a single AOA sensor on the ground is rather difficult, and requires specialised calibration equipment.

Testing an AOA disagree circuit requires two simultaneous readings, and doing so seems far beyond the scope of any normal airline maintenance requirements.

It is very hard to prove a 'negative' result by testing, and IMO it is not at all surprising that the missing AOA disagree function wasn't picked up earlier.
GordonR_Cape is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.