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 4th Oct 2019, 13:11
  #2881 (permalink)  
 
Join Date: Jun 2009
Location: Bedford, UK
Age: 70
Posts: 1,320
Received 42 Likes on 26 Posts
Originally Posted by ST Dog
Hazard criticality is determined first. Then based on that you design at the appropriate assurance level.
But DALs don't have failure rates.

And the criticality probabilities are per flight hour, not per year. And that's per flight hour of the entire fleet/type, not a single aircraft.

How the individual probabilities for parts of a system/subsystem add up depends on the relationships, determined in the FTA (fault tree).

And FWIW, in a FTA software failures have a probability of 1. There's really no way to calculate the probability of a defect in software, just as there is no defect free software (of any reasonable complexity).
Thanks, yes per hour. I was thinking primarily of the vane mechanics. Will have to do a sum to see what that means per year. A 1 in 10 million chance it will go wrong in any given hour still seems a stretch.
Mr Optimistic is offline  
Old 4th Oct 2019, 16:05
  #2882 (permalink)  
 
Join Date: Jan 2008
Location: uk
Posts: 857
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Mr Optimistic
Thanks, yes per hour. I was thinking primarily of the vane mechanics. Will have to do a sum to see what that means per year. A 1 in 10 million chance it will go wrong in any given hour still seems a stretch.
G-VHOT incident report suggests around 1 in 100k hours for AOA sensor (vane + resolver) failure rate, I'm sure I've seen similar numbers elsewhere as well. You then need the probability of failure being fail-high (there are other failure modes), which I have no information on, and the probability that fail-high on one sensor will trigger MCAS, which will be somewhere around 1/2 (maybe more - left side seems more susceptible (ramp damage?) and left side is also more likely active input as always selected after power-off reset). I agree that getting to 10^-7 seems a stretch, maybe I'm missing something though.
infrequentflyer789 is offline  
Old 4th Oct 2019, 16:36
  #2883 (permalink)  
 
Join Date: Jul 2019
Location: Mass
Posts: 23
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by CurtainTwitcher
. . .
Having said all that, Boeing pushed the system too far. The $10 billion in industry costs for an EICAS model is now a line ball breakeven, and 346 people have lost their lives. . . .
There seems to be an assumption in the article and in the comments that a retrofitted EICAS would have been likely to have prevented these two accidents. How realistic is that assumption?

In JT610, the crew knew they had a malfunctioning stab trim system and they figured out they could keep the plane in the air using MET. They were focused on the stab trim problem but they just couldn't figure out that they needed to shut off electric trim. How would EICAS have fixed this? Presumably, it would have focused the crew on the principal problem (malfunctioning AoA) and suppressed some of the cascading warnings (e.g., stick shaker, IAS disagree) but the crew was already focused on solving the problem. Would EICAS have displayed a "Turn off the freakin electric trim" message?

In ET302, the crew knew they had a stab trim malfunction and they shut off electric trim, but didn't relieve control column forces first. They then knew they were out of trim, but were apparently confused about what to do about it. What would EICAS have added? Would it have displayed a "Don't turn the electric trim back on" message? Or maybe a message "If you turn electric trim back on, trim up fast and then turn it off"?

I understand the general assumption that less noise/flashing lights may lead to better decision making, particularly at the very beginning of an incident. But both crews were past that and were controlling the plane (even climbing in the case of ET302), albeit with significant effort. Isn't it hugely speculative to say that EICAS would have made any difference in the outcome?
Notanatp is offline  
Old 4th Oct 2019, 19:18
  #2884 (permalink)  
 
Join Date: Aug 2019
Location: Rocket City
Posts: 47
Received 0 Likes on 0 Posts
Originally Posted by Bend alot
Any details on these sim tests?
I can't find it now. But it was the results of the line pilots testing the new software.
I recall an AA pilot being quoted.

ST Dog is offline  
Old 4th Oct 2019, 20:10
  #2885 (permalink)  
 
Join Date: Aug 2019
Location: Rocket City
Posts: 47
Received 0 Likes on 0 Posts
Originally Posted by infrequentflyer789
I agree that getting to 10^-7 seems a stretch, maybe I'm missing something though.
Well for a major criticality they only needed 10E-5 for uncommanded trim AND.
Recalling that even full AND stab trim can be overcome by the elevator normally, then uncommanded trim is not catastrophic.

But that ignores the elevator blow down effect. What was the likelihood of uncommented MCAS trim at speeds where blow down reduces elevator authority?
What is the likelihood of blowdown at all? Is it outside approved speeds?

I didn't see anything that suggested Boeing got the failure rate of the AoA input grossly wrong.

The NTSB did see the analysis that was done. We won't ever see it.
But the NTSB only faulted some of the assumptions, not the analysis with those assumptions.
Like assumptions about pilots recognizing and acting on the uncommanded trim w/o consideration of all the secondary effects of a failure.
Like all the disparate alarms and warning the AoA failure caused that hampered recognition and reaction to the uncommanded trim.



ST Dog is offline  
Old 4th Oct 2019, 20:16
  #2886 (permalink)  
 
Join Date: Feb 2015
Location: The woods
Posts: 7
Likes: 0
Received 4 Likes on 3 Posts
Maths?

Seems the mathematicians and statisticians have taken hold of the plot.

If the thing can break - and if, when it breaks it can kill you - then it’s time either to fix it, or find something better, no matter what the probability equations say.
bill fly is offline  
Old 4th Oct 2019, 20:38
  #2887 (permalink)  
 
Join Date: Sep 2019
Location: leftcoast
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
corrected wrong link from picklefork to failure rate of AOA discussion

What failure rate do you use for ramp rash on left side ? especially with a 8 to 12 inch higher nose wheel ?

or bird strike ? or freeze ?

Last edited by Grebe; 4th Oct 2019 at 22:14. Reason: corrected wrong quote- link
Grebe is offline  
Old 4th Oct 2019, 21:57
  #2888 (permalink)  
 
Join Date: May 2008
Location: denmark
Posts: 10
Likes: 0
Received 3 Likes on 1 Post
Originally Posted by Notanatp
There seems to be an assumption in the article and in the comments that a retrofitted EICAS would have been likely to have prevented these two accidents. How realistic is that assumption?
How is the relation between EICAS and flight control systems with the capability of performing sensor fault insulation? (IAS, AOA, INS etc)?
I would assume that flight control systems able to perform fault insulation would require some kind of EICAS to provide status about what is going on inside the boxes. There are a limit on how much information that can be shown in the PFD.
And large aircraft with EICAS would normally also have fight control systems able to perform fault insulation and detect bit-flips.

Originally Posted by Notanatp
In JT610.....Would EICAS have displayed a "Turn off the freakin electric trim" message?
No, it should have displayed a message indicating something like "CPT AoA fault" or "AoA disagree", and a list of affected systems, and actions to be performed.
The fight control system should have stopped using faulty data.
HighWind is offline  
Old 4th Oct 2019, 22:11
  #2889 (permalink)  
 
Join Date: Sep 2019
Location: leftcoast
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by ST Dog
Well for a major criticality they only needed 10E-5 for uncommanded trim AND.
Recalling that even full AND stab trim can be overcome by the elevator normally, then uncommanded trim is not catastrophic.

But that ignores the elevator blow down effect. What was the likelihood of uncommented MCAS trim at speeds where blow down reduces elevator authority?
What is the likelihood of blowdown at all? Is it outside approved speeds?

I didn't see anything that suggested Boeing got the failure rate of the AoA input grossly wrong.

The NTSB did see the analysis that was done. We won't ever see it.
But the NTSB only faulted some of the assumptions, not the analysis with those assumptions.
Like assumptions about pilots recognizing and acting on the uncommanded trim w/o consideration of all the secondary effects of a failure.
Like all the disparate alarms and warning the AoA failure caused that hampered recognition and reaction to the uncommanded trim.
"But the NTSB only faulted some of the assumptions, not the analysis with those assumptions."

Oh - ever hear of GIGO ? garbage in garbage out ??
Grebe is offline  
Old 5th Oct 2019, 02:22
  #2890 (permalink)  
 
Join Date: Jul 2014
Location: Harbour Master Place
Posts: 662
Likes: 0
Received 0 Likes on 0 Posts
This comes back to the natural experiment that Boeing have performed between the MAX and the NG. Same AoA vane (IIRC same part number), and almost everything identical. So if you want to assume constant component failure rate, why are there no NG accidents related to flight controls or instrument failures that I can recall?

Boeing has been undone by this high degree of commonality. The difference in the accident rates with almost identical systems was too stunning to ignore. When investigation reveal there was a new MCAS system (previous rebuke for calling it software accepted in post #2459 ) there was nowhere for Boeing to hide. A single system change related to flight controls could be pin pointed as the single cause of the accidents. Sure there were plenty of other factors (training, experience, organisational, SOPs), but they were all held constant between the NG and the MAX.

Pilots have been dealing with AoA vane failures and faults since 1997 at the same rate as the MAX with the introduction of the NG without a fatality. MCAS is the only significant variable that has been implicated in these accidents. It simply could not be ignored.

I posted earlier, Boeing's October 2018 own data, that attested to the safety of the NG. Some 100+ million flights without flight control issues, and then 0.65 million MAX departures with two flight control fatal accidents. It was not the AoA vane failure that was the cause, it was the MCAS system, and only the MCAS system that the airworthiness authorities could not sweep under the rug and dismissed as pilot error and ignore because everything was so common with the NG.

Put yourself in the shoes of a senior CAA official looking at these figures comparing the max to the NG. Could you honestly let it fly in your airspace? Because I would be certain, this is the data that each CAA would have had compiled very quickly




As an aside, I keep thinking that this whole failure is looking more similar to the first computer accident, the Therac-25 accident X-Ray machine as documented by Nancy Leveson.

In that accident, a sequence of models of new software dose control were held constant, and a physical interlock preventing a lethal does of radiation to the patient was eventually removed. This interlock was from the previous non-computerised model where the operator could accidentally overdose a patient. What the company failed to understand, was the software was actually faulty right from the start, the interlock was saving patients for years without the company understanding the fault.

To me, the MAX accidents have a similar theme, engineering assumptions made on the basis of almost identical systems, with small iterative flawed changes that lay dormant until the accident sequences.
CurtainTwitcher is offline  
Old 5th Oct 2019, 05:09
  #2891 (permalink)  
 
Join Date: Dec 2015
Location: Cape Town, ZA
Age: 62
Posts: 424
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Notanatp
There seems to be an assumption in the article and in the comments that a retrofitted EICAS would have been likely to have prevented these two accidents. How realistic is that assumption?

In JT610, the crew knew they had a malfunctioning stab trim system and they figured out they could keep the plane in the air using MET. They were focused on the stab trim problem but they just couldn't figure out that they needed to shut off electric trim. How would EICAS have fixed this? Presumably, it would have focused the crew on the principal problem (malfunctioning AoA) and suppressed some of the cascading warnings (e.g., stick shaker, IAS disagree) but the crew was already focused on solving the problem. Would EICAS have displayed a "Turn off the freakin electric trim" message?

In ET302, the crew knew they had a stab trim malfunction and they shut off electric trim, but didn't relieve control column forces first. They then knew they were out of trim, but were apparently confused about what to do about it. What would EICAS have added? Would it have displayed a "Don't turn the electric trim back on" message? Or maybe a message "If you turn electric trim back on, trim up fast and then turn it off"?

I understand the general assumption that less noise/flashing lights may lead to better decision making, particularly at the very beginning of an incident. But both crews were past that and were controlling the plane (even climbing in the case of ET302), albeit with significant effort. Isn't it hugely speculative to say that EICAS would have made any difference in the outcome?
I could have a lengthy argument about your statement, but that would take us off topic. Regulators clearly think that pilot workload played a role in the crashes, and the MAX will remain grounded until training issues are dealt with: https://www.reuters.com/article/us-e...-idUSKBN1WJ2IU
Global regulators are looking at “startle factors” that can overwhelm pilots as they consider revised protocols for the Boeing 737 MAX, Nicholas Robinson, the head of civil aviation for Transport Canada, told Reuters on Friday.

Boeing Co’s fastest-selling jetliner, the 737 MAX, was grounded worldwide in March after two fatal crashes that killed a total of 346 people within five months.

Pilot overload appears to have played a role in both crashes, in which crews struggled to regain control of the airplane while a new flight control system repeatedly pushed the nose down amid a series of other audio and sensory alarms and alerts.

“What we need to do is ensure that the aircrew in the MAX are able to handle that environment,” Robinson said in an interview with Reuters.

Transport Canada is among a core group of regulators that is evaluating the requirements for the 737 MAX to fly again after a seven-month grounding.
Their decisions could lead to sweeping changes to pilot flight operating manuals and classroom instruction and even mandates for costly simulator training, industry sources have said. However, no training decisions can be made until Boeing submits software updates to the FAA for review and approval, Robinson said.

Transport Canada is closely aligned with the European Union Aviation Safety Agency on return to service demands and has also raised questions over the architecture behind the 737 MAX’s angle of attack system.
GordonR_Cape is offline  
Old 5th Oct 2019, 09:43
  #2892 (permalink)  
 
Join Date: Nov 2007
Location: dublin
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by bill fly


A lot of mistakes in there. If AP is on MCAS doesn’t operate anyway. Complete confusion as to the controls required to disable trim. Also the words “you just...” smack of some other machos posting on here.
The guy was totally incorrect as has been observed here.
You (just) observe the aircraft getting out of trim due to rotation of trim wheels, and if flying manually you observe that the plane is getting more and more nose heavy and hard to hold. Trim nose up using ELEC trim, all the way back into trim and turn off the STAB switches. Then trim manually.

That's the full package.
Don't think that guy had flown a 737 at at all.
Cheers
Yan
yanrair is offline  
Old 5th Oct 2019, 10:14
  #2893 (permalink)  
 
Join Date: Oct 2017
Location: Tent
Posts: 924
Received 33 Likes on 16 Posts
Originally Posted by yanrair
The guy was totally incorrect as has been observed here.
You (just) observe the aircraft getting out of trim due to rotation of trim wheels, and if flying manually you observe that the plane is getting more and more nose heavy and hard to hold. Trim nose up using ELEC trim, all the way back into trim and turn off the STAB switches. Then trim manually.

That's the full package.
Don't think that guy had flown a 737 at at all.
Cheers
Yan
Not exactly as per the AD issued - My bold, seemed more an option(al extra) not a mandate.
Bend alot is offline  
Old 5th Oct 2019, 14:25
  #2894 (permalink)  
 
Join Date: Nov 2000
Location: Canada
Posts: 604
Received 0 Likes on 0 Posts

Reuters Exclusive: Regulators weigh ‘startle factors’ for Boeing 737 MAX pilot training – Transport Canada executive

News provided by Reuters – link to full story

MONTREAL/CHICAGO (Reuters) – Global regulators are looking at “startle factors” that can overwhelm pilots as they consider revised protocols for the Boeing 737 MAX, Nicholas Robinson, the head of civil aviation for Transport Canada, told Reuters on Friday.
Longtimer is offline  
Old 5th Oct 2019, 15:51
  #2895 (permalink)  
 
Join Date: Jan 2004
Location: Far West Wessex
Posts: 2,640
Received 6 Likes on 4 Posts
Originally Posted by yanrair
The guy was totally incorrect as has been observed here.
You (just) observe the aircraft getting out of trim due to rotation of trim wheels, and if flying manually you observe that the plane is getting more and more nose heavy and hard to hold. Trim nose up using ELEC trim, all the way back into trim and turn off the STAB switches. Then trim manually.

That's the full package.
Don't think that guy had flown a 737 at at all.
Cheers
Yan
But that isn't the trim runaway procedure, is it? Because the FTFA-fundies have been banging away at "just use the trim runaway procedure" here and elsewhere. Granted, this is a great way to handle a spurious MCAS operation, but first you have to be trained to understand MCAS as something that is not a runaway.
LowObservable is offline  
Old 5th Oct 2019, 16:32
  #2896 (permalink)  
 
Join Date: Apr 2019
Location: Up there...
Posts: 9
Likes: 0
Received 0 Likes on 0 Posts
Every week that passes we learn more information about the flawed & corrupt certification process for the 737-MAX, a plane that burns less fuel at the expense of safety.

Boeing prioritizes profits above safety, the FAA is infiltrated by Boeing, and the eventual loss of life has been deemed acceptable.

Boeing's PR machine still tries to disseminate the "just fly the aircraft" mantra in many ways to try and divert attention from a flawed 737-MAX design, while trying to imbue pilots with some sort of blame.

Last edited by Byros; 5th Oct 2019 at 20:38.
Byros is offline  
Old 5th Oct 2019, 16:37
  #2897 (permalink)  
Thread Starter
 
Join Date: Apr 2015
Location: Under the radar, over the rainbow
Posts: 852
Received 27 Likes on 16 Posts
Originally Posted by Longtimer
Reuters Exclusive: Regulators weigh ‘startle factors’ for Boeing 737 MAX pilot training – Transport Canada executive
From the article:

Their decisions could lead to sweeping changes to pilot flight operating manuals and classroom instruction and even mandates for costly simulator training, industry sources have said.

However, no training decisions can be made until Boeing submits software updates to the FAA for review and approval, Robinson said.
So, as of yesterday (if Reuters is correct) the software updates had not been submitted. And, if those "sweeping changes" are to be completed before the MAX flies in revenue service again . . .
OldnGrounded is offline  
Old 5th Oct 2019, 16:45
  #2898 (permalink)  
 
Join Date: Nov 2004
Location: UK
Posts: 582
Received 83 Likes on 27 Posts
Originally Posted by LowObservable
But that isn't the trim runaway procedure, is it? Because the FTFA-fundies have been banging away at "just use the trim runaway procedure" here and elsewhere. Granted, this is a great way to handle a spurious MCAS operation, but first you have to be trained to understand MCAS as something that is not a runaway.

Very true. And also you really need to be 'trained' to know that MCAS even exists (read "informed by Boeing of the very existence of MCAS")

What yanrair has effectively done is to invent his / her own procedure, to deal with a serious problem, with the benefit of 20:20 hindsight. How very convenient. How dreadfully inconvenient there was no knowledge (ensured by Boeing) of the existence of MCAS at the time of the 2 terrible crashes with hundreds of fatalities.

The procedure yanrair with perfect hindsight would love everybody to have used (runaway trim) was neither indicated in its usual form, nor was it the specified procedure to deal with failed MCAS, for no better reason than there NEVER HAS BEEN any procedure for dealing with the secret MCAS system that Boeing didn't want pilots to know about. Now, faced with the 'inconvenience' that even the frequently-touted runaway trim procedure would not have done what he /she has always it to do for rogue MCAS, they have gone off on another tangential 'test pilot' direction to invent and specify a revised procedure, based on the runaway trim, tailored and modified to suit their goals of 'knowing how they would easily have dealt with a rogue MCAS', all with the benefit of hindsight and many months after the tragic events, with all that time to think it all through.

Sadly, the affected crews didn't have any of that privileged knowledge, and they certainly didn't have months to consider all the possibilities to invent their own procedure to deal with a problem caused by a rogue system that was completely unknown, on the spot.

As I said, how very convenient for yanrair, but how highly inconvenient for the affected crews and their unfortunate passengers. Obviously, yanrair should have been there to help, with the benefit of all their expert knowledge of the future, which presumably would require some extraordinary psychic powers. What an amazing claim.

Last edited by pilotmike; 5th Oct 2019 at 16:57.
pilotmike is offline  
Old 5th Oct 2019, 18:02
  #2899 (permalink)  
 
Join Date: Nov 2007
Location: dublin
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Bend alot
Not exactly as per the AD issued - My bold, seemed more an option(al extra) not a mandate.
Hi Bend a lot.

The AD only says go to RUNAWAY STAB PROCEDURE if you have any undesired stab trim. I agree it does not specifically say "trim back into trim if you are out of trim" because, I assume they believe that you have already tried that. Boeing are being accused currently of expecting pilots to know how to fly in difficult situations. Well, I can only say that I have always believed that that is why they sit where they do.

The QRH STAB TRIM runaway does not specifically say to "trim back to normal trim" because that is what you do when you get out of trim and it is what I was taught, and generations of pilots before and after me.
It does say TRIM USING MAIN ELECTRIC TRIM AS NEEDED which means back into trim. It cannot really mean anything else.
Moreover, the QRH is a synopsis of procedures most of which are covered during training using the Boeing Flight Crew Training Manual and the Supplementary Procedures which are much more extensive in giving advice on how to cope with abnormal situations. You are supposed to have digested and retained the information therein. And practised it in the sim. until your are sick of PM calling "Runaway Stabiliser" and you can do it with your eyes closed.


And here we have the problem, as I and many others have seen it from the start. Training is not covering such matters in anything like the detail they used to. And that applies from Flight School right through MPL and type rating and annual and bi-annual check rides on the sim.
On a Piper Cub if the trim gets heavy, you trim it out. It is instinctive. But not if you believe that the automatics do all that fancy stuff for you and you keep trying to engage the autopilot when the thing to do is just fly the plane before it slowly and surely heads downhill until it becomes unrecoverable. I am not saying this is what happened in the two recent crashes. I am saying that given a situation like this, it IS covered in the manuals and it should be recoverable - and was the day before the first fatal crash.

Like most of us I am very keen to see how the Final Report handles the fact that the day before, on October 18th, the third pilot who did not even work for Lionair managed to save the day. Perhaps he was Chuck Yeager in disguise. Reincarnated. Or just a guy who worked for an outfit that taught him properly? Or inspired? We don't know because nobody is talking about it.
Or will that be left out because it is simply not possible to include such a story in a report written by the national AAIU of the country where a major national carrier is involved and it happened at your main base in your own country. A bit like the Final Report on Concorde.

We know that the excellent crew of QF 32 saved the day with first class airmanship and training, and they had an extra flight crew member who had been involved in the introduction of the plane to QF. An above average crew I would say even for QF which is itself a first class airline with an enviable safety record.
The fault was a dodgy fuel pipe / oil pipe - doesn't matter. It was something that was catastrophic and shouldn't have happened, just like MCAS. It was way outside the box and produced something over 50 separate faults which all had to be prioritised and dealt with. Talk about "startle factor" but these guys just got on with it. Not because they were super human but because the had been trained.
Imagine, hypothetically a month later another carrier flying the same plane has the same fault and they crash. The focus would by massively on why the first one didn't and the second one did, with the self same fault, and especially since the crew of the second plane had read all the reports about what caused the first one, and had been forewarned?
Don't see that happening at the moment and if any reader has info. on that please do let us have it because I have genuinely missed it.
Cheers
Yan

yanrair is offline  
Old 5th Oct 2019, 18:23
  #2900 (permalink)  
 
Join Date: Nov 2007
Location: dublin
Posts: 2
Likes: 0
Received 0 Likes on 0 Posts
Hi Pilot Mike
QUOTE
As I said, how very convenient for yanrair, but how highly inconvenient for the affected crews and their unfortunate passengers. Obviously, yanrair should have been there to help, with the benefit of all their expert knowledge of the future, which presumably would require some extraordinary psychic powers. What an amazing claim.UNQUOTE
See my reply further down to Bend a lot where I cover your issues. Except the one about the secret MCAS. Yes it was a secret, but it is only one of several electrical inputs to the STAB motor such as STS, Mach Trim, Auto trim, and electric trim. The RESULT of any of this including MCAS is that the stab moves in an uncommanded manner and puts the plane out of trim. The pilot does not need to know which system is causing the problem. That is the point. There is one procedure which covers any uncommanded trim and that is the RUNAWAY STAB PROCEDURE and the AD makes this very clear.
So taking your point, there is possibly some excuse for the first crew not knowing what to do but in the second after the AD and presumably lots of training input from the airline?
Yan
yanrair 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.