Author Topic: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'  (Read 178350 times)

0 Members and 1 Guest are viewing this topic.

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • !
  • Posts: 2699
  • Country: tr
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #325 on: March 13, 2019, 10:28:48 pm »
The thing that strikes me about this video is that there is little roll activity -- the plane just lies straight into the swamp.  I also note the absence of any smoke suggesting the problem is unlikely to be engine or cargo fire related.

Here's some info:

The further a society drifts from truth, the more it will hate those who speak it.
 

Offline pilotchup

  • Supporter
  • ****
  • Posts: 27
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #326 on: March 13, 2019, 10:33:10 pm »
Man this thread has blown up into a long one, so I haven't read everything here yet.
I'm interested in this, as is the rest of the aviation and some of the engineering community I bet.
I am a instrument rated pilot.
One thing I can say, is that airline pilots rarely fly in the traditional sense anymore, because of all the automation and systems. Not good nor bad, it's just how the industry is now. The extent that airline pilots "fly" is more like babysitting a system, with some manual control often done during take-off and landings depending on the capabilities of the airport in use and systems onboard. When something goes wrong, the pilots look at each other and say "whys it doing that" instead of disengaging the autopilot (breaking the circuit breaker if it is a safety system that is not disconnected with autopilot), grabbing the controls, and fly the darn plane. I may be ignorant here not reading all the circumstances and evidence yet, but just my initial 2 cents.. carry on!
 

Online Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #327 on: March 13, 2019, 10:44:30 pm »
One thing that is new here is that numerous reports of 737 Max pilots to weirdness when autopilot is turned on but in the past we were told MCAS only activates when AP is off so perhaps there's other flight control software problems beyond MCAS or MCAS plays a role while autopilot is engaged as well.  Honestly, we can put to bed the idea that this isn't a fly-by-wire AC even if it's not fly-by-wire by the conventional definition -- if the computer can wrestle control in AP or when AP is off AND the only way to prevent that is to turn systems off the idea that this is a stick flown AC is laughable. 

So, in the past we were told MCAS only activated when being stick flown but the new data says the same or similar problems can also occur on AP.

The 737 (everything from the -100 to MAX10) is not a fly by wire aircraft.  The difference is how the control surfaces are connected to the control inputs.  In the case of all 737s it's done mechanically/hydraulically, and the stick directly controls the hydraulics.  That's not to say there's aren't systems like the AP or MCAS (stick shaker) that can't control the stick.

On the A320s, the input is purely a joystick into an electronic brain, aka fly-by-wire; which then actuates electro-hydraulic systems.

And once again this is a distinction without a difference.  I've worked on AC (USAF) and know a thing or two about them and, yes, there is a definition related to fly-by-wire and the 737 does not meet that definition -- however, if the computer can still override the pilots inputs to the stick this distinction is largely irrelevant.  The pertinent point is that a computer is controlling things and it makes little difference if the pilot has a direct connection to the control surfaces or if the computer is in between -- if the computer can wrestle control from the pilot the end effect is the same.  In fact, the idea that a pilot has direct control of the control surfaces on a 737 is laughable anyway as there's a hydraulic system in between -- not like AC where the yoke directly connects to the control surfaces with cables.  Look, we've been down this semantic road before and its tiring to continue the pretense that an AC like the 737 isn't controlled by computer even if a formal definition would argue it isn't.  It is a sure bet that most if not all the update Boeing plans is SOFTWARE.


Brian

Saying something is fly-by-wire when it's fly-by-hydraulic is like saying an electric water heater is the same as an gas one, just because they both accomplish the same goal. I agree that the problem likely is not in the method of transmitting force to the actuators, but that's no excuse for using a term describing that method to describe something else. If it's a software problem, call it a software or computer problem. If its a sensor problem, call it a sensor problem. Using the wrong term will get you push-back, as you've discovered multiple times in this thread already.
 

Offline boffin

  • Supporter
  • ****
  • Posts: 1027
  • Country: ca
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #328 on: March 13, 2019, 11:25:37 pm »
Not mentioned here, but there have been a few other similar airdata computer/autopilot plunge accidents (bad data=bad programmed response); such as

A330-200  Air France 447 crashed off the coast of Brazil
A330-300 Qantas 72 which diverted to Learmonth

 

Offline raptor1956

  • Frequent Contributor
  • **
  • Posts: 868
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #329 on: March 13, 2019, 11:27:23 pm »
One thing that is new here is that numerous reports of 737 Max pilots to weirdness when autopilot is turned on but in the past we were told MCAS only activates when AP is off so perhaps there's other flight control software problems beyond MCAS or MCAS plays a role while autopilot is engaged as well.  Honestly, we can put to bed the idea that this isn't a fly-by-wire AC even if it's not fly-by-wire by the conventional definition -- if the computer can wrestle control in AP or when AP is off AND the only way to prevent that is to turn systems off the idea that this is a stick flown AC is laughable. 

So, in the past we were told MCAS only activated when being stick flown but the new data says the same or similar problems can also occur on AP.

The 737 (everything from the -100 to MAX10) is not a fly by wire aircraft.  The difference is how the control surfaces are connected to the control inputs.  In the case of all 737s it's done mechanically/hydraulically, and the stick directly controls the hydraulics.  That's not to say there's aren't systems like the AP or MCAS (stick shaker) that can't control the stick.

On the A320s, the input is purely a joystick into an electronic brain, aka fly-by-wire; which then actuates electro-hydraulic systems.

And once again this is a distinction without a difference.  I've worked on AC (USAF) and know a thing or two about them and, yes, there is a definition related to fly-by-wire and the 737 does not meet that definition -- however, if the computer can still override the pilots inputs to the stick this distinction is largely irrelevant.  The pertinent point is that a computer is controlling things and it makes little difference if the pilot has a direct connection to the control surfaces or if the computer is in between -- if the computer can wrestle control from the pilot the end effect is the same.  In fact, the idea that a pilot has direct control of the control surfaces on a 737 is laughable anyway as there's a hydraulic system in between -- not like AC where the yoke directly connects to the control surfaces with cables.  Look, we've been down this semantic road before and its tiring to continue the pretense that an AC like the 737 isn't controlled by computer even if a formal definition would argue it isn't.  It is a sure bet that most if not all the update Boeing plans is SOFTWARE.


Brian

Saying something is fly-by-wire when it's fly-by-hydraulic is like saying an electric water heater is the same as an gas one, just because they both accomplish the same goal. I agree that the problem likely is not in the method of transmitting force to the actuators, but that's no excuse for using a term describing that method to describe something else. If it's a software problem, call it a software or computer problem. If its a sensor problem, call it a sensor problem. Using the wrong term will get you push-back, as you've discovered multiple times in this thread already.


In every post I've made on this topic I've made it clear that the technical description of fly-by-wire does not match what the 737 Max is, the point I'm making is that this technical difference is irrelevant given the fact that the computer is controlling the control surfaces.  One could argue that fly-by-wire is inaccurate anyway if in fact the wire is an optical fiber!


Brian
 

Offline edy

  • Super Contributor
  • ***
  • Posts: 2385
  • Country: ca
    • DevHackMod Channel
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #330 on: March 14, 2019, 01:32:48 am »
So if Boeing grounded all the planes based on some new data found from the flight recorders, are we to assume there is some software glitch they discovered with their flight control system? Something in the software that is making an improper calculation/correction and perhaps even exaggerating the problem under certain conditions (like a run-away effect)? I'm sure there are millions of lines of code in there and somewhere earlier in the thread I believe someone mentioned that they "simulated" the behaviour of earlier planes in the new MAX series so that pilots would feel like they were flying an older plane they were used to.

I'm sure some of the experienced pilots in this thread can shed some light on what exactly modern sophisticated passenger jet software does and how it is supposed to aid pilots. Other than instrumentation, how much is the software flying the plane and how much is it just informing the pilot? This may boil down to a philosophical question regarding autonomous computer-controlled transportation in general and the statistical argument about safety versus human comfort level. For example, if you say had fully computer-controlled aircraft (from take off to landing) and found that the safety improved by a factor of 2x, would people still be willing to let a computer fly them, or would they rather have a human behind the wheel even knowing statistically-speaking that it would double their chance of a fatal crash? Sure, the numbers tell us something but psychologically how ready are we to allow computers to take full control?
YouTube: www.devhackmod.com LBRY: https://lbry.tv/@winegaming:b Bandcamp Music Link
"Ye cannae change the laws of physics, captain" - Scotty
 

Offline tooki

  • Super Contributor
  • ***
  • Posts: 11343
  • Country: ch
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #331 on: March 14, 2019, 02:01:52 am »
Though auto-takeoff and autoland (in addition to the cruise autopilot) have actually existed for many decades, making it theoretically possible for an aircraft to have every phase of flight automated, this isn’t really done in practice.

Autopilot isn’t a single thing. It’s an umbrella term for dozens of discrete automation functions that can be used all at once, or none at all, or in any combination. So things like auto-navigation (automatic turns), auto level (maintain flight level), autothrottle (maintain speed), auto brake (on landing) etc. But tons more are automatic safeguards, like stall prevention.

In this case, they added another computer-controlled automatic adjustment to compensate for the different lift characteristics of the MAX, to make it behave like a classic 737, so that a pilot doesn’t have to be completely recertified for a different aircraft type, which would be needed if it handled differently. The problem is simply that they didn’t even mention this new system to pilots, never mind explain how to disable it when faulty sensors are causing it to make erroneous corrections, as happened with Lion Air and may have happened with Ethiopian.

Back to your question: no flight is ever under total computer control. Far from it. The pilots are coordinating the bazillions of different settings and modes, some manually, others through automation. The programming is critical (like, you must correctly enter the flight plan, so that auto nav knows when to turn). And a flight is never flown according only to pre-decided steps: air traffic control is directing them what to do, when to do it, etc. So for example, you may choose to use auto level at cruise, but you’re still in contact with air traffic control, who tell you which flight level to use, and you must enter that into the autopilot yourself, and ensure it’s in a mode that allows it to change flight level.

People mistakenly think that pilots don’t do anything. But in fact, aviation is extremely demanding, with tons of tasks that must be done without fail. And so offloading some of the workload to the autopilot makes total sense. But pilots also want to fly. They didn’t do all that training and spend years earning poverty wages as junior pilots only to then sit in the cockpit and do nothing. So often, they do many things manually that they could automate. And other things, like takeoffs and landings, are never done automated, even if the hardware and software is present. They’re there as backup systems if, for instance, a pilot is incapacitated and the copilot is also occupied with some emergency and needs to offload a task.
 

Offline tautech

  • Super Contributor
  • ***
  • Posts: 28147
  • Country: nz
  • Taupaki Technologies Ltd. Siglent Distributor NZ.
    • Taupaki Technologies Ltd.
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #332 on: March 14, 2019, 02:24:06 am »
Though auto-takeoff and autoland (in addition to the cruise autopilot) have actually existed for many decades, making it theoretically possible for an aircraft to have every phase of flight automated, this isn’t really done in practice.

Autopilot isn’t a single thing. It’s an umbrella term for dozens of discrete automation functions that can be used all at once, or none at all, or in any combination. So things like auto-navigation (automatic turns), auto level (maintain flight level), autothrottle (maintain speed), auto brake (on landing) etc. But tons more are automatic safeguards, like stall prevention.
Yep, it's called FMS; Flight Management System.
Usually engaged/disengaged just after takeoff and just before landing and depending on guidance systems and weather conditions disengagement might be as low as 10's of feet above the runway.
Avid Rabid Hobbyist
Siglent Youtube channel: https://www.youtube.com/@SiglentVideo/videos
 

Offline raptor1956

  • Frequent Contributor
  • **
  • Posts: 868
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #333 on: March 14, 2019, 03:31:29 am »
Not mentioned here, but there have been a few other similar airdata computer/autopilot plunge accidents (bad data=bad programmed response); such as

A330-200  Air France 447 crashed off the coast of Brazil
A330-300 Qantas 72 which diverted to Learmonth


One of the points the proponents of Boeing AC have made relative to Airbus is that Boeing AC permitted the pilots greater control over the plane whereas Airbus went full in with the all-seeing all-controlling control system.  What these latest accidents in Boeing planes suggests is that Boeing doesn't appear to have a leg to stand on in this respect -- at least not any more.  It's too soon to say of the Amazon plane that crashed last week was also due to the computer doing something the crew either didn't comprehend or couldn't fight.  If the 767 crash winds up being a similar computer control issue then Boeing could be toast and i'm 100% serious about that as far as commercial aviation is concerned.  The last 20-30 years has seen software come to dominate the tech world where hardware used to, but sometimes the software types don't know as much as they think they do or they think pilots are so stupid they need to be kept on a leash. 


Brian
 

Offline Homer J SimpsonTopic starter

  • Super Contributor
  • ***
  • Posts: 1222
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #334 on: March 14, 2019, 04:31:23 am »
 
The following users thanked this post: WattsThat

Offline james_s

  • Super Contributor
  • ***
  • Posts: 21611
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #335 on: March 14, 2019, 04:41:14 am »
One of the points the proponents of Boeing AC have made relative to Airbus is that Boeing AC permitted the pilots greater control over the plane whereas Airbus went full in with the all-seeing all-controlling control system.  What these latest accidents in Boeing planes suggests is that Boeing doesn't appear to have a leg to stand on in this respect -- at least not any more.  It's too soon to say of the Amazon plane that crashed last week was also due to the computer doing something the crew either didn't comprehend or couldn't fight.  If the 767 crash winds up being a similar computer control issue then Boeing could be toast and i'm 100% serious about that as far as commercial aviation is concerned.  The last 20-30 years has seen software come to dominate the tech world where hardware used to, but sometimes the software types don't know as much as they think they do or they think pilots are so stupid they need to be kept on a leash. 

A classic example of this is the Therac-25 incidents. https://web.stanford.edu/class/cs240/old/sp2014/readings/therac-25.pdf
 
The following users thanked this post: tooki

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #336 on: March 14, 2019, 09:01:18 am »
The last 20-30 years has seen software come to dominate the tech world where hardware used to, but sometimes the software types don't know as much as they think they do or they think pilots are so stupid they need to be kept on a leash.
This is nonsense. The "software guys" do not make the use cases, requirements and safety concerned issues.
There should be a product owner, (sub)system architects and system engineers that can translate the use cases to requirements and that should involve safety features including redundant systems and fallback scenario's.
Then there should be an elaborate testing phase where all systems and software routines are rigurously tested, esp the safety critical ones with high impact, they should be continuously tested in simulators.
For a modern airplane I can imagine this would take years. Why do you think the software update took so long? It needs to be restested the whole shabang.
In any good business this is in place esp. where safety and reliability is a major issue.

BTW many accidents in the past where hardware related, a bolt that was made from the wrong steel and broke off, a sensor that was cluttered, so to say software is worse than hardware.... they both can be cause for an accident.
 
The following users thanked this post: tooki, cgroen

Offline GeorgeOfTheJungle

  • Super Contributor
  • ***
  • !
  • Posts: 2699
  • Country: tr
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #337 on: March 14, 2019, 10:30:27 am »
Boeing Statement On 737 MAX Software Enhancement
http://aero-news.net/index.cfm?do=main.textpost&ID=78526990-A846-4929-8D11-9D3C0FDD9DE3
Quote
For the past several months and in the aftermath of Lion Air Flight 610, Boeing has been developing a flight control software enhancement for the 737 MAX, designed to make an already safe aircraft even safer. This includes updates to the Maneuvering Characteristics Augmentation System (MCAS) flight control law, pilot displays, operation manuals and crew training. The enhanced flight control law incorporates angle of attack (AOA) inputs, limits stabilizer trim commands in response to an erroneous angle of attack reading, and provides a limit to the stabilizer command in order to retain elevator authority.
[...]
Boeing’s 737 MAX Flight Crew Operations Manual (FCOM) already outlines an existing procedure to safely handle the unlikely event of erroneous data coming from an angle of attack (AOA) sensor. The pilot will always be able to override the flight control law using electric trim or manual trim. In addition, it can be controlled through the use of the existing runaway stabilizer procedure as reinforced in the Operations Manual Bulletin (OMB) issued on Nov. 6, 2018

Emergency Airworthiness Directive (AD) 2018-23-51 is sent to owners and operators of The Boeing Company Model 737-8 and -9 airplanes.
http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgad.nsf/0/83EC7F95F3E5BFBD8625833E0070A070?OpenDocument
Quote
This emergency AD was prompted by analysis performed by the manufacturer showing that if an erroneously high single angle of attack (AOA) sensor input is received by the flight control system, there is a potential for repeated nose-down trim commands of the horizontal stabilizer. This condition, if not addressed, could cause the flight crew to have difficulty controlling the airplane, and lead to excessive nose-down attitude, significant altitude loss, and possible impact with terrain.
[...]
Runaway Stabilizer
Disengage autopilot and control airplane pitch attitude with control column and main electric trim as required. If relaxing the column causes the trim to move, set stabilizer trim switches to CUTOUT. If runaway continues, hold the stabilizer trim wheel against rotation and trim the airplane manually.
Note: The 737-8/-9 uses a Flight Control Computer command of pitch trim to improve longitudinal handling characteristics. In the event of erroneous Angle of Attack (AOA) input, the pitch trim system can trim the stabilizer nose down in increments lasting up to 10 seconds.
In the event an uncommanded nose down stabilizer trim is experienced on the 737-8/-9, in conjunction with one or more of the indications or effects listed below, do the existing AFM Runaway Stabilizer procedure above, ensuring that the STAB TRIM CUTOUT switches are set to CUTOUT and stay in the CUTOUT position for the remainder of the flight.
An erroneous AOA input can cause some or all of the following indications and effects:
•   Continuous or intermittent stick shaker on the affected side only.
•   Minimum speed bar (red and black) on the affected side only.
•   Increasing nose down control forces.
•   IAS DISAGREE alert.
•   ALT DISAGREE alert.
•   AOA DISAGREE alert (if the option is installed).
•   FEEL DIFF PRESS light.
•   Autopilot may disengage.
•   Inability to engage autopilot.
Initially, higher control forces may be needed to overcome any stabilizer nose down trim already applied. Electric stabilizer trim can be used to neutralize control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT. Manual stabilizer trim can be used before and after the STAB TRIM CUTOUT switches are moved to CUTOUT.

« Last Edit: March 14, 2019, 10:32:29 am by GeorgeOfTheJungle »
The further a society drifts from truth, the more it will hate those who speak it.
 

Online langwadt

  • Super Contributor
  • ***
  • Posts: 4395
  • Country: dk
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #338 on: March 14, 2019, 02:42:20 pm »
Not mentioned here, but there have been a few other similar airdata computer/autopilot plunge accidents (bad data=bad programmed response); such as

A330-200  Air France 447 crashed off the coast of Brazil

I'm not sure you can blame that on the computer, the computer switched to alternate law when it didn't believe the sensor input,
leaving the controls to the pilots with the task of not crashing a perfectly functional plane flying at 38000 feet
 
The following users thanked this post: Tadas

Offline boffin

  • Supporter
  • ****
  • Posts: 1027
  • Country: ca
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #339 on: March 14, 2019, 03:23:03 pm »
Not mentioned here, but there have been a few other similar airdata computer/autopilot plunge accidents (bad data=bad programmed response); such as

A330-200  Air France 447 crashed off the coast of Brazil

I'm not sure you can blame that on the computer, the computer switched to alternate law when it didn't believe the sensor input,
leaving the controls to the pilots with the task of not crashing a perfectly functional plane flying at 38000 feet

And the pilots then did the same dumb thing that the auto-pilot appears to have done on the 737-MAX aircraft, pushed the nose down to gain speed....
 

Offline newbrain

  • Super Contributor
  • ***
  • Posts: 1714
  • Country: se
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #340 on: March 14, 2019, 03:57:04 pm »
And the pilots then did the same dumb thing that the auto-pilot appears to have done on the 737-MAX aircraft, pushed the nose down to gain speed....
Had to check, because that's not what I remembered:
In fact, they did the exact opposite. Had they put the nose down, instead of up, the plane would (probably) not have crashed.
From wikipedia:
Quote
From there until the end of the flight, the angle of attack never dropped below 35 degrees.

Having flown with Ethiopian airlines twice a month for several years, I was quite shocked by this disaster.
Nandemo wa shiranai wa yo, shitteru koto dake.
 

Offline boffin

  • Supporter
  • ****
  • Posts: 1027
  • Country: ca
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #341 on: March 14, 2019, 05:44:48 pm »
And the pilots then did the same dumb thing that the auto-pilot appears to have done on the 737-MAX aircraft, pushed the nose down to gain speed....
Had to check, because that's not what I remembered:
In fact, they did the exact opposite. Had they put the nose down, instead of up, the plane would (probably) not have crashed.
From wikipedia:
Quote
From there until the end of the flight, the angle of attack never dropped below 35 degrees.

Having flown with Ethiopian airlines twice a month for several years, I was quite shocked by this disaster.

Sorry, I got it backwards, but it was still incorrect airdata info that led to (in this case) pilots doing the wrong thing.  I do remember being told "If you lost all instruments - 75% power, keep it level - and it will fly"

 
The following users thanked this post: WattsThat

Offline jmelson

  • Super Contributor
  • ***
  • Posts: 2758
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #342 on: March 14, 2019, 07:46:39 pm »
And the pilots then did the same dumb thing that the auto-pilot appears to have done on the 737-MAX aircraft, pushed the nose down to gain speed....
NO NO!  That't the PROBLEM!  They SHOULD have pushed the stick forward, to exit the stall and get flying.  But, at least ONE pilot held the stick full back until they hit the water!  The other pilot was pushing full forward, which was what was needed to break out of the stall.  But, because these sticks are not coupled and do not have any force feedback, the pilots did not know what the other one was doing.

Something that may have confused them is the Airbus suppresses the stall warning below 70 knots airspeed, so you don't have the horn blaring during landing.
But, this caused the horn to start blaring each time they began to escape from the stall.  Somehow the one pilot believed that by pulling back on his stick and keeping the horn from sounding, he was ESCAPING from the stall.  Total lack of understanding the systems.

Jon
 

Offline Kjelt

  • Super Contributor
  • ***
  • Posts: 6459
  • Country: nl
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #343 on: March 14, 2019, 07:53:40 pm »
  Total lack of understanding the systems.
Or are the systems designed ambiguous ?
 

Offline Dundarave

  • Regular Contributor
  • *
  • Posts: 151
  • Country: ca
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #344 on: March 14, 2019, 11:50:37 pm »

Emergency Airworthiness Directive (AD) 2018-23-51 is sent to owners and operators of The Boeing Company Model 737-8 and -9 airplanes.
http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgad.nsf/0/83EC7F95F3E5BFBD8625833E0070A070?OpenDocument
This emergency AD was prompted by analysis performed by the manufacturer showing that if an erroneously high single angle of attack (AOA) sensor input is received by the flight control system, there is a potential for repeated nose-down trim commands of the horizontal stabilizer. This condition, if not addressed, could cause the flight crew to have difficulty controlling the airplane, and lead to excessive nose-down attitude, significant altitude loss, and possible impact with terrain.
[...]
Runaway Stabilizer
Disengage autopilot and control airplane pitch attitude with control column and main electric trim as required. If relaxing the column causes the trim to move, set stabilizer trim switches to CUTOUT. If runaway continues, hold the stabilizer trim wheel against rotation and trim the airplane manually.
Note: The 737-8/-9 uses a Flight Control Computer command of pitch trim to improve longitudinal handling characteristics. In the event of erroneous Angle of Attack (AOA) input, the pitch trim system can trim the stabilizer nose down in increments lasting up to 10 seconds.
In the event an uncommanded nose down stabilizer trim is experienced on the 737-8/-9, in conjunction with one or more of the indications or effects listed below, do the existing AFM Runaway Stabilizer procedure above, ensuring that the STAB TRIM CUTOUT switches are set to CUTOUT and stay in the CUTOUT position for the remainder of the flight.
An erroneous AOA input can cause some or all of the following indications and effects:
•   Continuous or intermittent stick shaker on the affected side only.
•   Minimum speed bar (red and black) on the affected side only.
•   Increasing nose down control forces.
•   IAS DISAGREE alert.
•   ALT DISAGREE alert.
•   AOA DISAGREE alert (if the option is installed).
•   FEEL DIFF PRESS light.
•   Autopilot may disengage.
•   Inability to engage autopilot.
Initially, higher control forces may be needed to overcome any stabilizer nose down trim already applied. Electric stabilizer trim can be used to neutralize control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT. Manual stabilizer trim can be used before and after the STAB TRIM CUTOUT switches are moved to CUTOUT.

That's quite the software "workaround" for when the aircraft hardware (i.e. the angle-of-attack sensors, etc.) goes whacko.  I wouldn't want to be reading those instructions for the first time while the plane is trying its damndest to dive nose-first with "possible impact with terrain".

I'd be thinking that the recovery maneuvers as described above would need to be a simulator-based exercise repeated until it was thoroughly drilled in.  No time for "is the aircraft doing that thing with the bad AOA sensor or is something else wrong?" discussions in the cockpit.
 
The following users thanked this post: GeorgeOfTheJungle

Online Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #345 on: March 15, 2019, 12:38:31 am »

Emergency Airworthiness Directive (AD) 2018-23-51 is sent to owners and operators of The Boeing Company Model 737-8 and -9 airplanes.
http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgad.nsf/0/83EC7F95F3E5BFBD8625833E0070A070?OpenDocument
This emergency AD was prompted by analysis performed by the manufacturer showing that if an erroneously high single angle of attack (AOA) sensor input is received by the flight control system, there is a potential for repeated nose-down trim commands of the horizontal stabilizer. This condition, if not addressed, could cause the flight crew to have difficulty controlling the airplane, and lead to excessive nose-down attitude, significant altitude loss, and possible impact with terrain.
[...]
Runaway Stabilizer
Disengage autopilot and control airplane pitch attitude with control column and main electric trim as required. If relaxing the column causes the trim to move, set stabilizer trim switches to CUTOUT. If runaway continues, hold the stabilizer trim wheel against rotation and trim the airplane manually.
Note: The 737-8/-9 uses a Flight Control Computer command of pitch trim to improve longitudinal handling characteristics. In the event of erroneous Angle of Attack (AOA) input, the pitch trim system can trim the stabilizer nose down in increments lasting up to 10 seconds.
In the event an uncommanded nose down stabilizer trim is experienced on the 737-8/-9, in conjunction with one or more of the indications or effects listed below, do the existing AFM Runaway Stabilizer procedure above, ensuring that the STAB TRIM CUTOUT switches are set to CUTOUT and stay in the CUTOUT position for the remainder of the flight.
An erroneous AOA input can cause some or all of the following indications and effects:
•   Continuous or intermittent stick shaker on the affected side only.
•   Minimum speed bar (red and black) on the affected side only.
•   Increasing nose down control forces.
•   IAS DISAGREE alert.
•   ALT DISAGREE alert.
•   AOA DISAGREE alert (if the option is installed).
•   FEEL DIFF PRESS light.
•   Autopilot may disengage.
•   Inability to engage autopilot.
Initially, higher control forces may be needed to overcome any stabilizer nose down trim already applied. Electric stabilizer trim can be used to neutralize control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT. Manual stabilizer trim can be used before and after the STAB TRIM CUTOUT switches are moved to CUTOUT.

That's quite the software "workaround" for when the aircraft hardware (i.e. the angle-of-attack sensors, etc.) goes whacko.  I wouldn't want to be reading those instructions for the first time while the plane is trying its damndest to dive nose-first with "possible impact with terrain".

I'd be thinking that the recovery maneuvers as described above would need to be a simulator-based exercise repeated until it was thoroughly drilled in.  No time for "is the aircraft doing that thing with the bad AOA sensor or is something else wrong?" discussions in the cockpit.

It's not a software workaround, it's a procedural workaround. The procedure itself is actually pretty simple (STAB TRIM CUTOUT switches to CUTOUT, adjust trim by hand wheel). Most of the above is explanation, not procedure.

The software workaround is what Boeing is currently developing.

Any affected pilot reading those instructions for the first time during an event has failed their duty as a pilot, and their employer has failed at training. Stuff like this is not optional reading material.
« Last Edit: March 15, 2019, 12:40:08 am by Nusa »
 

Online langwadt

  • Super Contributor
  • ***
  • Posts: 4395
  • Country: dk
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #346 on: March 15, 2019, 12:42:37 am »

Emergency Airworthiness Directive (AD) 2018-23-51 is sent to owners and operators of The Boeing Company Model 737-8 and -9 airplanes.
http://rgl.faa.gov/Regulatory_and_Guidance_Library/rgad.nsf/0/83EC7F95F3E5BFBD8625833E0070A070?OpenDocument
This emergency AD was prompted by analysis performed by the manufacturer showing that if an erroneously high single angle of attack (AOA) sensor input is received by the flight control system, there is a potential for repeated nose-down trim commands of the horizontal stabilizer. This condition, if not addressed, could cause the flight crew to have difficulty controlling the airplane, and lead to excessive nose-down attitude, significant altitude loss, and possible impact with terrain.
[...]
Runaway Stabilizer
Disengage autopilot and control airplane pitch attitude with control column and main electric trim as required. If relaxing the column causes the trim to move, set stabilizer trim switches to CUTOUT. If runaway continues, hold the stabilizer trim wheel against rotation and trim the airplane manually.
Note: The 737-8/-9 uses a Flight Control Computer command of pitch trim to improve longitudinal handling characteristics. In the event of erroneous Angle of Attack (AOA) input, the pitch trim system can trim the stabilizer nose down in increments lasting up to 10 seconds.
In the event an uncommanded nose down stabilizer trim is experienced on the 737-8/-9, in conjunction with one or more of the indications or effects listed below, do the existing AFM Runaway Stabilizer procedure above, ensuring that the STAB TRIM CUTOUT switches are set to CUTOUT and stay in the CUTOUT position for the remainder of the flight.
An erroneous AOA input can cause some or all of the following indications and effects:
•   Continuous or intermittent stick shaker on the affected side only.
•   Minimum speed bar (red and black) on the affected side only.
•   Increasing nose down control forces.
•   IAS DISAGREE alert.
•   ALT DISAGREE alert.
•   AOA DISAGREE alert (if the option is installed).
•   FEEL DIFF PRESS light.
•   Autopilot may disengage.
•   Inability to engage autopilot.
Initially, higher control forces may be needed to overcome any stabilizer nose down trim already applied. Electric stabilizer trim can be used to neutralize control column pitch forces before moving the STAB TRIM CUTOUT switches to CUTOUT. Manual stabilizer trim can be used before and after the STAB TRIM CUTOUT switches are moved to CUTOUT.

That's quite the software "workaround" for when the aircraft hardware (i.e. the angle-of-attack sensors, etc.) goes whacko.  I wouldn't want to be reading those instructions for the first time while the plane is trying its damndest to dive nose-first with "possible impact with terrain".

I'd be thinking that the recovery maneuvers as described above would need to be a simulator-based exercise repeated until it was thoroughly drilled in.  No time for "is the aircraft doing that thing with the bad AOA sensor or is something else wrong?" discussions in the cockpit.

I read somewhere that US airlines have a custom software for the Max that permanently displays the AOA for MCAS on the main display, I  guess so
the pilots can instantly see if the value makes sense

 

Offline Dundarave

  • Regular Contributor
  • *
  • Posts: 151
  • Country: ca
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #347 on: March 15, 2019, 01:15:19 am »

It's not a software workaround, it's a procedural workaround. The procedure itself is actually pretty simple (STAB TRIM CUTOUT switches to CUTOUT, adjust trim by hand wheel). Most of the above is explanation, not procedure.

The software workaround is what Boeing is currently developing.

Any affected pilot reading those instructions for the first time during an event has failed their duty as a pilot, and their employer has failed at training. Stuff like this is not optional reading material.

I suppose the type of "workaround" is a semantic argument, but I'd say that if mission-critical system software such as this can't correctly handle faulty inputs like a defective AOA sensor, and instead needs timely and critical human intervention to avoid a disaster, then there's something wrong with the software, and the efforts needed to do the job that software should be doing constitute a workaround.

I would think that what Boeing is currently developing would be software that does what it should, and not simply "workaround" what it has already written.
 

Online Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #348 on: March 15, 2019, 02:12:19 am »
A software workaround is when your software is changed to handle a failure in some other system.

So when hardware fails, whatever the software does to compensate is, by definition, a workaround -- something designed to work as well as possible until the failing system can be repaired. There are many other workarounds designed into avionics software, as there are many systems that have the potential to fail. This is one that just wasn't done right the first time around. And got through testing and review.
 

Offline raptor1956

  • Frequent Contributor
  • **
  • Posts: 868
  • Country: us
Re: Lion Air crash: Jakarta Boeing 737 'had prior instrument error'
« Reply #349 on: March 15, 2019, 02:17:41 am »
News reports suggest that part of the reason they were able to tie the two crashes together and point towards both crashes having a similar/same cause and before they had the FDR or CVR was the position of the horizontal stabilizer jack screw.  If the plane was forced down by MCAS or similar the jack screw would be wound up towards one end of travel -- the end forcing the plane down.  That, on top of the refined GPS altitude data pointing to a bucking bronco ride into the ground is pretty clear evidence that Lionair and this crash have a common cause.

The French now have the recorders and hopefully we'll get a preliminary readout on them tomorrow or soon and I would be very interested to know what the AOA, altitude and airspeed indications were and what differences, if any, one side had versus the other.


Brian
 
The following users thanked this post: GeorgeOfTheJungle


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf