Indonesian aircraft missing off Jakarta
Dave re # 1875, et al, AoA disagree alert; Reuters article.
The preliminary accident report indicates that maintenance had knowledge of AoA problems and took appropriate action; page 17 ‘AOA SIGNAL FAIL’ and ‘AOA SIGNAL OUT OF RANGE’ and then changed the vane unit (presumably the left one, due to the associated left bias in other air-data alerts.)
The Reuters argument appears to depend on the supposition that a ‘disagree’ alert can determine which of the two vanes is in error. A disagreement involves the comparison of inputs, i.e. determining a difference (split), but not the validity of any one value, (which one is correct).
The Flight and Maintenance log together with BITE can provide additional information - as reported.
There appears to be no better maintenance action available than that taken; fitting an additional cockpit alert is of little / no value to maintenance.
A further hobby-horse in the article is a pilots display of AoA.
This theme also is taken up with the report of Southwest retrofitting the display option, quoting their reasons as:-
“… guard against any erroneous sensor data that may activate the jet’s controversial stall protection system.”
“… a valuable supplemental crosscheck in the event there is an erroneous AoA signal present” **
The implication in this reasoning again implies that it is possible to determine which value of AoA is correct.
Further misunderstanding comes from the FAA AD checklist, which relates the failed vane to the highest AoA (thus fly with the lowest AoA value). This is only correct for a situation involving MCAS trim operation, highest AoA is incorrect.
It is equally possible to have an AoA disagree alert due to one vane under-reading where the high value is correct (also with disagree alerts in other systems). The crew have no means of knowing which display is correct. This is the same issue as above, ‘disagree’ involves comparison, not validity.
Other instrument systems which have comparator logic have a third system to resolve an errant input; ‘2 out of 3’ logic (airspeed, alt). You cannot expect a dual (single failure exposed) system to provide triple redundancy, which appears necessary for MCAS etc.
** https://theaircurrent.com/aviation-s...737-max-fleet/
The preliminary accident report indicates that maintenance had knowledge of AoA problems and took appropriate action; page 17 ‘AOA SIGNAL FAIL’ and ‘AOA SIGNAL OUT OF RANGE’ and then changed the vane unit (presumably the left one, due to the associated left bias in other air-data alerts.)
The Reuters argument appears to depend on the supposition that a ‘disagree’ alert can determine which of the two vanes is in error. A disagreement involves the comparison of inputs, i.e. determining a difference (split), but not the validity of any one value, (which one is correct).
The Flight and Maintenance log together with BITE can provide additional information - as reported.
There appears to be no better maintenance action available than that taken; fitting an additional cockpit alert is of little / no value to maintenance.
A further hobby-horse in the article is a pilots display of AoA.
This theme also is taken up with the report of Southwest retrofitting the display option, quoting their reasons as:-
“… guard against any erroneous sensor data that may activate the jet’s controversial stall protection system.”
“… a valuable supplemental crosscheck in the event there is an erroneous AoA signal present” **
The implication in this reasoning again implies that it is possible to determine which value of AoA is correct.
Further misunderstanding comes from the FAA AD checklist, which relates the failed vane to the highest AoA (thus fly with the lowest AoA value). This is only correct for a situation involving MCAS trim operation, highest AoA is incorrect.
It is equally possible to have an AoA disagree alert due to one vane under-reading where the high value is correct (also with disagree alerts in other systems). The crew have no means of knowing which display is correct. This is the same issue as above, ‘disagree’ involves comparison, not validity.
Other instrument systems which have comparator logic have a third system to resolve an errant input; ‘2 out of 3’ logic (airspeed, alt). You cannot expect a dual (single failure exposed) system to provide triple redundancy, which appears necessary for MCAS etc.
** https://theaircurrent.com/aviation-s...737-max-fleet/
Join Date: Aug 2005
Location: London
Posts: 78
Likes: 0
Received 0 Likes
on
0 Posts
No, they probably didn't, but the point is that they didn't need to. Continued forward pressure on the side stick would have been enough. Completely different to the many 737 LOC-Is (and some non FBW Airbus types, A300s and A310s) where the crews would certainly have been better off if they understood what the stab was doing, or in some cases actually using it.
Join Date: Jul 2014
Location: Germany
Posts: 344
Likes: 0
Received 0 Likes
on
0 Posts
A further hobby-horse in the article is a pilots display of AoA.
This theme also is taken up with the report of Southwest retrofitting the display option, quoting their reasons as:-
“… guard against any erroneous sensor data that may activate the jet’s controversial stall protection system.”
“… a valuable supplemental crosscheck in the event there is an erroneous AoA signal present” **
This theme also is taken up with the report of Southwest retrofitting the display option, quoting their reasons as:-
“… guard against any erroneous sensor data that may activate the jet’s controversial stall protection system.”
“… a valuable supplemental crosscheck in the event there is an erroneous AoA signal present” **
Also on takeoff roll crosschecking speeds AND AoA seems like a sensible idea.
I'm not sure any of that would have helped in this case. Not sure if the pilots in question ever had a simulator session with let's say a faulty stick shaker.
(Also announcing runaway trim before running the simulator training for that seems like a bad idea but that's my understanding of how simulator training is done)
Their obsession with going fast to the extent of overspeeding the flaps suggests to me they didn't even run the UAS checklist.
(edit: maybe more importantly no constant thrust was set but was changed frequently according to the FDR. this would only happen after the UAS checklist if they determined they could rely on the speed reading in some way. which begs the question why they were going that fast then)
Apart from that the AoA disagree display could have easier landed in the maintenance log filled out by the pilots.
Also the replacement of the AoA sensor introduced the problem in this case.
Before the replacement the captains AoA sensor signal was probably intermittent leading to altitude and speed being displayed as blank.
So with an according tech log entry maybe the AoA vanes would have been checked after the first flight where the AoA disagree was properly handled.
Just flushing all errors when the AoA vanes disagree seems like a maintenance action that is really strange.
Not sure what the maintenance manual says but it seems to me like that's a bad idea.
Last edited by wiedehopf; 1st Dec 2018 at 23:39.
Join Date: Jun 2014
Location: Chocolatetown
Age: 63
Posts: 83
Likes: 0
Received 0 Likes
on
0 Posts
I'm pretty sure they ran the Airspeed Unreliable Checklist since they flew to 5,000 ft and retracted flaps per the PI-QRH for UAS. Why they added Flaps 5 is a bit curious as is why they failed to realize that flaps quieted the stabilizer trimming for a bit. UAS is one thing they may have anticipated based on the AFML?
Last edited by climber314; 2nd Dec 2018 at 12:52.
Join Date: Oct 2017
Location: Edmonton
Posts: 2
Likes: 0
Received 0 Likes
on
0 Posts
Maybe it's time for ...
1) Less computers in airliners
2) Going back to the two (IMHO) more reliable pathways for training airline pilots both of which embed stick and rudder skills:
1) military
2) years and thousands of hours as flight instructors
I find myself wondering if Big B will survive this financially if it's proven they hid this system for a reason, and also why this model was able to be certified under the old type certificate.
As as a frequent passenger, very glad to see that WestJet has the AOA Disagree indicator.
2) Going back to the two (IMHO) more reliable pathways for training airline pilots both of which embed stick and rudder skills:
1) military
2) years and thousands of hours as flight instructors
I find myself wondering if Big B will survive this financially if it's proven they hid this system for a reason, and also why this model was able to be certified under the old type certificate.
As as a frequent passenger, very glad to see that WestJet has the AOA Disagree indicator.
Join Date: Jul 2007
Location: USA
Posts: 82
Likes: 0
Received 0 Likes
on
0 Posts
I think you will find that a stick pusher was part of the Trident design as early as 1964, some 8 years before the accident in Staines!
Boeing may have decided to increase the level of protection on the MAX series due to undesirable behaviour exhibited during flight testing, but may not have been keen to publicise the fact that their design had poor stalling qualities. Had Boeing been more upfront and required additional training and understanding of the system from pilots moving onto the type this may not have happened.
Most aircraft have their little quirks and as long as they are known about and understood by the pilots and engineers they can be mitigated.
I wonder if part of Boeings problem is a disconnect between head office and where the aircraft are built and test flown. Long way between Seattle and Chicago, seen the same situation causal in a fatal industrial accident, engineers worked in head office and not accessible physically on the distant plant site, other than by phone.
Join Date: Mar 2014
Location: WA STATE
Age: 78
Posts: 0
Likes: 0
Received 0 Likes
on
0 Posts
I wonder if part of Boeings problem is a disconnect between head office and where the aircraft are built and test flown. Long way between Seattle and Chicago, seen the same situation causal in a fatal industrial accident, engineers worked in head office and not accessible physically on the distant plant site, other than by phone.
As an Boeing engineer myself- the only time I got to go to a corner office in Seattle was to show a short film of a test I was running ( in seattle area ) for some Saturn 5 hardware ( early 1960's ) which destroyed the part involved - which then resulted in a major redesign by New Orleans types of that part generally in accordance with my suggestions.
Do you have any evidence for either of those statements, given that we haven't seen any FDR traces prior to the sensor replacement ?
re stick "nudgers" and stick pushers
Stick pushers are sometimes fitted when there are "stability issues approaching the stall" - in such a case, though, in order to fix such issues the pusher needs to be set to trigger BEFORE the adverse characteristics are manifest, with attendant likely adverse impact on stall speeds. A pusher can also sometimes be fitted when the characteristics AT the aerodynamic stall are acceptable, but the post-aerodynamic stall characteristics degrade so quickly that a "barrier" is required to prevent excursions into an unsafe zone. Bottom line, pushers can be fitted for a variety of reasons.
Regarding why no "nudger" was fitted and MCAS adopted instead: A "nudger" would operate on the elevator, not the stab. If the issue is insufficient elevator authority/power to dependably recover from a stall under worst case conditions, nudging the stick will achieve nothing. In such a case, the solution is either more elevator authority (major redesign) for "help the elevator with the stab". The latter seems to be the MCAS solution.
Stick pushers are sometimes fitted when there are "stability issues approaching the stall" - in such a case, though, in order to fix such issues the pusher needs to be set to trigger BEFORE the adverse characteristics are manifest, with attendant likely adverse impact on stall speeds. A pusher can also sometimes be fitted when the characteristics AT the aerodynamic stall are acceptable, but the post-aerodynamic stall characteristics degrade so quickly that a "barrier" is required to prevent excursions into an unsafe zone. Bottom line, pushers can be fitted for a variety of reasons.
Regarding why no "nudger" was fitted and MCAS adopted instead: A "nudger" would operate on the elevator, not the stab. If the issue is insufficient elevator authority/power to dependably recover from a stall under worst case conditions, nudging the stick will achieve nothing. In such a case, the solution is either more elevator authority (major redesign) for "help the elevator with the stab". The latter seems to be the MCAS solution.
The point about the stick nudger was they have been approved in the past for other aircraft types (including other Boeings). None of us know the the degree of instability of the 737MAX (if any, outside the STS speed range), but it is up to the manufacturer to present a fix to the certification authority. I would be very surprised if there was a high degree of instability considering that only one configuration seems problematic. Also, the stability may be marginal, but compliant, in all other stall approach cases. If a low authority stick nudger could be sufficient there is still nothing to stop the manufacturer presenting the MCAS system for approval. To me if a low authority device would have been adequate, I am not so sure about the engineering justification, other than that all the required equipment was in place, of using a hugely powerful control to achieve that end. However, from an airworthiness view it can still be a perfectly acceptable approach.
I have difficulty with the concept that the elevator authority was perfectly adequate for all stalls approaches except one which that requires tailplane trim assist. That implies a huge loss of pitch control authority. So I will stick with my guess that this is a claytons stick nudger solution for mild longitudinal instability, implemented in the quickest, cheapest, way possible and nothing to do with loss of longitudinal controllabillity.
Perhaps I should have said a "stability augmentation device", of which a stick nudger is but one type.
I have not previously heard the proposal of stick nudgers being used as a stall recovery device, or stall identification device, perhaps I have misunderstood.
Last edited by zzuf; 2nd Dec 2018 at 03:13.
Post AF447 Stall Tests
Please could we stop comparisons with AF447. Airbus control system (modified 'C*') is completely different to any Boeing, especially the conventional 737, and the Elevator and Stab effectively act as one (stab following up the elevator when needed). Amazingly the A330 had enough pitch down authority to unstall up to some horrendous A of A, and had the crew only persisted in a nose down demand they might have recovered, though the nose down attitude needed was probably beyond their imagination. Some very courageous AB test pilots tried it out well into post stall (20 degrees or so from memory) and lived to tell the tale.
They stopped the Trim actuation (MCAS) by stab cutout switches.
What is notable from the A-SHOR report is that the use of the Airspeed Unreliable and ALT DISAGREE NNCs is mentioned but not the use of the Runaway Stabilizer NNC.
It is pretty clear that the KNKT has a clear cut view on whether JT43 should have continued on to Jakarta after having to disable the electric trim and with a stick shaker active. I have not yet found one commercial pilot who would have pressed on under similar circumstances. From an admittedly small but highly experienced pool, all of them have made it clear that they would have returned to Denpasar. It was interesting that the JT43 PIC appears to have noted that none of the three NNCs he ran contained the instruction “Plan to land at the nearest suitable airport”.
I'd be interested to see what Lion Air company policy states on such matters.
Join Date: Nov 2004
Location: bkk
Posts: 285
Likes: 0
Received 0 Likes
on
0 Posts
Years ago on a widebody, just after a hangar visit, as we went through 10 degrees nose up on rotation, the stick shaker went off (due to a faulty/stuck AOA gauge).There were other issues as well but I can tell you the stick shaker is VERY distracting.It went off continuously until short final.How any pilot can decide to continue with a commercial flight like this is hard to understand.There is no need to be told "land ASAP etc" it is kind of obvious.This Boeing model did not have MCAS but once the AP was engaged a similar safety system activated, which thought we were stalling (due to false AOA ) and at quite high speed (280k) started to violently "pump" the elevator to try to control pitch above stall (alpha floor protection).This was a very confusing and scary experience and took the 3 guys in the FD by surprise.The FD experience was around 80 years that day.The upshot is that it starts to border on criminal negligence to keep flying un-airworthy airplanes with these kind of problems that are ALREADY known about.Something is very very wrong with a safety culture that doesnt stop this kind of thing.
Join Date: Nov 2018
Location: Jakarta
Posts: 20
Likes: 0
Received 0 Likes
on
0 Posts
Where #2 was
There is no mention, vague or otherwise, of the trim problem or the requirement to disable both electric trim systems in 'the [Aircraft Flight Maintenance] log'. There was a reference to the Speed Trim System '... running to the wrong direction, suspected because of speed difference ...' in an entry made through the Company's A-SHOR electronic reporting system. At this stage we don't know who got to see that report.
What is notable from the A-SHOR report is that the use of the Airspeed Unreliable and ALT DISAGREE NNCs is mentioned but not the use of the Runaway Stabilizer NNC.
It is pretty clear that the KNKT has a clear cut view on whether JT43 should have continued on to Jakarta after having to disable the electric trim and with a stick shaker active. I have not yet found one commercial pilot who would have pressed on under similar circumstances. From an admittedly small but highly experienced pool, all of them have made it clear that they would have returned to Denpasar. It was interesting that the JT43 PIC appears to have noted that none of the three NNCs he ran contained the instruction “Plan to land at the nearest suitable airport”.
I'd be interested to see what Lion Air company policy states on such matters.
There is no mention, vague or otherwise, of the trim problem or the requirement to disable both electric trim systems in 'the [Aircraft Flight Maintenance] log'. There was a reference to the Speed Trim System '... running to the wrong direction, suspected because of speed difference ...' in an entry made through the Company's A-SHOR electronic reporting system. At this stage we don't know who got to see that report.
What is notable from the A-SHOR report is that the use of the Airspeed Unreliable and ALT DISAGREE NNCs is mentioned but not the use of the Runaway Stabilizer NNC.
It is pretty clear that the KNKT has a clear cut view on whether JT43 should have continued on to Jakarta after having to disable the electric trim and with a stick shaker active. I have not yet found one commercial pilot who would have pressed on under similar circumstances. From an admittedly small but highly experienced pool, all of them have made it clear that they would have returned to Denpasar. It was interesting that the JT43 PIC appears to have noted that none of the three NNCs he ran contained the instruction “Plan to land at the nearest suitable airport”.
I'd be interested to see what Lion Air company policy states on such matters.
There;s two flight DPS to CGK that night. both have almost same time departure after delayed.
JT33 PK-LJJ
Departure: Scheduled: 18:15 (10:15 UTC) Actual: 22:12 (14:12 UTC) 237 min delayed
Arrival: Scheduled: 19:10 (12:10 UTC) Actual: 22:37 (15:37 UTC) 207 min delayed
JT43 PK-LQP
Departure:
Scheduled: 19:30 (11:30 UTC) Actual: 22:21 (14:21 UTC) 171 min delayed
Arrival: Scheduled: 20:20 (13:20 UTC) Actual: 22:56 (15:56 UTC) 156 min delayed
Since there's almost same time departure, there is some misinformation that JT33 was PK-LQP, on Instagram story @nindypermatas via website: belitung.tribunnews.com/2018/10/29/warganet-ini-bagikan-pengalaman-naik-lion-air-jt610-sebut-ada-masalah-pada-penerbangan-sebelumnya
I detected that she was on 18.15 plane --> JT33. Its not the PK-LQP one.
The true of the JT43 pax story was based on website: tribunnews.com/regional/2018/11/01/pengakuan-penumpang-lion-air-pk-lqp-denpasar-jakarta-semua-teriak-allahuakbar.
JT43 was late also. almost two hours said Supriyanto Sudarto, one of the pax.
the pax already inside the plane. however still some people surrounding the plane body. Those people looks busy and getting into the plane.
After flight there is some flight altitude problem, abrupt down and going up, down again, up again.
Then one of the (presumably) pilot take something from the briefcase.
Its already late long and night, if the plane divert to DPS, then they need to provide hotel to the pax. Too Costly maybe.
The PK-LQP was landed at DPS before as JT775
Scheduled Arrival 09:10 (01:10 UTC) Actual 10:10 (02:10 UTC)
Plenty time to fix AOA sensor
Source:
planefinder.net/data/aircraft/PK-LQP
flightera.net/flight_details/Lion+Air/JT43/WADD/2018-10-28
flightera.net/flight/Lion+Air/JT775?offset=20181031#flight_list
Last edited by Realbabilu; 2nd Dec 2018 at 17:53. Reason: JT63 --> JT43
1.6 Aircraft Information
1.6.1 General
Registration Mark : PK-LQP
...
Certificate of Airworthiness Issued : 15 August 2018
...
Time Since New : 895 hours 21 minutes
Cycles Since New : 443 cycles
1.6.1 General
Registration Mark : PK-LQP
...
Certificate of Airworthiness Issued : 15 August 2018
...
Time Since New : 895 hours 21 minutes
Cycles Since New : 443 cycles
Join Date: Sep 2017
Location: Europe
Posts: 1,674
Likes: 0
Received 0 Likes
on
0 Posts
Boeing appears to have kept very quiet about the MCAS for a reason, with barely a reference to it anywhere. Why, would it have jeopardized the grandfathering of type certification?
Somewhere buried is the reason to stay quiet and it has little to do with the pilots don't need to know, but it is perhaps something that the corporates preferred the regulator not know, at least prior to certification.
Originally Posted by MickG0105
I have not yet found one commercial pilot who would have pressed on under similar circumstances. From an admittedly small but highly experienced pool, all of them have made it clear that they would have returned to Denpasar.
Originally Posted by MickG
It was interesting that the JT43 PIC appears to have noted that none of the three NNCs he ran contained the instruction “Plan to land at the nearest suitable airport”.
I'd be interested to see what Lion Air company policy states on such matters.
I'd be interested to see what Lion Air company policy states on such matters.
Last edited by Capn Bloggs; 2nd Dec 2018 at 05:30. Reason: "no ILS" comment corrected from 27 to 09
Capn, I am not a commercial pilot but in my limited flying I tended to be very conservative. It was a bit of a joke for a while at my flying school that I declared a PAN to expedite a landing for an external metallic banging noise of unknown origin that only manifested itself in turns. Once on the ground I found that the culprit was the RHS seatbelt had been left hanging out through the closed door (courtesy of my instructor when he had disembarked). So, no prizes for guessing what I would have done.
You sure about that? I think there is.
I'm not expecting that company policy would cover every possible contingency, it's policy not procedure. And I most assuredly would not expect it to include mention of a system that only Boeing knew about at the time. What I would be interested in is whether the circumstances of the JT43 incident as it actually occured (rather than as it was reported) should have been the subject of an internal Lion Air safety report and/or an Air Safety Report to the regulator.
You sure about that? I think there is.
Let's be realistic. What could it say? Cover every possible contingency with a directive for the crew to follow? Of course not. It certainly wouldn't cover this particular scenario. It seems half the world didn't even know about MCAS, let alone an operator putting out directives about what pilots should do if it goes rogue.
Of the many things that JT63 was and wasn't, one of them was that it wasn't JT63, it was JT33, and another is that it wasn't the accident aircraft, PK-LQP. You're talking about a different flight and a different aircraft.
Last edited by MickG0105; 2nd Dec 2018 at 03:48. Reason: Additional info