Tristarsteve
Posts: 3588
Joined: Tue Nov 22, 2005 11:04 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 12:41 pm

In the case of a faulty ADIRU #1 or #2, is it also possible the pilots failed to select ADIRU #3 (which could have given them an accurate air speed)?


Does the Max have an ADIRU3?
Just asking as old B737 only had 2 ADIRU, not 3.
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 12:44 pm

sgrow787 wrote:
I'll ask to see if I can get access to their flight simulator lab. I'm not a pilot but I know one of the Max project engineers who is. Any ideas any of you want me to pass on (besides me simulating erratic or failed airspeed and AOA sensor data)? I'll definitely need the FR24 data for this.

Can you simulate a climb that fit the zeke data with a faulty left AoA that produce an offset of +20 degs from the right AoA value ? Does this unique modification plausibly explain all the mess in case the STAB TRIM CUTOFF is not used ?
 
AirbusA370
Posts: 53
Joined: Sat Dec 06, 2008 9:14 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 12:59 pm

Is the Elevator Feel Computer also part of the equation? It will increase the nose down pressure on the controls when it thinks there is a stall. Can the pilots override it easily?

So we have
- Stick shaker
- MCAS trimming nose down constantly
- Elevator Feel Computer pushing the controls
and two pilots trying the opposite...
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 1:42 pm

mandala499 wrote:
I still somehow suspect that this problem could have been from an ADR error. I mean the erroneous left AOA moved up and down like the right AOA, just 20 degs apart. ADR or ADM problems here I suspect.

Thinking again about this as an electronic engineer. It's plausible that a faulty electronic component in the analog front end of the AoA sensor could introduce a constant offset to the signal. In don't know for this specific ADR, but I have see failure like this before on others industrial systems. That said, on critical systems, care is done to compensate and/or detect this kind of fault mode. Could be an unexpected failure mode.
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 2:30 pm

PixelFlight wrote:
mandala499 wrote:
I still somehow suspect that this problem could have been from an ADR error. I mean the erroneous left AOA moved up and down like the right AOA, just 20 degs apart. ADR or ADM problems here I suspect.

Thinking again about this as an electronic engineer. It's plausible that a faulty electronic component in the analog front end of the AoA sensor could introduce a constant offset to the signal. In don't know for this specific ADR, but I have see failure like this before on others industrial systems. That said, on critical systems, care is done to compensate and/or detect this kind of fault mode. Could be an unexpected failure mode.


Now I have see the FDR graph on http://avherald.com/files/lionair_b38m_ ... t_data.pdf and surprise: it's not a constant offset. What catch my eyes was that the offset of the AoA goes in the opposite direction when the system was powered up. Then, within less than 5 minutes, the left AoA offset increased in two moments while the right AoA stayed flat, all that a couple of minutes before takeoff. The right AoA values look like what can be expected.

Update: At the moment when the computed air speed start to increase (takeoff acceleration), the AoA reacted in opposition: the left AoA decreased while the right AoA increased by about the same amount.

Update 2: the ANGLEOFATTACKINDICATEDLFDR on the first graph don't match the ANGLEOFATTACKINDICATEDLFDR on the last graph ! On the last one, the offset of the left AOA start with 10 degs, goes down slowly, then have a big negative spike to -10 degs and finally take the +20 degs offset at the start of the takeoff acceleration. Disturbing. How it that possible ? Well, according to the timestamp, it look like the last graph is from the previous flight.
Last edited by PixelFlight on Fri Nov 23, 2018 2:55 pm, edited 2 times in total.
 
joeblow10
Posts: 193
Joined: Mon Jun 04, 2018 11:58 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 2:36 pm

The way AvHerald and the authorities are wording it... it almost sounds like the pilots on the prior flight with problems did turn off the MCAS since the automatic trim never activated
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 3:06 pm

PixelFlight wrote:


The flight path agrees closely to what I was able to generate through filtering the ADS-B data.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 3:11 pm

zeke wrote:
PixelFlight wrote:


The flight path agrees closely to what I was able to generate through filtering the ADS-B data.

Yes, as expected. Congratulation for your work.
 
fsabo
Posts: 197
Joined: Fri Dec 04, 2015 8:41 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 3:23 pm

PixelFlight wrote:
mandala499 wrote:
I still somehow suspect that this problem could have been from an ADR error. I mean the erroneous left AOA moved up and down like the right AOA, just 20 degs apart. ADR or ADM problems here I suspect.

Thinking again about this as an electronic engineer. It's plausible that a faulty electronic component in the analog front end of the AoA sensor could introduce a constant offset to the signal. In don't know for this specific ADR, but I have see failure like this before on others industrial systems. That said, on critical systems, care is done to compensate and/or detect this kind of fault mode. Could be an unexpected failure mode.


There are very simple ways to detect errors in the analog front end or A/D converter. If boeing's design was fooled by a bad op-amp or A/D converter then that is truly a shitty design. I cannot believe they would make such a fundamental error.
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 4:00 pm

fsabo wrote:
PixelFlight wrote:
mandala499 wrote:
I still somehow suspect that this problem could have been from an ADR error. I mean the erroneous left AOA moved up and down like the right AOA, just 20 degs apart. ADR or ADM problems here I suspect.

Thinking again about this as an electronic engineer. It's plausible that a faulty electronic component in the analog front end of the AoA sensor could introduce a constant offset to the signal. In don't know for this specific ADR, but I have see failure like this before on others industrial systems. That said, on critical systems, care is done to compensate and/or detect this kind of fault mode. Could be an unexpected failure mode.


There are very simple ways to detect errors in the analog front end or A/D converter. If boeing's design was fooled by a bad op-amp or A/D converter then that is truly a shitty design. I cannot believe they would make such a fundamental error.

Agree. Now with the FDR graph in view, I realize that the AoA sensors could maybe move in any direction when the aircraft was taxiing because of the vibration (to be confirmed). But if correct, why only the left AoA sensor do this while the right AoA stay far more stable in the same conditions ?
 
N47
Posts: 43
Joined: Tue Sep 26, 2017 7:38 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 4:42 pm

PixelFlight wrote:
fsabo wrote:
PixelFlight wrote:
Thinking again about this as an electronic engineer. It's plausible that a faulty electronic component in the analog front end of the AoA sensor could introduce a constant offset to the signal. In don't know for this specific ADR, but I have see failure like this before on others industrial systems. That said, on critical systems, care is done to compensate and/or detect this kind of fault mode. Could be an unexpected failure mode.


There are very simple ways to detect errors in the analog front end or A/D converter. If boeing's design was fooled by a bad op-amp or A/D converter then that is truly a shitty design. I cannot believe they would make such a fundamental error.

Agree. Now with the FDR graph in view, I realize that the AoA sensors could maybe move in any direction when the aircraft was taxiing because of the vibration (to be confirmed). But if correct, why only the left AoA sensor do this while the right AoA stay far more stable in the same conditions ?


Being that the AoA is measuring an angle, perhaps the reference angle was offset either mechanically or electronically. So the two probes would be physically aligned with the wind but because one was aligned with the fuselage incorrectly (mechanically) or was sending the wrong voltage (electronically) they had two different readings.

What is surprising is how the previous flight operated with the stick shaker on for the entire flight, and it came on very shortly after take-off. As i'm not an airline transport pilot, i would like to know what is the usual SOP for stick shaker coming on and not going away? Do you just continue to fly all the way up to cruise and not have any coffee for that flight or do you try to land ASAP?
 
F9Animal
Posts: 4219
Joined: Thu Dec 16, 2004 7:13 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 5:10 pm

Now that we seem to have a general idea on the possible cause of this.... What will likely take place to ensure it doesnt happen again? Also, Lion has a really banged up image. Can we expect to see the flight crew blamed here?

As for Lion Air, is the future looking bleak?
I Am A Different Animal!!
 
PlanesNTrains
Posts: 9527
Joined: Tue Feb 01, 2005 4:19 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 5:16 pm

F9Animal wrote:
Now that we seem to have a general idea on the possible cause of this.... What will likely take place to ensure it doesnt happen again? Also, Lion has a really banged up image. Can we expect to see the flight crew blamed here?

As for Lion Air, is the future looking bleak?


1. I guess a likely software “fix” of some sort by Boeing.
2. The flight crew may bear some blame depending on where the investigation takes us.
3. Lion Air will likely be seen partly as a victim in this unless the investigation reveals some Skelton’s in their closet or poor airmanship.

I just think there’s too much going on that it’s hard to see the outcome clearly at this stage. YMMV.
-Dave


MAX’d out on MAX threads. If you are starting a thread, and it’s about the MAX - stop. There’s already a thread that covers it.
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Re:

Fri Nov 23, 2018 5:18 pm

fsabo wrote:
There are two AoA sensors and even though they disagreed the system picked one and decided to belive it. That algorithm was a significant factor in this crash. Perhaps if the two sensors don't agree then don't believe either one is the better option.

This is exactly what I think too: the sensors/ADR can fail, no matter what (the failure mode of the sensor/ADR still matter), the problem is the algorithm of the FC to safely continue the flight in case of a such failure. There are a lot of others values that the FC can use to predict the probability of the flight dynamic state. It must use that prediction to filter out incoherent input value and rely on the others values. This will of course degrade the probability curve, reducing the safest flight domain, but still allow to flight. Unfortunately today civil FC still use dangerous obsolete voting algorithm that blindly trust a single value, or an average value in the best case.
 
sgrow787
Posts: 165
Joined: Fri May 16, 2014 8:12 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 5:36 pm

PixelFlight wrote:
Can you simulate a climb that fit the zeke data with a faulty left AoA that produce an offset of +20 degs from the right AoA value ? Does this unique modification plausibly explain all the mess in case the STAB TRIM CUTOFF is not used ?


Due to the holidays this might take me a couple of weeks depending on his schedule and that of the lab's. It's also possible the MCAS sim module is not active in the 737 simulation, or plain doesn't exist. Makes sense that a flight condition where the plane is climbing and falling erratically, with pilot yanking back on the stick with all his force, is not going to be able to use the headup display, which is a slab of special glass called a combiner that has a specific eye relief window. That said, the whole idea behind the HGS is to allow the pilot to look out the front window instead of down into the cockpit, usually for landings and high traffic situations, but also for maybe if I'm heading straight for the ground.

Because this thread is 32 pages long and over 1500 messages, the only method I have to find Zeke's FR24 data is to sniff his profile and search his posts. Hmmm..

Update: Just noticed PixelFlight's update on the flight data ..."Now I have see the FDR graph on http://avherald.com/files/lionair_b38m_ ... t_data.pdf ". I'll take a look, thanks.
Last edited by sgrow787 on Fri Nov 23, 2018 6:03 pm, edited 1 time in total.
 
buzzard302
Posts: 152
Joined: Sat Jun 06, 2015 12:06 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 5:38 pm

F9Animal wrote:
Now that we seem to have a general idea on the possible cause of this.... What will likely take place to ensure it doesnt happen again? Also, Lion has a really banged up image. Can we expect to see the flight crew blamed here?

As for Lion Air, is the future looking bleak?


Reading all the information thus far, I would only blame the crew for going ahead with the doomed flight. The maintenance issue was known on previous flight(s) and should have been resolved before the next flight. If it was not corrected, they should not have taken off.

Other than that, there are lots of other factors involved that the crew was facing which perhaps they should have never had to deal with. Perhaps a software update from Boeing will be in the future, as well as updated pilot training.
 
mandala499
Posts: 6587
Joined: Wed Aug 29, 2001 8:47 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 5:57 pm

sgrow787 wrote:
Because this thread is 32 pages long and over 1500 messages, the only method I have to find Zeke's FR24 data is to sniff his profile and search his posts. Hmmm..

You can find it in page 15 of this topic...

I'm taking a few days rest again from this JT610 (and my routine work) madness... I expect the preliminary report to be similar but many will use the moment to stir up interest in the general public again... gotta be ready... (and someone will probably claim I'm on the payroll of a vested interest in this accident, the coverup kind... LOL )
When losing situational awareness, pray Cumulus Granitus isn't nearby !
 
Tristarsteve
Posts: 3588
Joined: Tue Nov 22, 2005 11:04 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 8:41 pm

Agree. Now with the FDR graph in view, I realize that the AoA sensors could maybe move in any direction when the aircraft was taxiing because of the vibration (to be confirmed).


AOA probes are completely useless on the ground below about 60 kts. They can be fully up, or fully down depending on the wind. It is not until good airflow starts over them during the take off run that they move to the correct position.
If you go and touch one on the aircraft on the ground (not allowed!!) you will see that they are very easy to move, and then stay where you leave them.
 
Bradin
Posts: 276
Joined: Wed Jun 15, 2016 5:12 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 9:45 pm

Out of curiosity -

Given that there are at least 200+ 737-MAX deliveries made, do we know how many other 737-MAX that have had conflicting speed information?
 
sgrow787
Posts: 165
Joined: Fri May 16, 2014 8:12 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangal Pinang

Fri Nov 23, 2018 9:46 pm

zeke wrote:
I have managed to complete the filtering of the data on the internet into a sensible flight path. I have made some screenshots of the flight path from different viewpoints to help visualise it.

The corkscrew motion I mentioned previously can been seen in the track from around the 500 second mark (relative to 2018-10-28 23:21:00Z). Vmo (red horizontal line is at 340 KTAS where Vmo is technically 340 KCAS) was almost reached in the turn after takeoff, TAS increased until it passed Vmo in the terminal dive.

Image



Thanks Zeke! I was able to copy-n-paste your nicely formatted data dump into a spreadsheet on my Mac. How to use it with the sim is obviously the next hurdle. The FDR data from the Aviation Herald will be useful as well, especially if I decide to inject values directly at the sensor inputs. I don't know if the flight simulator will let me do that. We do have a hw/sw integration lab that has a python-based test framework that will, but what's the point if you can't check the MCAS response?

I also don't know if the flight sim is equipped with stick shaker and elevator feel shift. That might be something that only airlines have for training pilots.

Anyway, it's going to fun regardless.

Question: Why was the airplane allowed to climb unfettered (by MCAS) from 1500ft to 5000ft?
Question2: At ~100 sec after takeoff, the plane (and passengers) experienced a 200 meter drop in altitude for about 20 seconds. Could that have been the first occurrence of MCAS kicking in and they didn't figure it out until 20 sec later? That is a drop of 10 meters per second, which is 32 feet per second, which is zero gravity inside the fuselage of a plane falling at that rate!
Last edited by sgrow787 on Fri Nov 23, 2018 10:15 pm, edited 2 times in total.
 
zippy
Posts: 142
Joined: Sun May 11, 2014 9:46 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 9:56 pm

Bradin wrote:
Out of curiosity -

Given that there are at least 200+ 737-MAX deliveries made, do we know how many other 737-MAX that have had conflicting speed information?


Yes, Southwest purportedly replaced alpha vanes in an attempt to solve an autothrottle will not engage problem.

http://www.aero-news.net/index.cfm?do=m ... 39238cd542
 
fadecfault
Posts: 164
Joined: Mon Nov 05, 2007 8:44 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 9:58 pm

So according to the graphs both flights experienced a runaway trim condition and stickshaker with the previous crew hitting the cutout switches (per Reuters) and even using the manual trim wheel to trim nose up towards the end.
The later crew kept trying to correct with electric trim but something happened at the end and they applied too little trim. You can see the duration of electric at the end of graph vs the middle and you can also see the trim position keep going nose down. The CVR will need to be found to figure out what caused them to be distracted.
Such a sorry story, too many lives lost for no reason.
The views and opinions written here are my own and do not reflect those of my employer.
 
mxaxai
Posts: 1068
Joined: Sat Jun 18, 2016 7:29 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangal Pinang

Fri Nov 23, 2018 11:03 pm

sgrow787 wrote:
Question2: At ~100 sec after takeoff, the plane (and passengers) experienced a 200 meter drop in altitude for about 20 seconds. Could that have been the first occurrence of MCAS kicking in and they didn't figure it out until 20 sec later? That is a drop of 10 meters per second, which is 32 feet per second, which is zero gravity inside the fuselage of a plane falling at that rate!

Just a nitpick: Velocity is irrelevant with regards to whta you feel. The only thing that counts is acceleration, which was certainly felt during those maneuvers.
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Fri Nov 23, 2018 11:21 pm

fadecfault wrote:
So according to the graphs both flights experienced a runaway trim condition and stickshaker with the previous crew hitting the cutout switches (per Reuters) and even using the manual trim wheel to trim nose up towards the end..


The trim does not run away, STS cuts out when a nose up control input is made opposite to the trim or when the trim switches are used.

MCAS stops when the trim switches are used.

So in both of these cases the pilots thinks they can still control the trim as it still works the way it should when the trim switches are used.

MCAS is an autopilot function that makes trim inputs faster than the pilots trim switches.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
hivue
Posts: 1903
Joined: Tue Feb 26, 2013 2:26 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangal Pinang

Fri Nov 23, 2018 11:22 pm

sgrow787 wrote:
Question: Why was the airplane allowed to climb unfettered (by MCAS) from 1500ft to 5000ft?


Likely because one or more of the conditions required for activation of MCAS were not being met during that period of time.
"You're sitting. In a chair. In the SKY!!" ~ Louis C.K.
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangal Pinang

Fri Nov 23, 2018 11:47 pm

hivue wrote:
sgrow787 wrote:
Question: Why was the airplane allowed to climb unfettered (by MCAS) from 1500ft to 5000ft?


Likely because one or more of the conditions required for activation of MCAS were not being met during that period of time.


If you look at the previos flight and the accident flight you will see that at takeoff they needed around 7 units of trim and as it accelerates around 5 units. The acceleration to 5000 normally wants nose down trim.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
o0OOO0oChris
Posts: 97
Joined: Sun May 13, 2012 10:27 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangal Pinang

Sat Nov 24, 2018 12:35 am

sgrow787 wrote:

Question: Why was the airplane allowed to climb unfettered (by MCAS) from 1500ft to 5000ft?
Question2: At ~100 sec after takeoff, the plane (and passengers) experienced a 200 meter drop in altitude for about 20 seconds. Could that have been the first occurrence of MCAS kicking in and they didn't figure it out until 20 sec later? That is a drop of 10 meters per second, which is 32 feet per second, which is zero gravity inside the fuselage of a plane falling at that rate!


MCAS only operates in clean config, flaps retracted. As Mandala499 said and you can see in the FDR data, the first dive was when flaps where retracted to zero and MCAS kicked in.
When they stopped the dive, thes selected Flaps 5 and kept that all the way up to 5000 feet (for whatever reason), inhibiting MCAS. When the flaps where retracted again, MCAS trimmed down again.
 
User avatar
litz
Posts: 2278
Joined: Wed Dec 24, 2003 6:01 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 12:48 am

Man, you gotta wonder what was going on in that cockpit ... the confusion had to have been paramount.

But this : "When the flaps where retracted again, MCAS trimmed down again." ..... all they had to do was put the flaps back out and return back to the airport.

One has to wonder why they didn't ....
 
User avatar
QuarkFly
Posts: 368
Joined: Mon Aug 15, 2016 4:20 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 1:13 am

This whole ideal that you have...

1) Auto-flight
2) Manual flight but semi-auto features, e.g, STS and MCAS
3) Fully manual flight - when you use the frickin cut-out switches!...causing what flight envelope consequences?

This defeats the whole B philosophy of not overriding pilots' full control during manual-flight -- when manual really does not mean fully manual. And it shows that 737 upgrades over the years may have gone past the point where a traditional flight control system can present the pilots with a controllable aircraft at all times. Frankly, certification authorities should have at least pinged B for this Trim System design...which is only in manual mode if the 737Max FCS allows the pilot to be in manual mode.

Some hypocrites at Boeing !! More ...

http://aviationweek.com/commercial-avia ... t-training

For those who don't pay for AW ...

https://www.thedailybeast.com/before-fa ... m-in-tests
Last edited by QuarkFly on Sat Nov 24, 2018 1:23 am, edited 1 time in total.
Always take the Red Eye if possible
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 1:18 am

litz wrote:

But this : "When the flaps where retracted again, MCAS trimmed down again." ..... all they had to do was put the flaps back out and return back to the airport.

One has to wonder why they didn't ....



They would have put the flaps out again as a result of the low level stall recovery.

MCAS is only flaps up, STS will activate when clean as well.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 1:43 am

QuarkFly wrote:
This whole ideal that you have...

1) Auto-flight
2) Manual flight but semi-auto features, e.g, STS and MCAS
3) Fully manual flight - when you use the frickin cut-out switches!

This defeats the whole B philosophy of not overriding pilots' full control during manual-flight -- when manual really does not mean fully manual. And it shows that 737 upgrades over the years may have gone past the point where a traditional flight control system can present the pilots with a controllable aircraft at all times. Frankly, certification authorities should have at least pinged B for this Trim System design...which is only in manual mode if the 737Max FCS allows the pilot to be in manual mode.



The problem as I see it on the 737 in pitch is a single sensor and single flight control computer (FCC) configuration which can go astray by a single bad sensor without reasonable test of another source. It is really unacceptable that this is not a dual channel design like the yaw implementation on the 737.

The 737 has two FCCs, each with two processors. FCC number 1, and the first processor on the FCC is normally always independently doing the STS and MCAS pitch commands without being checked as being reasonable by the other processors on the same FCC or even the other FCC.

I have stated before that I think it is counterintuitive to turn off the autopilot trim cutout switch in order to turn off STS, MCAS, and Mach trim when in manual flight.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
BobbyPSP
Posts: 345
Joined: Fri Oct 08, 2004 12:29 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 2:00 am

Galwayman wrote:
fadecfault wrote:
Good read about the Lion Air culture on NYtimes
https://www.nytimes.com/2018/11/22/world/asia/lion-air-crash-safety-failures.html

Interesting tidbit:
"The problem should have been fixed within 10 days. But Captain Hasan said that the carrier had a habit of simply moving the faulty radar to another plane. As the clock wound down on the next 10 days, the radar would then be switched to another plane, he said, in a dangerous game of hot potato."

Being in Maintenance for nearly 20 years this says a lot about the safety culture of an airline.
We would NEVER do this.


This is one of the worst years for Boeing 737 aircraft crashes , I think there have been 5 maybe this year and Boeing are running around issuing directives like crazy

This NY article feels like a smear campaign to distract from the crisis at Boeing ....imho


A bit of research would show you with the exception of this Lion Air Flight, they were all overruns/excursions due to bad weather, with the exception of WN that had an uncontained engine failure.

Variety of 700’s and 800’s. Nothing that makes it a 737 issue.
 
F9Animal
Posts: 4219
Joined: Thu Dec 16, 2004 7:13 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 2:52 am

PlanesNTrains wrote:
F9Animal wrote:
Now that we seem to have a general idea on the possible cause of this.... What will likely take place to ensure it doesnt happen again? Also, Lion has a really banged up image. Can we expect to see the flight crew blamed here?

As for Lion Air, is the future looking bleak?


1. I guess a likely software “fix” of some sort by Boeing.
2. The flight crew may bear some blame depending on where the investigation takes us.
3. Lion Air will likely be seen partly as a victim in this unless the investigation reveals some Skelton’s in their closet or poor airmanship.

I just think there’s too much going on that it’s hard to see the outcome clearly at this stage. YMMV.


The sad part is, this could have been prevented. I couldnt imagine what the 2 up front saw in their final moments. It amazes me how technology can still have horrible results. :(
I Am A Different Animal!!
 
ikramerica
Posts: 14871
Joined: Mon May 23, 2005 9:33 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 3:01 am

buzzard302 wrote:
F9Animal wrote:
Now that we seem to have a general idea on the possible cause of this.... What will likely take place to ensure it doesnt happen again? Also, Lion has a really banged up image. Can we expect to see the flight crew blamed here?

As for Lion Air, is the future looking bleak?


Reading all the information thus far, I would only blame the crew for going ahead with the doomed flight. The maintenance issue was known on previous flight(s) and should have been resolved before the next flight. If it was not corrected, they should not have taken off.

Other than that, there are lots of other factors involved that the crew was facing which perhaps they should have never had to deal with. Perhaps a software update from Boeing will be in the future, as well as updated pilot training.

As you state the flight never should have departed. The plane was not repaired. The previous flight most likely should have diverted or returned home. Was dispatching the flight an effort to get the aircraft home at all costs?
Of all the things to worry about... the Wookie has no pants.
 
ikramerica
Posts: 14871
Joined: Mon May 23, 2005 9:33 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 3:08 am

zeke wrote:
QuarkFly wrote:
This whole ideal that you have...

1) Auto-flight
2) Manual flight but semi-auto features, e.g, STS and MCAS
3) Fully manual flight - when you use the frickin cut-out switches!

This defeats the whole B philosophy of not overriding pilots' full control during manual-flight -- when manual really does not mean fully manual. And it shows that 737 upgrades over the years may have gone past the point where a traditional flight control system can present the pilots with a controllable aircraft at all times. Frankly, certification authorities should have at least pinged B for this Trim System design...which is only in manual mode if the 737Max FCS allows the pilot to be in manual mode.



The problem as I see it on the 737 in pitch is a single sensor and single flight control computer (FCC) configuration which can go astray by a single bad sensor without reasonable test of another source. It is really unacceptable that this is not a dual channel design like the yaw implementation on the 737.

The 737 has two FCCs, each with two processors. FCC number 1, and the first processor on the FCC is normally always independently doing the STS and MCAS pitch commands without being checked as being reasonable by the other processors on the same FCC or even the other FCC.

I have stated before that I think it is counterintuitive to turn off the autopilot trim cutout switch in order to turn off STS, MCAS, and Mach trim when in manual flight.

Agreed completely on the first point. There is no reason for this. It’s based on the old analog design ideas.

As for counter intuitive procedures, that’s why it needs to be trained. Much about flying, driving, sailing or moving any large machinery is counterintuitive when dealing with outlying events.
Of all the things to worry about... the Wookie has no pants.
 
maint123
Posts: 175
Joined: Sun Nov 04, 2018 4:18 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 4:21 am

In a nutshell , boeing in order to save on developing a new aircraft , tried to squeeze as much fuel efficiency out of 737 by making 737 aerodynamically unstable and stabilizing it through a software solution.
This software was intoduced quitely to save on retraining and certification costs and bypass regulations.
Now to fly this unnecessarily more unstable plane , the geniuses here are suggesting that the pilots undergo more complicated training.
Nice.
 
Bradin
Posts: 276
Joined: Wed Jun 15, 2016 5:12 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 4:30 am

zippy wrote:
Bradin wrote:
Out of curiosity -

Given that there are at least 200+ 737-MAX deliveries made, do we know how many other 737-MAX that have had conflicting speed information?


Yes, Southwest purportedly replaced alpha vanes in an attempt to solve an autothrottle will not engage problem.

http://www.aero-news.net/index.cfm?do=m ... 39238cd542


I'm no pilot or 737 expert, but could one explain how the lack of enabling autothrottle correlates with conflicting speed information?
 
Chemist
Posts: 540
Joined: Tue Oct 20, 2015 4:46 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 5:13 am

Seems like there is, as in most accidents, a combination of issues:
1 - Poor repair
2 - Pilots not following Boeing procedures/training (cutout electric trim when runaway)
3 - Poor aircraft system design (electronic new system on MAX)

It's really just which is listed as the primary cause and which are contributing causes. All 3 need to be rectified.
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 5:24 am

Chemist wrote:
Seems like there is, as in most accidents, a combination of issues:
1 - Poor repair
2 - Pilots not following Boeing procedures/training (cutout electric trim when runaway)
3 - Poor aircraft system design (electronic new system on MAX)

It's really just which is listed as the primary cause and which are contributing causes. All 3 need to be rectified.


1) No evidence to suggest the maintenance was not performed per the manual. They have the AOA sensor removed prior to this flight which means it was replaced.

2) The FDR data shows it did not run away

3) I would suggest this goes back well before the MAX, about 5 years ago VQ-BBN crashed after STS put a boot load of trim in when high AOA was detected, combined with the pilot input resulted in the aircraft crashing from 2000 ft.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News
 
airtechy
Posts: 706
Joined: Sun Dec 24, 2006 7:35 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 9:04 am

I'm still trying to understand how any failure of the AOA vanes could affect the airspeed indicators .. as seems to have been stated. Unless another "algorithm" is involved, I thought this was purely a function of the pitot pressure and the static air sensor. I can understand how the stick shaker needs both AOA and airspeed.

Also whether it's a stuck control column trim switch, an incorrect autopilot input, and incorrect STS or MCAS command .. all of these result in runaway trim which is fixed by the cutout switches or grabbing the trim wheel as all of these commands are applied .. AFAIK .. to the trim wheel drive motor.

Unfortunately, it seems one component failure caused a false stall indication (with stick shaker) and the equivalent of runaway trim and the crew couldn't deal with it for one reason or the other. Hard to see how this could happen in what was supposedly daytime VFR conditions.
 
User avatar
AirlineCritic
Posts: 1629
Joined: Sat Mar 14, 2009 1:07 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 9:14 am

litz wrote:

But this : "When the flaps where retracted again, MCAS trimmed down again." ..... all they had to do was put the flaps back out and return back to the airport.

One has to wonder why they didn't ....


Because they didn't know! With perfect hindsight we now know what would have been necessary -- either flaps or cutout switches.
 
User avatar
AirlineCritic
Posts: 1629
Joined: Sat Mar 14, 2009 1:07 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 9:20 am

buzzard302 wrote:
Reading all the information thus far, I would only blame the crew for going ahead with the doomed flight.


Really? This is your conclusion? My conclusion is that we first of all don't know all the details and conclusions, but some combined blame on flight crew confusion and airplane is likely.

But you conclude that it is all on crew... despite not knowing what maintenance fixed (they may well have done everything by the book; we don't know yet). And per WSJ, Boeing is designing a software fix. Does that sound like only-crew-to-be-blamed situation to you?
 
WIederling
Posts: 8489
Joined: Sun Sep 13, 2015 2:15 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 9:55 am

litz wrote:
One has to wonder why they didn't ....


You had a couple of hundred hours

to get here and a gross of posters collecting all the data and massaging it into a useful form.
Quite a bit of that information principally unavailable to the crew as Boeing never divulged
to Lion Air what we now know.

The crew had ?two? minutes?
Murphy is an optimist
 
buzzard302
Posts: 152
Joined: Sat Jun 06, 2015 12:06 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 11:24 am

AirlineCritic wrote:
buzzard302 wrote:
Reading all the information thus far, I would only blame the crew for going ahead with the doomed flight.


Really? This is your conclusion? My conclusion is that we first of all don't know all the details and conclusions, but some combined blame on flight crew confusion and airplane is likely.

But you conclude that it is all on crew... despite not knowing what maintenance fixed (they may well have done everything by the book; we don't know yet). And per WSJ, Boeing is designing a software fix. Does that sound like only-crew-to-be-blamed situation to you?


It's not a conclusion at all. It's a comment and an opinion. If you were a pilot that day on that plane and knew that the previous flights were struggling to maintain control, would you have jumped up front and put yours and all those passengers lives at risk? I suppose pilots have no authority to question the safety of the airplane they are about to fly?

If you read some of my other comments throughout this thread, I have been very specific that it is a chain of events that lead to this accident. Starting from maintenance, to the pilots, and including Boeing as well. I never intended to conclude it was only-crew-to-be-blamed. My previous post wording was not well chosen
 
User avatar
PixelFlight
Posts: 508
Joined: Thu Nov 08, 2018 11:09 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 11:47 am

airtechy wrote:
I'm still trying to understand how any failure of the AOA vanes could affect the airspeed indicators .. as seems to have been stated. Unless another "algorithm" is involved, I thought this was purely a function of the pitot pressure and the static air sensor.

Angle Of Attack Effect On Indicated Airspeed
viewtopic.php?t=738067
 
o0OOO0oChris
Posts: 97
Joined: Sun May 13, 2012 10:27 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 11:54 am

zeke wrote:
They would have put the flaps out again as a result of the low level stall recovery.


That makes sense. They had stickshaker running and UAS, makes sense where that there was a mindset to put the flaps out again.

zeke wrote:
Chemist wrote:
Seems like there is, as in most accidents, a combination of issues:
1 - Poor repair


1) No evidence to suggest the maintenance was not performed per the manual. They have the AOA sensor removed prior to this flight which means it was replaced.


I have been thinking about this. Per maintenance record, AOA was not replaced before the accident flight, so the new vane was also on the previous flight.
What I don`t understand is why didn`t the previous crew mention the continously going stickshaker in the maintenance record. I think this is pretty significant.

I try to think how the maintenance tried to figure out what was wrong. (Though I may be wrong, I am no aircraft maintenance engineer). The going stickshaker may have been written up in a previous log and as the stickshaker ties the problem to the AoA Sensor, it was replaced. Now the maintenance guys see that no stickshaker is mentioned, and as the AoA vane was replaced, they may thought they eliminated that probable cause and had to look for the culprint in something else. As they had no further clue, they just performed the actions to the mentioned problems by the pilots to the book.

And maybe the max maintenance manual was just as poor in documenting the MCAS system as the FCOM, they may have not have enough knowledge to analyse the situation.

But the previous crew not mentioning a continously going stickshaker as a hint for maintenanence baffles me.

Maybe I am wrong and a engineer can shed his view on this?
Last edited by o0OOO0oChris on Sat Nov 24, 2018 11:55 am, edited 1 time in total.
 
devplane
Posts: 2
Joined: Sun Aug 10, 2014 7:30 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 12:28 pm

airtechy wrote:
I'm still trying to understand how any failure of the AOA vanes could affect the airspeed indicators .. as seems to have been stated. Unless another "algorithm" is involved, I thought this was purely a function of the pitot pressure and the static air sensor. I can understand how the stick shaker needs both AOA and airspeed.


I looked a bit into the effects of AOA on pitot tubes. I found some datasheet of a pitot tube which shows an air speed error in the range of 2% due to an angle between the air flow and the pitot tube direction (https://www.techmark.de/files/techmark/ ... sonden.pdf). Form the FDR data the air speed error look to be >5%. So if the air speed is corrected for an angled approach flow, there is a possibility that the speed mismatch is due to the AOA mismatch. Although according to this datasheet the air speed error is still a bit to large.

On the other hand if the air speed error is not due to the AOA error, I think the ADIRU was faulty. Because two similar fault (constant offset) in two different independent sensors appears extremely unlike.

It woud be interssting to know the dependency of the air speed on approach flow for the pitot tube of the 737.
 
airtechy
Posts: 706
Joined: Sun Dec 24, 2006 7:35 am

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 1:28 pm

PixelFlight wrote:
airtechy wrote:
I'm still trying to understand how any failure of the AOA vanes could affect the airspeed indicators .. as seems to have been stated. Unless another "algorithm" is involved, I thought this was purely a function of the pitot pressure and the static air sensor.

Angle Of Attack Effect On Indicated Airspeed
viewtopic.php?t=738067


I understand the effect of angle of attack on airspeed. That was not the question. What does that have to do with a plane .. this plane .. that was within the normal angle of attack range? Only the AOT vane thought it was not .. evidently thinking it was at a stall angle of attack.

Unless there was another failure or somehow the airspeed indication is compromised by a failed AOA, I suspect the airspeed was correct the whole flight.
 
stratclub
Posts: 1328
Joined: Fri Jan 05, 2018 10:38 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 1:41 pm

kalvado wrote:
WIederling wrote:
( I don't think "MCAS of no immediate concern for training" is an acceptable position.)

Still it makes some sense. I would compare MCAS with car airbag to some extent. If you need it, you are significantly outside normal mode, and while it is a planned emergency feature- it makes little sense to practice the use, better spend training time on avoiding those marginal situations.
It is a bad idea to withhold information, even worse idea to implement it poorly. But I don't think crash axe is part of sim arrangement as well.
Still wondering what are the other differences hidden by the software, though.

The MCAS implementation was completely bogus from what I have read. Give my airliner a life threatening function and not tell me about it? Really? How completely stupid is that? My aircraft auto trims it's self into a stall situation. Really? I'm pretty sure I would like to know about that so that I could just plain old just fly the aircraft if the automation tries to kill me which it seems to have done in Lion Airs case.
Last edited by stratclub on Sat Nov 24, 2018 1:59 pm, edited 1 time in total.
 
User avatar
zeke
Posts: 13831
Joined: Thu Dec 14, 2006 1:42 pm

Re: Lion Air 737MAX8 Crashed Jakarta to Pangkal Pinang

Sat Nov 24, 2018 1:54 pm

mandala499 wrote:
This incident however raises my eyebrow as I suspect it was the ADR portion of the ADIRU that was faulty with PK-LQP... No evidence of that yet,


I think you are correct, especially after they replaced the AOA probe. The AOA probe is a mechanical unit with an analogue to digital (AotD) converter on the inside. The previous flight data suggests that the AOA that was computed by the ADIRU was faulty on the previous flight, replacing the AOA gave similar AOA values.

I think when they send the AOA probe back to the manufacturer it will be functing correctly.

The AOA probe normally have an inbuilt AtoD converter, the output varies however one common way to transmit the value of the angle of attack is to vary the frequency with the angle of attack value, this is known as frequency modulation. If the ADIRU had an issue demodulating the input from the AOA probe it would explain the erroneous results. The FDR data for the AOA will not be the data actually transmitted by the AOA, it will be the AOA computed by the ADIRU.

I have had a chance to look at the published FDR to date

Image

1 - power up, trim set for takeoff at 7.0 units up
2 - on power up the LH and RH AOA are different, this is very normal, they are often pointing different directions on the ground.
3 - Flaps set for takeoff
4 - Flight control check
5- AOA probe moves during taxi
6 - LH and RH airspeed agree

Image

1 - control column moved forward on thrust application (normal)
2 - IAS rising normally both sides, AOA also aligns with free-stream airflow with a split between sides
3 - Rotation, LH stick shaker comes on straight away AOA(left) approximately reading 22 degrees, AOA (right) zero degrees. RH stick shaker not activated
4 - STS adding trim inputs
5 - flap retraction
6 - First MCAS input, pilot stops MCAS trim with trim switch input, very high control force at this time.
7- Pilot appear to reduce the back pressure slightly like it was a stall recovery procedure, slight increase in fuel flow observed. Alitiude traded for speed.
7a - LH - Stick shaker stops
7b - Flaps selected like a low level stall recovery
8 - LH Stick shaker starts again
9 - Reaching 5000 ft 250-350 kts flaps retracted

Image

1-25

MCAS trims stabilizer nose down 25 times, each time the pilot stops the trim with trim switch input, this inhibits MCAS for 5 seconds each time. The average trim position after MCAS trim was 4.01 units, after the pilot corrected the MCAS trim the average trim was 4.96 units.

After MCAS, After Pilot Trim
4.92, 6.09
4.53, 6.03
4.58, 5.47
4.69, 5.25
4.47, 5.47
4.80, 5.81
4.64, 6.42
4.69, 5.59
4.36, 5.31
4.47, 5.59
3.91, 5.31
5.03, 5.42
4.47, 5.47
4.47, 5.59
4.19, 4.92
3.35, 5.31
4.47, 5.59
4.19, 4.58
4.13, 5.03
3.63, 5.36
4.19, 5.36
3.46, 3.58
3.02, 3.35
1.34, 1.62
0.28, 0.45

26 - 26 Not sure what is happening here. The IAS is increasing, the have run out of stabilizer trim, they appear to increase thrust in an attempt to induce a pitch up. Stabilizer almost at full travel, suggest the elevator was ineffective.
Human rights lawyers are "ambulance chasers of the very worst kind.'" - Sky News

Popular Searches On Airliners.net

Top Photos of Last:   24 Hours  •  48 Hours  •  7 Days  •  30 Days  •  180 Days  •  365 Days  •  All Time

Military Aircraft Every type from fighters to helicopters from air forces around the globe

Classic Airliners Props and jets from the good old days

Flight Decks Views from inside the cockpit

Aircraft Cabins Passenger cabin shots showing seat arrangements as well as cargo aircraft interior

Cargo Aircraft Pictures of great freighter aircraft

Government Aircraft Aircraft flying government officials

Helicopters Our large helicopter section. Both military and civil versions

Blimps / Airships Everything from the Goodyear blimp to the Zeppelin

Night Photos Beautiful shots taken while the sun is below the horizon

Accidents Accident, incident and crash related photos

Air to Air Photos taken by airborne photographers of airborne aircraft

Special Paint Schemes Aircraft painted in beautiful and original liveries

Airport Overviews Airport overviews from the air or ground

Tails and Winglets Tail and Winglet closeups with beautiful airline logos